Turbo Machinery Presentation Collection

Table of Contents
WHAT TO DO WHEN A CLASSIC RCA DOESNT GIVE THE ANSWER APPLY DETAILED ENGINEERING ANALYSIS
by Michael J. Drosjack
Senior Principal, Rotating Equipment Engineering
John W. Felten
Senior Staff Engineer, Rotating Equipment Engineering Shell Global Solutions (US) Inc. Houston, Texas
George H. Seamon
Principal Design Engineer, Expanders
and Tim R. Griffin

Senior Technical Specialist, Expanders and Axial Compressors Dresser-Rand Company Olean, New York
Michael J. Drosjack is a Senior Principal in the Rotating Equipment Department at Shell Global Solutions (US) Inc., in Houston, Texas. He is responsible for providing technical support for rotating and reciprocating machinery to Shell and Shell affiliated companies worldwide, as well as commercial customers. Since joining Shell in 1975, he has had assignments on projects involving specification, evaluation, installation, and startup of machinery along with extensive field troubleshooting, particularly in the area of vibration measurement, vibration analysis, and rotordynamics. Dr. Drosjack received his B.S. degree (Mechanical Engineering, 1970) from Carnegie-Mellon University, and his M.S. (1971) and Ph.D. (1974) degrees (Mechanical Engineering) from The Ohio State University. He is a member of ASME, the Vibration Institute, the Machinery Subcommittee of the Ethylene Products Committee, participates in API task forces, and has been a speaker and panelist for NPRA. He has been a Turbomachinery Symposium Advisory Committee member since 1986. John W. Felten is a Senior Staff Engineer in the Rotating Equipment Group at Shell Global Solutions (US) Inc., in Houston, Texas. He is responsible for providing technical support for rotating and reciprocating equipment to Shell and Shell affiliated companies worldwide, as well as commercial customers. He joined Shell in 1987 and has had assignments in multiple refineries encompassing day-to-day rotating equipment support and troubleshooting, new projects, and revamps. Mr. Felten received his B.S. degree (Engineering with a Mechanical Specialty, 1986) from Colorado School of Mines. He has presented expander-related material in both NPRA and ASME.
7
George Seamon is a Principal Design Engineer for Dresser-Rand Company, in Olean, New York. For the last 19 years, he has been responsible for the aerodynamic and mechanical design and development of hot gas expanders for FCC and nitric acid service. Prior to that, he spent six years on the design of gas turbines and four years on the design of the GHH type hot gas expander. Before joining Dresser-Rand, Mr. Seamon worked for 10 years with General Electric and Pratt & Whitney on heat transfer, aerodynamic, and mechanical design of the turbine section of jet engines. Mr. Seamon graduated with a BSME/AE degree from Northwestern University (1967).
ABSTRACT
A methodology is presented to identify the most likely cause of failure for a cat cracker hot gas expander in which the classic failure mechanisms did not fit the evidence collected after the failure. Because the classic root cause analysis (RCA) did not provide a probable theory for the failure cause, a comprehensive engineering analysis was required to positively rule out causes that proved improbable and focus on the most likely theory. Included is a review of the failure mechanisms considered in the search for the cause of a blade failure in the expander. This paper provides a discussion of how common failure mechanisms were eliminated leading to a theory in which catalyst deposits that built up and eventually shed off of expander blades led to supersonic flows in a converging-diverging nozzle created by blade deposits. The cycling of the blade loading from successive buildup and shedding of multiple deposits on the same blades and the resulting changes from subsonic to supersonic flows in the blade throats are suspected to have created blade cracks in an unexpected location leading to the eventual failure of this machine.
PROCEEDINGS OF THE THIRTY-FIFTH TURBOMACHINERY SYMPOSIUM 2006
Table of Contents
INTRODUCTION
A typical cat cracker process unit has the process flow configuration shown in Figure 1. The primary machine is the main air blower that provides air for both combustion and transport of the catalyst and feed. A reaction is induced in either the riser or reactor and the desired products separated in downstream units. Residual hydrocarbon is burned off the catalyst in the regenerator after the reaction occurs. The gas coming off the regenerator is hot (1250 to 1350F) and at a pressure higher than the carbon monoxide (CO) boiler, the next downstream device. In many cat crackers, a hot gas expander is used as an energy recovery device in this stream. The layout of a typical expander is shown in Figure 2. This machine, in effect a one stage power turbine, has a very difficult environment to work in. It is subjected to hot gas with trace elements and varying amounts of extremely hard catalyst fines.
The unit is a Dresser-Rand (Ingersoll-Rand) Model E-148. Design conditions were:
Inlet pressure: Inlet temperature: Flow rate: Exhaust pressure: Power generated:
44.14 psia 1200F 775,000 lbs/hr 15.56 psia 24,400 hp
The expander is directly coupled to the main air blower. Also included in the train were a motor/generator to address variations in the power generated or consumed and a helper steam turbine for use in startup. Shown in Figure 3 is a photo of the train.
Figure 3. Photo of the Expander Train. Due to severe imposed duties, hot gas expanders have a history of being problematic devices. History indicates that operational failures are more common than in machines in more benign operating environments. This installation had experienced several operational failures in the past and the refinery was able to react to the failure in a practiced manner. On June 6, 2003, a unit shutdown occurred. A high radial vibration trip was the result of a sudden jump in vibration caused by a blade failure. Before the shutdown, there were no obvious indications that this machine was in distress either from the vibration or process data monitoring. Loss of this machine caused the fluid catalytic cracking (FCC) unit to be shut down. Repair and restart resulted in seven-figure production losses. As a result of the failure, an RCA was immediately initiated. Available data from the vibration and process monitoring systems were gathered. Analysis began to identify anomalies or tell-tale indications that something was wrong or had been changing with the machine or process. Once the machine was entered, visual examination was made, measurements gathered, and photos taken. The rotor and stator were moved to the shop for additional examinationoriginal equipment manufacturer (OEM) design staff and additional technical staff were deployed to perform the analysis. The broken blade was found in the piping systema rather unusual location. All of the blades were subjected to comprehensive metallurgical analysis. Because hot gas expanders have a well-documented failure history with numerous RCAs and engineering studies, the analysts had an expectation that they would reach a NORMAL conclusion. That is, a failure mode and cause that had been previously evaluated on many other failures. That expectation was not met. Metallurgical work indicated that the failure was caused by fatigue, a typical occurrence in expander blade failures. However, the failure location on the blade and the initiation site were not
Figure 1. Typical Refinery Cat Cracker Process.
Figure 2. Typical FCC Expander Configuration.
Table of Contents WHAT TO DO WHEN A CLASSIC RCA DOESNT GIVE THE ANSWER APPLY DETAILED ENGINEERING ANALYSIS 9
consistent with the normal loading associated with expander failures. In addition to the failed blade, two other blades were found that were cracked in a similar location. Historical vibration and process data were reviewed very carefully. There were a few anomalies, but none that really tied to NORMAL causes. As a result, the RCA team progressed through the standard RCA practices and found themselves looking for loads and forces outside those considered typical. Thus, they had to deviate from the NORMAL analysis protocols and employed more sophisticated analytical tools.
smoothly as all equipment returned to service and operated properly. The process unit was lined out and the process conditions for the expander were stabilized. To illustrate the process conditions upon the restart, Figures 4, 5, and 6 depict flue gas flow, expander inlet pressure, and expander inlet temperature, respectively.
ROOT CAUSE FAILURE ANALYSIS

Organization of RCA and Participants The blade failure on the expander was an extremely sudden and unexpected event. Parallel to beginning the repair on the machine, an RCA team was formed to determine what had happened and why. As with any RCA investigation, it is vitally important to form a team with a breadth of experience and viewpoints. To this end, this RCA comprised personnel from refinery staff, the end-user, and the OEMs expander engineering resources. This personnel represented engineering disciplines such as:
Figure 4. Flue Gas Flow.
Process engineering Reliability engineering Rotating equipment engineering Metallurgical engineering Expander design Operations
With the basic RCA team formed, the charter and ground rules were set to define its function. Of vital importance to any team are communication and appropriate team interaction. There were the normal cautions to ensure open dialogue between different engineering groups and disciplines. An additional dimension in this case was the inclusion of team members from multiple companies. Because there was a shared goal in understanding the failure, communication was kept open and information freely shared. This resulted in a very cohesive RCA team. A refinery senior reliability engineer led the RCA team. After preliminary face-to-face meetings held at the refinery to initiate the effort, the RCA team continued its work with teleconferencing and networking via an Internet-based desktop sharing program. This permitted the team to review data and common files together without having to travel across the country to sit face-to-face. This paper will detail the data-gathering and analytical work performed to reach the ultimate conclusion. Included in this analysis are: Figure 5. Expander Inlet Pressure.
Figure 6. Expander Inlet Temperature. Along with the stable process operating conditions, the expander performed well mechanically. Vibration levels were primarily less than 1.0 mil for the duration of the run (Figure 7).
Vibration and process data review Photos and physical examination of the machine Metallurgical failure analysis Classic blade stress analysis Finite element analysis Evaluation of classic blade failure mechanisms Load evaluationstress testing of the blade Computational fluid dynamics (CFD) analysis and load calculation
for abnormal throat geometries
Identification of potential abnormal loading scenarios

Vibration and Process Data Review The FCC unit was started in November 2002, after completion of a planned maintenance outage. The startup of the unit went
Figure 7. Expander Disk End Vibration. The vibration levels for this expander run were relatively low compared to previous runs as shown in Figure 8. During the scheduled maintenance outage in 2002, normal maintenance activities were performed (rotor change out), and there were no
10
Table of Contents
design changes made to the expander, cat cracking unit regenerator, or third stage separator system.
remnant was not found within the expander casing. After a thorough inspection, it was found in a vertical section of the exhaust duct approximately 30 feet above the expander exhaust flange. The failed blade was located on an expansion joint inner flange surface along with foreign objects. The objects, shown in Figure 11, were actually two pieces of what looked to be either a washer or some type of fabricated part. A metallurgical analysis of the object confirmed that it was neither a component of the expander, nor a component of the associated piping and vessels.
Figure 8. Historical Expander Disk End Vibration Amplitudes. About one-and-a-half months after the restart, it was noted that the expander inlet flow function was trending downward as shown in Figure 9. Flow function is a measure analogous to an orifice coefficient and is related to the effective opening of the expander flow path. Mathematically, the expander flow function is calculated by using inlet temperature, inlet pressure, and mass flow per the following equation:
Flow Function =
Mass Flow ( Inlet Temperature) Inlet Pr essure
0.5
Figure 10. Rotor with Failed Blade.
Figure 9. Expander Flow Function. A downward trend in the flow function tends to indicate that the flow path area was getting smaller. This reduction in area was assumed to be caused by deposition of catalyst fines. This expander had experienced deposition previously. Operations had procedures to routinely inject walnut hulls online to blast deposition off. Visual inspection through the expander view ports verified that the walnut hulls were cleaning the shroud area effectively. The deposition was reasoned to be occurring on the stator vanes and/or rotor blades. Injection of walnut hulls on a routine basis was continued during the run to try and maintain the flow path. Without warning on the morning of June 6, 2003, the expander tripped because of high radial vibration. Photos and Physical Examination of the Machine The machine was made ready for inspection and maintenance. Deposition was visually confirmed as a significant amount of deposits were found within the expander case. In a normal shutdown of the expander, the machine is run to deinventory the cat cracking unit and deposits would be blown out of the machine before inspection. Because the shutdown occurred suddenly and the machine did not continue to run, the deposits that had formed in the flow path were not removed. When the machine was opened for inspection, it was discovered that the rotor had experienced a blade failure as shown in Figure 10. In addition to this blade failure, the stator was damaged, as was the shroud, diffuser, and inner exhaust case. The failed blade Figure 11. Foreign Objects Found Near Blade. Two other blades were found to have cracks near the platform. These blades were located 12 blades and 22 blades, respectively, away from the failed blade (there are 62 blades in this rotor). These cracks were not visible to the naked eye until dye penetrant tests were conducted on each blade (Figure 12).
Figure 12. Dye Penetrant Inspection Results of One of the Blades Found to Be Cracked but Intact.
There was significant damage to the components within the flow path of the expander. The stator was damaged with trailing edges broken on a majority of the vanes (Figure 13).
Figure 14. Fracture Surface of Blade 1. The crack was examined in the unetched condition with the optical microscope as well as the scanning electron microscope. The crack appeared relatively straight (there is a slight curvature toward the top of the blade) and primarily transgranular. There were a few secondary cracks that had initiated off the main crack, but these had only progressed a grain or two away from the main crack. Deposits were found in the crack, although it could not be classified as deposit-filled. Some of these deposits could be attributed to polishing compounds that were used, but evidence of sulfur was present in several locations along the wall of the crack as well as in some of the secondary cracks. The transgranular nature of the crack combined with the observation of striations on the fracture surface indicates a failure mechanism of mechanical fatigue. platform: Similar analysis was performed on blades 13 and 47. The crack identified through dye penetrant testing was opened and analyzed using an SEM. The cracking mechanism was determined to be high cycle fatigue, with the fracture origin in approximately the same location as the fracture origin in blade 1. No evidence was found of any erosion or impact damage that might have initiated cracking.
Figure 13. Stator with Numerous Damaged Vanes. The other components that were damaged, i.e., shroud, diffuser, and inner exhaust case all showed indications of being struck by the failed blade. Failure Analysis Upon completion of the initial inspection, the rotor with blades and the complete stator were moved from the field for further inspection and analysis. The analysis team worked together to analyze these components. Metallography work was performed by the end-user and the OEM to ensure that all possible avenues were explored. This also reduced the time required to complete the analysis. The metallography work yielded the following:
Blades 13 and 47Additional blades found with crack near
Blade 1The failed blade: The fracture surface of blade 1 was examined by the scanning electron microscope (SEM). Chemical analysis was performed on the blades fracture surface before cleaning. The uncleaned blade had areas with large amounts of silicon, aluminum, and oxygen, in addition to the expected elements. This was attributed to catalyst deposits on the fracture. Polished metallurgical samples indicated that small amounts of sulfur were present on the fracture face. The fracture itself was confirmed to be the result of high cycle fatigue followed by ductile overload. Fatigue striations were confirmed in several areas on the fracture face, although large portions of the fracture were damaged and unreadable because of abrasion, handling, and catalyst deposits that could not be removed during the cleaning process. The origin of the fracture was clearly identified. The coating was still intact, although somewhat thinner near the origin of the crack. No evidence of any other cracking mechanism (environmental cracking) was detected at the origin or anywhere else on the fracture. The fracture surface of blade 1 is shown in Figure 14. The platform and the base of the airfoil are visible in the figure. The crack initiated by fatigue on the suction side of the blade, about 13/8 inches from the leading edge. The crack advanced by fatigue about 80 percent of the way through the blade thickness, toward the pressure side, and then arrested for sufficient time to form a bluish oxide on the crack surface before beginning to advance again by fatigue. The crack propagated almost through the entire blade thickness by fatigue before converting to ductile rupture at which point the blade separated quickly from its base.
Stator
vanes: Forty out of the 50 stator vanes were found damaged. After inspecting the stator vanes, two vanes were selected for further analysis. The two stator vanes selected, 7 and 37, had fracture surfaces typical of the entire group. The fracture surfaces were characterized in the SEM. Some areas on vane 7 were distinguishable as brittle fracture, but most of the surfaces of both vanes were covered with catalyst deposits. Cleaning in acetone and in inhibited hydrogen chloride (HCL) did not remove this deposit.
Hardness and impact testing: The blades and vanes were tested for hardness as well as Charpy impact results. In all cases the tests yielded results that correlated with expected design properties. There were no deviations found in blade, vane, or coating material properties. Conclusions of metallography: Three rotor blades cracked because of high-cycle fatigue. All three cracks initiated at approximately the same location, on the suction side between the leading edge of the blade and the point of maximum camber, just above the platform. All three blades came from different heats and the chemical analysis and hardness results indicated normal chemistry. This yields the conclusion that the original condition of the blades did not play a role in the failure.
12
Table of Contents
The fatigue crack on the severed blade was observed to be transgranular over its entire length. Secondary cracking was almost nonexistent. Although sulfur was present on the fracture surface, the presence of sulfur is ubiquitous in expanders, and the conclusion is that sulfur played no role in the blade failure. Classic Blade Stress Analysis A Goodman diagram is a classic tool used in analysis of a failure of this type. While a finite element analysis provides more detailed quantitative results, the Goodman diagrams permit a means to compare to other expanders (and gas turbines) in the fleet of an OEM. A modified Goodman diagram is shown in Figure 15.
The calculated stresses displayed in these Goodman diagrams include simple pull over area calculations to which gas bending forces are added. In the airfoil root these are standard classic stress calculations. The firtree area is very geometry specific and a stress concentration factor is used that is based on the radius of curvature of the firtree neck. Each lug has added stress components from local bending and shear. The combined stresses for the Goodman diagrams are evaluated at six locations of high stress on the blades. Evaluation of past failures indicates that these six areas cover the highest percentage of previous failures. The Goodman diagram for the airfoil root is the most critical area of evaluation for this failure. In the airfoil, the leading edge and trailing edge normally have the highest stress because gas-bending loads add to the centrifugal stress at these locations. The max camber point is about halfway around the suction (convex) side of the airfoil. It gets its name from being the maximum distance from the minimum bending axis, which is a line running through the airfoil center of gravity and almost parallel to a line that is tangent to the leading edge and trailing edge. Figure 17 shows the minimum bending axis and the max camber point. The effect of gas bending is to lessen the centrifugal stresses at the max camber point.
Figure 15. Modified Goodman Diagram Showing Airfoil Root Stresses. The vertical axis is alternating stresses with the maximum value being the endurance limit of the material at operating temperature. One-third of that endurance limit is identified as the allowable alternating stress when steady-state stresses are zero. The horizontal axis plots steady-state stresses; for high temperature operation the maximum value with zero alternating stress is 100,000 hours creep rupture capability. Eighty percent of the 100,000 hour rupture strength is considered the allowable steady-state stress when alternating stresses are zero. A line connects the two points of allowable alternating and steady-state stresses; it represents the maximum design limit in the case of combined alternating and steady stress. The expected stress levels for a machine are plotted on this diagram. Distance from the limit is representative of safety margin. Figure 16 provides a comparison of the stress levels in this unit to others in operation. From this plot it can be seen that the design of this particular expander appears to be conservative. There are many other expanders and gas turbines operating with calculated stresses much closer to the allowable limits, i.e., lower safety factors.
Figure 17. Photo Showing Location of Minimum Bending Axis and Maximum Camber Point. Evaluation of recorded failures indicates that the majority of failures that have occurred in the airfoil section are at either the leading or trailing edge. A few have been at the max camber point. A crack initiation in the location observed on the failed blade is unique and cannot be easily explained by classic stress analysis or normal blade loading. That is, stresses caused by normal loading are always going to be higher at the three locations plotted in the Goodman diagram than they would be at the point of crack initiation in this failure. This means the classic stress calculations would not predict a failure in the location where this one occurred. Finite Element Analysis A finite element analysis (FEA) model was used to perform a more detailed analytical examination of the stresses in the blade under various loading conditions. The results for the model with normal steady-state loads are shown in Figures 18 and 19.
Figure 16. Comparison of the Normal Stresses in the Failed Unit to Higher Stressed Units in Operation.
Figure 20. Photo Showing Location of Straingauges for Bench Test. Figure 18. Steady-State Stress Analysis Results, A.
Figure 19. Steady-State Stress Analysis Results, B. Load EvaluationStress Testing of Blading The FEA model was calibrated using a bench test. A blade taken out of the unit had four straingauges installed around the perimeter of the root at the following locations: Figure 21. Locations of Load Application for Bench Test.
Leading edge Crack initiation site Max camber Trailing edge

Figure 20 shows these strain gauge locations. Figure 21 shows the locations where point loads were applied to the blade tip and Figure 22 shows one of the loads being applied during the bench test. Strains were measured with each load case and are shown in Table 1. The FEA model was run at zero rpm with point loads similar to the bench tests. Nodes of the finite elements closest to the straingauge location were chosen and the calculated stresses noted for each load case. Strain was calculated and compared directly to the measured strains. These values are also shown in Tables 1 and 2 and demonstrate a very good correlation of the model to the bench test. This confirms that the results from the FEA model can be taken as reliable. This provides a foundation for using the FEA model for further investigation.
Figure 22. Photo of a Load Being Applied to a Blade During the Bench Test.
14
Table of Contents
Table 1. Measure Strain from Bench Test.
Table 2. Calculated Strain from FEA Model.
Figure 23. Fourier Analysis Results for Stator Vane Throat VariationsUndamaged Stator. Upon disassembly, the stator was found to have significant damage. A theory was tested to evaluate a potential excitation from the damaged stator (assuming the damage had been a precursor to the blade failure). The stator damage was surveyed in detail and the resulting throat dimensions analyzed to predict potential harmonic excitations. The results are shown in Figure 24. Although the exit pulses from the stator are understandably stronger in this case than for the new stator, the pulses predicted at the blades natural frequencies were too weak to have been the cause of the failure.
Classic Failure Mechanisms On the basis of prior failure histories with FCC expanders, there is a methodology that is followed to address classic failure mechanisms. One of the most common mechanisms is the excitation of blade resonance frequencies by either strut-passing-frequency-flow excitations or stator-driven-flow excitations. FCC expanders are further challenged by potential natural frequency interferences because, in the course of a run, natural frequencies of the blades can be lowered or raised by metal removal driven by catalyst erosion. The design of this expander incorporated uneven nosecone strut spacing to minimize the risk of a strong excitation. In addition:
Figure 24. Fourier Analysis Results for Stator Vane Throat VariationsDamaged Stator. A failure caused by any stator damage should have resulted in blade crack initiation at the leading edge, trailing edge, or max camber. Thus the stator-driven forces were eliminated as a principal cause of failure. Foreign object damage to the blades also was considered as a possible failure cause, in particular because foreign material was found in the line downstream of the unit. However, the two other blades with cracks did not show any evidence of impact such that foreign object damage was ruled out as a likely cause of failure. Blade tip rubs on shroud catalyst deposits have caused failures in FCC expanders. This unit had experienced blade tip rub damage in the past from contact with catalyst deposits on the shroud. However, during this run there were no physical indications of tip rubs. Initial Results of RCANo Answer Data, such as discussed above, were gathered and evaluated by the RCA team. The evaluation protocol was practiced and applied rapidly. A cause-and-effect diagram was utilized to record results and validate conclusions. Data were used to validate or eliminate branches of the trees. Issues concerning the strength of the blades and vanes were
Three other sets of identical blades operated successfully in this

machine without fatigue failures (from 1997 to 2002).
The current failure was not at the leading edge, trailing edge, or The rotor blades had no metal loss, and therefore had not changed in natural frequency during the course of the run.
max camber point where a first bending mode would show up as high stress.
Thus the strut passing resonance excitation was eliminated from consideration. A harmonic excitation caused by a variation in the stator vane throat geometry was evaluated. A Fourier analysis was performed using the stator throat dimensions of the original stator to determine the potential excitation frequency and amplitude. The results of this analysis are shown in Figure 23. They show that the harmonics identified were not strong enough to have caused the failure. And if there had been a strong excitation, a blade failure would have occurred in the first few days of operation.
Table of Contents 15
eliminated as possible causes with metallurgical data. The results of tests conducted to confirm mechanical strength as well as coating properties were within the original design parameters. Analysis of the stator indicated that stator damage did not precede a rotor blade failure. Ring testing of rotor blades for resonance confirmed that the blades natural frequencies had not been shifted by erosion into a region where they could be excited by normal operation. Detailed inspection of the machine assembly verified that there were no installation or assembly errors that could have lead to the blade failure. The possibility of the failure being the result of a foreign object was investigated and rejected. It was discovered that the failure originated in an area of the blade that was not highly loaded under normal operation. The crack that ultimately led to the failure was found to have originated on the suction side of the airfoil between the leading edge and max camber point. As the classic modes of failure were eliminated, they were crossed out on the cause-andeffect diagram. Figure 25 displays a portion of the diagram as modes of failure were eliminated.
case were approximated based on the results of a CFD analysis that included a convergent/divergent passage created by a blade deposit that is discussed in more detail in the following section. These CFD results provided the most plausible theory, and that theory is considered to be the most likely failure scenario.
Figure 26. Stress Visualizer Results for Two Load Cases. Identifying Potential Loading Mechanisms Extensive CFD analysis was done with normal airfoil shapes and varying inlet and outlet flow conditions. None of those cases generated gas loadings that resulted in a predicted failure at the crack initiation site. Theories were developed to explain how a catalyst deposit might affect flow patterns across the blades. At first, consideration was given to uniform buildup on the stator vanes and/or rotor blades. This did bring about some shifts in incidence angles, stage reaction, and pressure profiles, but still was insufficient to cause the failures. Typically, catalyst deposits on rotor blades resulted in higher rotor vibrations; however, in this case it was theorized that these same deposits could have aerodynamic effects on blade loading. Therefore, this CFD analysis was used to evaluate the impact of the actual shape and location of deposits. Figure 27, although not to scale, illustrates the size and shape of actual deposits found in other units. Investigations of expander deposits show that they are always made up of catalyst less than two microns in diameter. Particles of that size approach the blade row with a similar trajectory to the flue gas molecules. However, their weight-to-drag ratio is high so they do not make the 100 degree turn in the blade row. Thus they plow through the turn and pile up on the blade pressure side. There is little buildup upstream of where the deposit is depicted because that surface is shrouded by the leading edge. There is also little buildup near the trailing edge because the adjacent blade casts a shadow over that region. This results in a shallow mound shape that will be in position to bring about a converging diverging (C/D) passageway upstream of the actual blade throat.
Figure 25. Sample Portion of the Cause-and-Effect Diagram. At the end of the classic RCA, the team had crossed off the potential causes without reaching a positive conclusion. This failure did not follow the rules of most of the failures within the experience base of the analysis team. Because the classic modes of failure were being ruled out as the analysis continued, it became apparent that an unusual event must have occurred to initiate the blade failure. The next steps for the root cause analysis team were to look outside classic failure modes. The investigation needed to go deeper into possible causes. This required a focus on scenarios that simultaneously led to high blade stresses in the crack initiation sites while not causing increased stresses at the expected blade failure locations.
SECOND PHASE OF THE RCA

Identifying Potential Loading Scenarios The approach taken was to try to first identify a loading scenario that would create higher stresses on the blade at this specific failure location. A tool was developed to facilitate evaluation of load cases that included centrifugal and alternating gas-bending loads. It was called the stress visualizer, and it permitted more cases to be explored in a short time. This approach provided the flexibility of applying calculated stresses from nonstandard gas loads on top of the more precise FEA steady-state stresses. Figure 26 shows a comparison of stress safety factor for two different load cases. As expected, in the base case (normal flows/loads) the crack initiation site had significant safety margin, while the leading and trailing edges had the lowest safety margin. However, if loads were adjusted, the margin at the crack initiation site could be significantly reduced under potential loading scenarios as shown in Figure 26. The pressure loads in this second
Figure 27. Sketch Showing Typical Size and Shape of Deposits Formed on Rotor Blades.
16
Table of Contents
If blade deposits form a converging diverging (C/D) passageway, the flow can easily reach the speed of sound in the new throat that is upstream of the blade row discharge. In supersonic flow, a diverging passage produces the opposite effect on flow as it does for subsonic flow. The flow downstream of a C/D nozzle minimum area will accelerate, if the conditions are right. A CFD model was then built with a similar deposit shape to that seen in other units, and the results on the first pass were dramatic. Figure 28 shows the Mach number variation for the baseline model with no deposit. In contrast, Figure 29 shows the same plot with a typical deposit where the exit Mach number is 1.9.
with a C/D passage. It can be seen that the static pressure can be as low as 4.3 psia at Mach 1.9. This meant that the rotor blades could be subjected to large variations in pressure loading with variations in catalyst deposit size and shape.
Figure 30. Static Pressure Across Blade with and without Deposit. There is a mechanism for changing the shape of the deposit similar to the phenomenon of snow buildup on a mountainside. In both cases the bond is mechanical and the buildup will continue to grow until the weight of the deposit is too much for the strength of the bond. In the case of a snow buildup, that force is gravity, and the snow pile slides off in the form of an avalanche. On a rotor blade, the catalyst is in a centrifugal force field that is about 11,000 times the force of gravity. Initially the bond strength of the deposit can be great enough to overcome the centrifugal load. However, as the deposits grow in size eventually the increased centrifugal load causes deposits to shed. Figure 28. Mach Number Distribution without Deposit. Introduction of a Sonic Shock Theory into the RCA Because the CFD analysis was able to predict abnormally high Mach number due to blade deposits, the RCA went back through the process data. The intent of the re-review of the process data was to determine if process conditions could help prove or disprove the CFD theory on supersonic flow in the blade row. Available data would not allow for exact measurements of Mach number within the expander. However there were enough data to understand the relative trend in Mach number. Figure 31 illustrates the trend in Mach number for the expander two months before the failure.
Figure 29. Mach Number Distribution with Deposit. To demonstrate the effects on blade surface static pressure, Figure 30 was developed via compressible flow equations. It compares a blade under normal Mach number distribution to one
Figure 31. Calculated Mach Number at Blade Midspan (No Deposits). The calculated Mach number shown in Figure 31 used available pressure data and assumed a clean expander. Based on a clean expander, a threshold redline was plotted to represent when the
flow would have exceeded Mach 1.0 over a significant portion of the blade trailing edge. As shown in Figure 31 the Mach number makes a sudden change upward about three days prior to the failure. This significant change just days before the failure led to the highest Mach number recorded during the entire run of the machine. The data showed that the Mach number remained supersonic for a period of time just before the failure. These process data supported the theory of a high Mach number in the blade row. With the previously discussed review of flow function indicating deposition within the expander flowpath, the RCA was able to add process data support to the CFD modeling. As this theory of deposition creating a C/D passage leading to high Mach number in the blade row progressed, the cause-andeffect trees were updated to reflect the investigation. An example of the cause-and-effect tree for supersonic flow within the blade row is shown in Figure 32.
Monitor cat carryover/distribution Continue frequent walnut hull injection Monitor catalyst quality (particle size)
Assemble an expander reliability team that meets monthly (more frequently if required) and reviews the results of the data generated. This team will have the authority to mandate the appropriate level of flow path cleaning to avoid another failure of this type. In addition, more comprehensive solution development was undertaken as a separate effort. Closure This effort demonstrated that the old reliable classic approaches to engineering might be insufficient to solve problems. The RCA followed well-known, historically proven steps that were unable to generate an answer. Instead, the study team had to employ more sophisticated engineering tools in an almost experimental mode to permit a positive conclusion to be reached. The result, through the use of state-of-the-art engineering tools, was the identification of a very plausible failure mode and cause that had not been considered in the past by the investigators. The future is exciting.
BIBLIOGRAPHY
Balfoort, J. P., 1974, Power Recovery Systems and Hot Gas Expanders, Proceedings of the Third Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 1-7. Figure 32. Sample Portion of the Cause-and-Effect Diagram Addressing Sonic Shock Theories. Drosjack, M. J. and Felten, J. W., 2004, CCU Expander Reliability: A Machinery Issue or a Process Engineering Issue, 2004 NPRA FCCU Conference. ODea, D. M., 1980, Reliable Operation of Flue Gas Expanders, Chemical Engineering Progress, AIChE. Stettenbenz, L. M., 1970, Minimizing Erosion and Afterburn in the Power Recover Gas Expander, ASME Paper No. 70-PET-6.
CONCLUSIONS
Failure Mode The suspected primary cause of the failure was identified as catalyst deposition in a local rotor blade passageway producing a converging/diverging nozzle that drives flow supersonic. This phenomenon can produce buffeting that creates high alternating stresses at the crack initiation site. The stresses alternate because of unsteady shock wave formation at those times when the flow is supersonic or when flow fluctuates between subsonic and supersonic. Unstable supersonic flow can be produced by a single condition or a combination of conditions. For instance, it can happen because of wake disturbances (e.g., broken stator vanes), circumferential variations in catalyst deposits on the shroud, disturbances in the upstream flow, flow variations from day-tonight, etc., either alone or (more likely) in combination. Thus the investigation led to the discovery of this sonic shock theory that could plausibly lead to crack formation in the blades at the location in which this failure occurred. After having developed this potential theory, further review of the process data showing high Mach numbers just prior to failure provided additional confidence in the theorys probability. Future Action In order to minimize the potential for a recurrence of this failure in the future, the analysis team generated the following list of mitigation steps:
ACKNOWLEDGEMENT
The authors would like to acknowledge the dedicated personnel at the Shell companies and Dresser-Rand who contributed to this RCA including Shells Art Tinney (Refinery Reliability Engineering), Jason Kaufman (Refinery Pressure Equipment Engineering), Michael McGranahan (Refinery Operations), Marie Miglin (Metallurgical Engineering), and Anthony Soby (Rotating Equipment Engineering). Also to be recognized for their significant contributions are Dresser-Rands Greg Holland (Aero-Thermo Design), Jason Kopko (Aero-Thermo Design), Paul Chilcott (Solid Mechanics Engineering), and Dan Claus (Metallurgical Engineering). The authors would also like to thank the Shell companies and Dresser-Rand for permission to publish this work. The technical information contained in this presentation is controlled for export by the US Government. Please do not forward the contents without obtaining authorization from the author/presenter.
Monitor the expander for:
Mach number Efficiency Flow function Stack emissions/opacity Vibration
18
Table of Contents
Table of Contents
ADVANCED CONTROL TECHNIQUE OF CENTRIFUGAL COMPRESSOR FOR COMPLEX GAS COMPRESSION PROCESSES
by Kazuhiro Takeda
Research Manager, Research and Development Center
and Kengo Hirano

Instrument and Control Engineer, Plant Design Section Mitsubishi Heavy Industries, Ltd. Hiroshima, Japan
Kazuhiro Takeda is presently the Vibration and Control System Laboratory Research Manager in Mitsubishi Heavy Industries Hiroshima Research and Development Center, in Hiroshima, Japan. He is working mainly on research and development of process control systems for chemical and power plants. Dr. Takeda received a B.S. and M.S. degree (Mechanical Engineering) from Kanazawa University in Japan, and received a Ph.D. degree (System Engineering) from Okayama University in Japan. Kengo Hirano is presently an Instrument and Control Engineer working in Mitsubishi Heavy Industries Turbo Machinery Engineering Department plant design section, in Hiroshima, Japan. He is an I&C specialist for compressor process control. Mr. Hirano graduated from Miyakonojo National College of Technology (Electric Engineering).
Each gas turbine is related to each individual compressor, which has to supply fuel gas for the individual gas turbine according to the load by modulating the control valve opening. The compressor also has to stabilize the common header pressure fluctuation by changing the gas turbine loads. For this process, one control technique is proposed with feed forward (FF) control for the load changing of gas turbine and feedback (FB) control for common header pressure stabilization, as well as application of the above technique in a plant operating in the field. With the FF and FB control, if load shedding of one gas turbine occurs and the gas turbine load suddenly goes down, the other gas turbine can continue operating safely. The compressor control performance is verified with a compressor dynamic process simulator during the design and manufacturing stages prior to shipping. In particular, in order to ensure smooth startup at site, the control panels are connected to the process simulator and the above control technique is applied. Confirmation that all individual functions are working properly is made before shipping.
INTRODUCTION
In recent years, centrifugal compressors have been used for various gas compression processes. In those processes, a conventional control technique, i.e., the combination of master controller and antisurge controller, has proved very effective. But other compressor types, like fuel gas compressors, require quick response and reliable controllability. For these compressors, there is a limit to how far the controllability can be tuned using conventional control techniques. In this paper, an advanced control technique using feed forward control is proposed. This control technique places more emphasis on controllability. First, details of the control technique are introduced. Then the controllability of the control system is checked by studying the dynamic simulation results. After confirmation of results, the control system is applied in the field, where the controllability is verified by studying field test data.
ABSTRACT
Centrifugal compressors are widely used in petrochemical industries and natural gas fields. One of the major control problems associated with these compressors is pressure fluctuation. In recent years, various control methods have been designed and applied in the field to reduce pressure fluctuation and improve the overall operating stability. This paper describes recent advanced complex gas compression control techniques used for centrifugal compressors, and a verification method using a newly developed control system and dynamic simulator. One of the control techniques used for fuel gas compressors that supply fuel gas to gas turbine combined cycle (GTCC) power plants is described in this paper. Two compressors supply the fuel gas to two independent gas turbines through one common header.
25
FUEL GAS COMPRESSOR WITH COMMON HEADER

Figure 1 shows a typical integrally geared fuel gas compressor with inlet guide vane (IGV). This compressor has two compression stages, one gear wheel, and one pinion gear. Each gear is mounted on each shaft. The round arrows show the rotating direction of each shaft, and straight arrows show the gas flow direction.
Table of Contents 26 PROCEEDINGS OF THE THIRTY-FIFTH TURBOMACHINERY SYMPOSIUM 2006
the control philosophy and the operating sequence are simple. When changing the gas turbine is required, the compressor operation can simply be changed simultaneously by reading the header pressure of the related gas turbines. However, when the fuel gas is provided to multiple gas turbines through a common header, the compressor control system will be different. Figure 3 shows an example of a fuel gas compressor arrangement for a GTCC plant, with two gas turbines and three fuel gas compressors connected by a common header. The fuel gas compressor flow capacity is the same as that of the gas turbine. One of the fuel gas compressors is a standby unit. When both gas turbines are under operation fed by fuel gas through the common header, a sudden change in operation by one gas turbine will cause a pressure fluctuation in the common header. The above pressure fluctuation will affect the operating performance of the other gas turbines. Figure 1. Schematic Diagram for Centrifugal Compressor. Table 1 shows the fundamental features of the fuel gas compressor. The handling gas is natural gas from the well. The compressor has two stages, and is driven by a three-phase induction motor. The IGV controls the compressor flow. The seal type is a dry gas seal. Table 1. Fundamental Feature of Compressor.
Figure 3. Fuel Gas Compressor Units Arrangement for GTCC Plant Using Common Header. In this condition, there is a limit to using the conventional control method (like the master controller method for controlling the common header pressure in order to get effective controllability of the common header pressure fluctuation).
COMPRESSOR ADVANCED FLOW CONTROL

Figure 2 shows the compressor arrangement. The main motor drives the wheel gear. The gear train, consisting of the wheel gear and the pinion gear drives the two-stage compressors. The main motor also drives the lube oil pump mounted on another pinion shaft. The gas enters into the first stage of the compressor through the IGV, is compressed in the first and second stages, and then the discharge gas goes out to the outlet section. Figure 4 shows an example of a performance curve of a compressor driven by a constant speed motor with the individual inlet guide vane opened. The horizontal axis shows the flow rate, and the vertical axis shows the pressure ratio. The IGV opens from 20 percent to 100 percent to supply the required flow in this case. When the IGV position is 20 percent opened (at minimum opening point), the antisurge valve (ASV) will open to reduce the flow of the gas turbine. The pressure ratio of the fuel gas compressor is almost constant in this operating condition. Also control valves control the common header pressure, so the compressor operating point shifts toward the dotted line of the performance curve.
Figure 2. Compressor Arrangement. Individual fuel gas compressors are usually dedicated to each gas turbine in a GTCC plant so that both can work as a pair. Hence, Figure 4. Compressor Performance Curve and ASV/IGV Control Band.
Table of Contents ADVANCED CONTROL TECHNIQUE OF CENTRIFUGAL COMPRESSOR FOR COMPLEX GAS COMPRESSION PROCESSES 27
On the performance curve, the IGV and ASV control bands are divided into two by a boundary line, formed by the cross section point of the pressure ratio line (dotted line) and IGV minimum opening line (dotted/dashed line) for ASV control. One band is on the left side of the boundary line and the other band is on the right side. Therefore, at first, the load signals from the related gas turbines are entered in the function generator called feedforward load function. Figure 5 shows the above-proposed control system. Figure 5 (1) shows the schematic control diagram for the compressor unit, and (2) shows the load sharing control diagram.
MV0 is limited to the range of zero percent to 100 percent. If FF value deviates from the actual and design conditions, FB values can cover the full range of FF value according to this formula. The IGV and ASV positions are decided using the split range function generator Fx1 and Fx2. Figure 6 shows an example of feedforward load function. The horizontal axis shows the load signal of the fuel gas flow of the assigned gas turbine. The vertical axis shows the FF signal as a percentage.
Figure 6. Example of Feedforward Load Function (Fxload). There are two boundary lines in Figure 6. One is the boundary line (fine double-dotted/dashed line) for the minimum ASV position, and the other is the boundary line (thick dotted/dashed line) for opening/closing the ASV and IGV as shown in Figure 4. In Figure 6, zero percent load signal means that the gas turbine stops. When the G/T controller gives zero percent load signal, the compressor is in its full recycle operating condition, and the discharge pressure of the compressor should be the same as the set value of the header pressure. The reason for this is that if the gas turbine starts, the compressor has to supply fuel gas to the gas turbine as soon as possible. Therefore, the boundary line for the minimum ASV is used as a limit line to keep the discharge pressure of the compressor above the set value of the header pressure. Figure 7 shows an example of the split range function for the ASV (Fx1) and IGV (Fx2). The horizontal axis shows the output of the calculation (CAL) unit (MV0) and the vertical axis shows the positions of the ASV (MV1) and IGV (MV2).
Figure 5. Compressor Units Load Sharing Control Method by FF+FB/FF Selector. This control system has two operating modes: FF+FB mode and FF mode. The compressor control panel or distributed control systems (DCS) selects these modes with a selector switch (SEL) mounted on the master control panel or DCS. Then the gas turbine selector (SEL) selects the gas turbine and the compressor unit. The output of SEL gives a FB value for FF+FB mode, and a neutral value (50) for FF mode. The FF+FB mode is controlled by the pressure control value (FB) and feedforward load function (Fxload) output value (FF), whereas the FF mode is controlled directly by FF value. This control system avoids any interference between the compressor units. The manipulating value (MV0) is calculated (Equation [1] below) in the calculation unit (CAL) as shown in Figure 5 (1) using FB value and FF value.
MV0 = FF + 2 (FB 50)
Figure 7. Example of Split Range Function for ASV and IGV. Thus, using the combination of the feedforward load function, ASV function (Fx1), and IGV function (Fx2), the ASV controls the low load from zero percent to the boundary flow, by adjusting the FF signal from minimum ASV position to 50 percent open, and the IGV controls the high load from the boundary flow to maximum flow by adjusting the FF signal from 50 percent to 100 percent. Figure 8 shows an example of the compressors load sharing control system by a master controller for the common header pressure. This alternate control system can be used instead of the above-mentioned system as shown in Figure 5. The difference
between the two is that the present control system uses the master controller for the common header pressure instead of the FF+FB/FF selector shown in Figure 8 (2). In this way, the FB value is calculated by the master controller, and is used in the CAL unit of all compressor units shown in Figure 8 (1).
After G/T A keeps the load close to the load condition before G/T B Trip G/T B load suddenly goes down to zero percent at trip and at the same time comp B stops. Case 2: G/T A and B Load Shedding at the Same Time
Before After
G/T A at 100 percent load relates to comp A in FF+FB mode G/T B at 100 percent load relates to comp B in FF mode
G/T A and B load suddenly goes to 30 percent (minimum load) during load shedding Figure 9 shows the simulation result of Case 1. These graphs show trend data of the G/T demand flow, header pressure, compressor flow, and recycle flow. In this simulation, G/T B trip occurs 10 seconds after simulation start. After G/T B trip, A takes care of the G/T load and adjusts the header pressure with the FF+FB function. Comp B stops and decreases the compressor flow by opening the ASV and coasting down.
Figure 8. Compressor Units Load Sharing Control Method by Master Controller for Common Header Pressure. These load sharing control systems are different from common load sharing control system. These load sharing systems concentrate on the controllability rather than the efficiency of the compressors. Fundamentally, the required gas turbine fuel flow modulates the control valves directly using feedforward load function with this control system. The FB value only adjusts the common header pressure to the set point.
SIMULATION
In order to verify the proposed FF+FB/FF selector control system, a dynamic simulation is performed. Some examples of the verification simulation are shown below in G/T parallel operating condition. Case 1: G/T B Trip Figure 9. Simulation Result for Case 1 (G/T B Trip). From the result of this G/T B trip simulation, it is found that the common header pressure fluctuation is within a small range (0.66/0.29 bar [9.57/4.21 psi] from SV), and the pressure became stable in three minutes. Figure 10 shows the simulation result of Case 2. In this simulation, G/T A and B load shedding occurs 10 seconds after simulation start. Comp A and B load goes down to 30 percent with feed forward load function following G/T load, and MV0 becomes small. The ASV opens to increase the flow, so comp A and B
G/T A at 75 percent load relates to compressor (comp) A at 75 percent load in FF+FB mode G/T B at 55 percent load relates to comp B at 55 percent load in FF mode
Before
recycle flow increases and comp A adjusts the header pressure with the FF+FB function.
the dynamic simulator used in the previous SIMULATION section, and run in real time. The switching panel initiates the start permissive condition, trip cause, etc. The PLC, process simulator, and switch panel are interconnected using hard wire cables.
Figure 11. Schematic System Configuration Diagram of Simulation Test Using Dynamic Simulator. Figure 12 shows the simulation test condition using the dynamic simulator to verify the PLC program. In this verification test, the detail sequence and actual control functions can be verified. Table 3 shows examples of the test item using the dynamic simulator.
Figure 10. Simulation Result for Case 2 (G/T A and B Load Shedding at the Same Time). From the result of this G/T A and B load shedding simulation, the common header pressure fluctuation is found to be within 3 bar (43.5 psi) (1.86/1.10 bar [26.98/15.95 psi] from SV), and the pressure became stable in two minutes in spite of the large gas turbine (G/T) load change from 100 percent to 30 percent in a relatively short time. Table 2 shows a summary of these simulation results. Table 2. Simulation Result. Figure 12. Simulation Test Condition Using Dynamic Simulator to Verify PLC Program. Table 3. Examples of Simulation Test Item.
VERIFICATION OF CONTROL SYSTEM USING DYNAMIC SIMULATOR

The FF+FB/FF selection control system is applied to the operating compressor units. The control system is installed in the programmable logic controller (PLC). Before this control system is applied to the actual compressor, the verification test using the dynamic simulator is performed in order to verify the control function and sequence during the factory acceptance test (FAT). Figure 11 shows the system configuration of the test condition. The control system is programmed using a PLC program loader, and is downloaded to the PLC located in the control panel. The operating condition can be confirmed in the operating panel. The simulation model of the gas turbines, compressors, and other auxiliaries is installed in the process simulator that is the same as
An example of the run back test shown in Table 3, Item No. 5, is described as follows. Case 3: G/T - A and B Run Back by Comp Trip
Before
G/T A at 75 percent load relates to comp A at 75 percent load in FF+FB mode G/T B at 75 percent load relates to comp B at 75 percent load in FF mode
After
G/T A and B load suddenly goes down to 30 percent (minimum load) by run back when comp A trips Compressor B supplies the total fuel to gas turbine A and B Figure 13 shows the result of the run back test. This graph shows trend data of the G/T A and B demand flow, header pressure, and compressor A and B flow. In this simulation test, comp A trip occurs 60 seconds after simulation test start. After comp A trips, comp A flow goes down quickly. G/T A and B load goes down to 30 percent load. Comp B flow goes down at once, then goes up to recover the decreasing common header pressure.
Figure 14. Actual Plant Load Shedding Test Result. From this result of the actual test, the common header pressure fluctuation is found to be within 2 bar (29 psi) (1.34/0.61 bar [19.44/8.85 psi] from SV), and the pressure became stable in three minutes. Table 5 shows a summary of this test result. Table 5. Load Shedding Test Result.
Furthermore the feedforward load function was verified in the field. Figure 15 shows the verification result of the function. This graph shows the original design function line, modified function line, and actual operating points. The modified function line is based on the actual operating points, using a linear equation.
Figure 13. Example of Simulation Test Result (Run Back Test). From this result of the simulation test, the common header pressure fluctuation is found to be within 3 bar (43.5 psi) (1.06/1.80 bar [15.37/26.11 psi] from SV), and the pressure became stable in two minutes. Table 4 shows a summary of this simulation result. Table 4. Simulation Test Result for Run Back Test. Figure 15. Feedforward Load Function and Actual Operating Points. The difference between the original and modified function was the molecular weight of the gas. The original function is set by the compressor performance curve based on the design base molecular weight, 16.6. As for the actual gas, methane was less and ethane was more than the design base gas composition, and the actual gas molecular weight was 17.9 (about 108 percent from design). Because the actual gas base molecular weight became heavier than design base, the compressor performance has become better. Therefore the inclination of the function became small, and the IGV control band became narrow for the feedforward load function. If the gas composition and molecular weight change, the feedforward function will actually shift. But in this control system, the FB signal covers the deviation from the design. It is possible to correspond by this control system if the molecular weight change is within 30 percent. This modified function is applied to the actual compressor controller.
ACTUAL CONTROL PERFORMANCE

In the following paragraph, the application of the FF+FB/FF selector control system in a plant operating in the field is described. Figure 14 shows the load shedding test result. This graph shows the trend data of the comp A flow, G/T A demand flow, and header pressure. In this test, G/T A relates to comp A, and comp A is in FF+FB mode. G/T B and comp B are stopped. After G/T A load shedding occurred, G/T Flow decreases. At the same time, comp A flow is decreased by feedforward load function. Header pressure at first goes up then comes down.
CONCLUSION
In this paper, a compressor load sharing control system with a FF+FB/FF selector has been proposed. This control system is mainly provided for a GTCC plant fuel gas compressor. The key factor in this control system is the controllability of the compressor. This control system uses FF+FB mode and FF mode for each compressor unit. The gas turbine fuel flow is input as the FF signal and the control valves are modulated by the FF signal. The FB signal comes from the header pressure controller, and adjusts the control valves. The functions of the control system are verified using a dynamic simulator. From the result of the simulation, it was confirmed that this control system has good controllability. Furthermore, this control system was verified using a simulation test, so it is concluded that the above control system shows good controllability in a plant operating in the field.
NOMENCLATURE
ASV CAL Comp FB FF = Antisurge valve = Calculation unit = Compressor = Feedback = Feed forward
Fx G/T GTCC IGV L/S LSS MV Pd PIC PLC Ps PV Qd SEL SV Td Ts UIC
= Function generator = Gas turbine = Gas turbine combined cycle = Inlet guide vane = Load shedding = Low signal selector = Manipulating value = Discharge pressure = Pressure indicating controller = Programmable logic controller = Suction pressure = Process value = Discharge flow = Selector = Set value = Discharge temperature = Suction temperature = Antisurge controller
BIBLIOGRAPHY
Shinskey, F. G., 1996, Process Control SystemsApplication, Design, and Tuning, McGraw-Hill Publishing Company.
Table of Contents
THE ROLE OF ONLINE AERODYNAMIC PERFORMANCE ANALYSIS

by Robert L. De Maria
Plant Reliability Superintendent Dakota Gasification Company Beulah, North Dakota
and M. Theodore Gresh

Vice President Flexware, Inc. Grapeville, Pennsylvania
Robert L. (Bob) De Maria is Plant Reliability Superintendent at the Beulah Synfuels Plant, in Beulah, North Dakota. He has responsibility for the reliability of the plants rotating and stationary equipment and piping systems. He has 35 years of experience with the specification, maintenance, and reliability of rotating equipment in the petrochemical industry, which includes 20 years of experience at Dakota Gasification Company. Mr. De Maria received a B.S. degree (Mechanical Engineering, 1970) from Stevens Institute of Technology. He has authored several papers and magazine articles on rotating equipment.
program resulting in better maintenance and reliability decisions. Case studies of machine problems and the need to extend the time between overhauls are discussed. The benefits that are realized from implementation of online performance monitoring are presented.
INTRODUCTION
Just like racing driver Jimmie Johnson strives to keep his car in top condition in order to keep out in front of the pack, every plant manager must work hard to keep his plant in top operational condition. It is that little bit of added performance that is the cream, the profits that make the difference between surviving and thriving in todays competitive economic climate. While a good, effective maintenance program is not free, and the cost of maintaining that program cuts into the bottom line, the added revenue from high onstream factors more then offsets this cost. The goal is to have a low-cost program with real benefits. Instant feedback on information with minimal intervention is key. Basing a maintenance program on equipment condition rather than operating time will go a long way toward saving money. While there is always the risk of too much information, knowing how the equipment is running is just as important as knowing plant profits on a real-time basis. Online performance monitoring of critical turbomachinery equipment adds value by knowing instantly when something goes wrong or is starting to go wrong, and the data provided helps scheduling maintenance and troubleshooting for root cause. Getting to problems quickly cuts losses and adds to the bottom line. And, if the equipment is running fine, then why spend the time and money to open it and inspect/replace all the wearing parts?
M. Theodore (Ted) Gresh is Vice President of Marketing and Engineering, Flexware, Inc., in Grapeville, Pennsylvania. He has been involved in the design of high efficiency centrifugal compressor staging, field-testing of compressors and steam turbines, and troubleshooting field performance problems for more than 30 years. While most of this time was in the Technical Services Department of Elliott Company, he is presently with Flexware, Inc., a company focused on consulting services, training seminars, and software for turbomachinery performance analysis. Mr. Gresh received a B.S. degree (Aerospace Engineering, 1971) from the University of Pittsburgh. In addition to numerous papers and magazine articles, he has published a book on the subject of compressor performance, and has several patents related to turbomachinery. He is a registered Professional Engineer in the State of Pennsylvania.
MAINTENANCE AND OPERATIONAL IMPROVEMENTS

In addition to confirming you are getting what the equipment manufacturer promised you would get, online performance monitoring will give you instant feedback for maintaining peak efficiency. While there are other indirect means to monitor compressor efficiency, directly calculating efficiency is the best method to assure the problem is in the compressor and not somewhere else. Monitoring the performance can help with maintenance scheduling such as cleaning (Figure 1) and scheduling a major turnaround. When it comes to considerations of a rerate for increased capacity, the current and historical data will be invaluable in making the decision on how to expand.
55
ABSTRACT
Continuous online aerodynamic performance monitoring of turbomachinery is crucial to plant operation and maintenance and is a key part of a thorough machine reliability program. Integrating machine real-time performance along with mechanical parameters (vibration data, thrust, bearing temperatures, and oil condition) makes for an all-encompassing condition-based maintenance
Online Performance Monitoring System Online performance monitoring utilizes compressor operating data and compares the processed results to the manufacturers expected performance to determine the compressor health. Pressures, temperatures, speed, and flow rates are processed to determine operating work input, head, and efficiency (Tables 1 and 2). These calculated results are then displayed on a chart (Figure 3) and compared to original equipment manufacturer (OEM) expected values. A historical data log (Figure 4) of raw data and calculated results is maintained for use in determining maintenance schedules and troubleshooting problems. Table 1. Input Data for GB1701 Methanation Compressor Pulled out of the Plant Historian Software. Input Data Then Moves Through the Spreadsheet below and Then to the Calculation Engine. Table 1 below Shows Live Compressor Information Being Received and (Table 2) the Calculation Results. Figure 1. Polymer Buildup in the Diffuser Passage of a Cracked Gas Centrifugal Compressor. Online Monitoring Shows the Rate of Decay in Performance and Aids in Scheduling Maintenance Such as Spray Washing or Disassembly for Manual Cleaning. (Courtesy Gresh, 2001) Plant operators are constantly making adjustments to improve the plant processes. Knowing the effect on the compressor via instant feedback from online performance monitoring can expedite their work immensely saving valuable man-hours. Performance monitoring can help with mechanical as well as efficiency issues. Some very basic information gathered from performance monitoring like where the compressor is operating on the performance curve can help prevent impeller failures or at lease help in troubleshooting efforts. Documenting surge events and choke events are vital in finding the root cause of a failure and provide the data necessary to change procedures for preventing future failures. Startups and other transients like trip situations can lead to severe upsets that may not show up as a significant event on the vibration monitors but may push the compressor into deep choke or surge, and unless online monitoring is implemented, will go unnoticed (Figure 2). Improper startup of a refrigeration compressor can result in liquid ingestion and damage to the compressor internals especially the impellers. Normal performance testing will not show these phenomena since a performance test is typically conducted during stable, steady-state conditions.
Mechanical Reliability
Avoid Deep Choke (Stonewall) Avoid Surge Monitor Transients
Table 2. Output Data for GB1701 Methanation Compressor.
Figure 2. Monitoring and Documenting Where the Compressor Is Operating on its Performance Curve Can Enhance Mechanical Reliability. Damage to Equipment Is Known to Occur in the Choke Region as Well as in Surge. (Courtesy Gresh, 2001)
Figure 3. Head, Efficiency, and Work for GB1701 Methanation Compressor. Data Shows Compressor to Be Operating as Expected.
Figure 4. Historical Data Chart for GB 1701 Methanation Compressor Showing Efficiency (Upper Curve) and Delta Efficiency (Lower Curve). In a typical online performance monitoring system, compressor operating data are transferred from a plant data collection system that, for most new plants, includes all the necessary data required for determining the compressor performance such as head, power, and efficiency. Some older plants do not collect all of this necessary data and some pressures and temperatures will need to be added to the existing data collection system in order to implement the online monitoring system. These data are transferred via dynamic data exchange (DDE) to the performance software calculation engine where the data are processed. While some may consider it possible to monitor data by comparing discharge pressure or pressure ratio over time, this is not really a good procedure and can create more confusion than help. A change in inlet temperature or pressure will affect the discharge pressure and even the pressure ratio. So, it is wise to monitor head and efficiency. When calculating head, the best method is by using Benedict-Webb-Rubin (BWR) equations of state. This will provide the best means of comparing calculated results with the OEM performance curves for the compressor. Along with head and efficiency, the work input is also calculated. By monitoring the work input, it is possible to monitor the quality of the compressor data being collected. Work input is representative of the energy transferred from the compressor impeller blades to the gas. Any degradation in the interstage seals, corrosion, or fouling will show up as a loss in efficiency and head, but the work input is fixed and remains the same regardless of these losses. Thus, monitoring work input is a good means to confirm that the input data are accurate. Case History A While online performance monitoring does not in itself enhance the mechanical integrity of the equipment, it is another tool for obtaining additional information otherwise not available or difficult to obtain or decipher. Such information can offer additional guidance to operators for avoiding conditions that could potentially cause equipment failure. Specifically, online performance monitoring provides an easy method of monitoring where the compressor is operating on its performance curve making it easy to assure the compressor is operating within the OEM limits. If a failure does occur, the information available from online performance monitoring will be valuable in the troubleshooting efforts. In 1997, a newly installed propylene refrigeration compressor in southern Louisiana tripped on vibration after just 20 hours of operation (Kushner, et al., 2000). The compressor was a twosection centrifugal with all new internals and driven by a constant speed motor. Inspection of the compressor found that the cause of the high vibration was mechanical failure of the last (fifth) stage impeller. A triangular section of the impeller back plate (Figure 5) was missing and several cracks were found near each vane tip.
Figure 5. Damage at Tip of Fifth Stage Impeller of Propylene Compressor. (Courtesy Kushner, et al., 2000) The second section of the compressor and specifically the last stage impeller were found to be operating well beyond the end of the curve, in deep choke conditions (Figure 6).
Figure 6. Head Ratio Versus Inlet Flow Ratio for Last (Fifth) Stage Impeller of Propylene Compressor. (Courtesy Kushner, et al., 2000) While the compressor was an older two-section machine, the installation was new. The compressor had been rerated for the new site with new diaphragms and new rotor. The motor was oversized significantly for the application so that it could be a spare for another compressor. During startup, liquid propylene was sprayed into the inlet piping upstream of the knockout drums to provide gas to operate the compressor. Unfortunately, the flowmeters, which were new, were installed correctly, but the wrong factor was input into the computer resulting in a calculated flow rate that was significantly less than the actual flow rate. So, when the operators noted the flow was looking low, they were concerned about surging the compressor, and increased the liquid injection to increase the gas to the compressor and bring it further away from surge. Unknowing to the operators, they were pushing the compressor deep into choke. Further adding to the problem was the knockout drum design. The drum was horizontal and did not have demister pads. Since several cracks were found near each vane tip on the toe of the fillet welds, resonance of an impeller natural frequency was at the top of the list of possible causes of the failure. The design of the impeller was a cutback design. The blade outer diameter was less than the impeller hub and cover outer diameter. So, the resolution of the problem was to cut the hub and cover back to the blade diameter, stiffening the impeller and raising the impeller natural frequencies. There were no metallurgical problems found with the failed impeller. The metallurgical examination by the OEM showed that cracking was not a result of preexisting (manufacturing) defects. Stress levels at the crack initiation site were calculated to be 15.4 ksi.
The only impeller resonance found was for plate modes at the impeller outside diameter (OD). While these frequencies could be excited by return vanes at the impeller inlet, they should dissipate greatly, according to the vendor, by the time flow reached the impeller OD where the failure occurred. In the vendors opinion, it would be difficult to have this high excitation unless there were liquids involved. The diffuser did not have vanes that might excite the impeller tip. Flow through the compressors second section during this 20 hour period was calculated to be 7950 icfm. This is 5 percent beyond the end of the OEM performance curve and 15 percent higher than the original rated flow. Borer, et al. (1997), did research that clearly shows that flow within a volute and around its tongue or cutwater causes a nonuniform static pressure field that will act as an excitation force at the impeller OD. While this pressure gradient is present throughout the compressor operating range, computational fluid dynamics (CFD) analysis shows that it becomes stronger in the choke region (Figures 7 and 8). Also, full load testing with straingauges demonstrated that overload dynamic stresses are significantly higher than dynamic stresses near surge.
billion standard cubic feet of natural gas annually. An abundant lignite resource underlying the rolling North Dakota plains supplies the plant with the fuel source. The plant consumes more than six million tons of coal each year (Figure 9).
Figure 9. The Coal Gasification Process. (Courtesy Stelter, 2001) Coal gasification involves dismantling the molecular structure of coal and reassembling the resultant hydrogen and carbon as methane. The heart of this plant is a building containing 14 gasifiers. Lignite is fed into the top of the gasifiers. Steam and oxygen are fed into the bottom of the coal beds causing intense combustion (2200F). The resulting hot gases break down the molecular bonds of the coal and steam, releasing compounds of carbon, hydrogen, sulfur, nitrogen, and other substances to form a raw gas that exits the gasifiers. Ash discharges from the bottom of the gasifiers. The raw gas is cooled and the tar, oils, phenols, ammonia, and water are condensed from the gas stream. Other byproducts are used as boiler fuel for steam generation. Methanation takes place by passing the cleaned gas over a nickel catalyst causing the carbon dioxide to react with free hydrogen to form methane. Final cleanup removes traces of carbon monoxide. Synthetic natural gas leaves the plant through a 2-foot-diameter pipeline, traveling 34 miles south. There it joins a 1249 mile interstate natural gas pipeline system, which transports the gas to pipeline companies. These companies supply thousands of homes and businesses in the United States. In addition to natural gas, the coal gasification plant produces numerous products from the coal gasification process that have added great diversity to the plants output. These products include ammonium sulfate and anhydrous ammonia, which are fertilizers that supply valuable nitrogen and sulfur nutrients for agricultural crops. Other products include phenol for the production of resins in the plywood industry, cresylic acid for the chemical industry, liquid nitrogen for refrigeration and oil field services, methanol for solvents, naphtha for gasoline blend stocks, carbon dioxide for enhanced oil recovery, and krypton and xenon gases for the nations lighting industry. In 2004, this plant began implementation of a continuous online monitoring system for critical machinery because of the following events causing the company +$10,000,000 in lost profits:
Figure 7. CFD Profile of Pressure Gradient Around the Periphery of a Discharge Volute for a Compressor Operating near Surge. Note How the Pressure Is Relatively Uniform. (Courtesy Borer, et al., 1997)
30,000
Figure 8. CFD Profile of Pressure Gradient Around the Periphery of a Discharge Volute for a Compressor Operating in Overload Conditions. Note the Large Pressure Change at the Volute Cutoff. (Courtesy Borer, et al., 1997) Based on this, prudence dictates operation within the boundaries of the OEM curve avoiding choke as well as surge. The Coal Gasification Plant The commercial-scale coal gasification plant in Beulah, North Dakota, began operating in 1984 and today produces more than 54
hp condensing mechanical drive steam turbine was opened for inspection after 15 years of operation and was found to have severe erosion in the nozzle block area resulting in extending the scheduled outage
30,000 hp axial air compressor blade failure resulting in a 50

percent reduction in production (Case History B )
14,000
hp centrifugal gas compressor was partially fouled causing a 10 percent reduction in plant rates (Case History C) Setting Up and Monitoring Compressors While the Beulah coal gasification plant has realized for some time the value of adding online performance monitoring to their
condition-based maintenance program, limited instrumentation has hampered progress. There are 22 critical unspared compressor trains that directly affect the plant revenue stream plus other spared compressor and expander trains. All of these machines are not now monitored continuously because of instrumentation deficiencies. Most of these machine trains have been in service for more then 20 years. As the machine control panels and instrumentation are updated, all of the compressor trains will eventually have continuous performance monitoring. In this interim period, manual data are taken from local instrumentation to calculate machine performance. Data required for online monitoring includes: inlet and discharge pressure and temperature at each compressor section nozzle, flow rate, and speed. Additionally, it is necessary to have an accurate gas analysis and have a copy of the compressor manufacturers predicted performance curves. Gas data are entered into the computer along with the curve data for reference. Accurate data are essential to obtaining good results with any condition-monitoring program. Calibrating all instruments involved will assure accurate data and correct results. One way to confirm the data are accurate is to monitor the compressor work input. If the work input falls on the curve, then the data are most likely accurate. If the measured work input does not fall on the work input curve, then the first thing that needs to be done is to check the accuracy of all instruments. This includes the flowmeters, pressure and temperature readings, and the gas analysis. At the Beulah plant, data are pulled from the plant historian system to a Microsoft Excel spreadsheet. From there the online monitoring system takes the data via DDE connection and processes the data to obtain the appropriate compressor parameters such as head and efficiency. A single point is displayed on the performance curve (Figure 3) and the curves are compensated for speed. A historical trend tracks various data (Figure 4) versus time to show degradation of the compressor efficiency and thus aid maintenance scheduling. Tracking the delta efficiency (the difference between the predicted and the actual efficiency) and delta head is a very good indication of compressor health. Tracking the delta work input (this value should be zero or near zero) is a good way of tracking the overall quality of the online performance monitoring system. Case History B In June 2002, an axial air compressor blade failed catastrophically in the first row of the low pressure section (Figure 10). The compressor is driven by a 30,000 hp fixed speed synchronous motor through a parallel shaft gear increaser at an operating speed of 4530 rpm. This compressor is one of two machines operated in the air separation facility for oxygen production used in the plant gasifiers and thus crucial to plant operation. The other machine is direct driven by a steam turbine. Both machines were placed in continuous operation in 1984. These machines operate at about 140,000 scfm and a discharge pressure of 90 psia. A similar failure had also occurred in the motor driven train in1995. Subsequent to this failure a protective coating was installed on the low pressure blading to protect it from erosion and corrosion. The impact to plant revenue of this second failure prompted an extensive investigation into the root cause.
A thrust bearing failure on the steam turbine train in 1993 resulted in an upgrade to a directed lube tilt pad design in both compressors. In 1994 the speed was increased from 4425 rpm to 4530 rpm, which increased capacity by 12 percent. There were no incidents of high vibration, or high thrust bearing or journal bearing temperatures prior to the blade failure. Analysis of the failed blade showed it to be of the correct material and mechanical properties and was not the result of foreign object damage, corrosion, or erosion. Since only one minute averaged data were available from the plant historian, the operating point on a dynamic basis was not available making it difficult to determine that the machine had been operating in a choke condition. A finite element analysis (FEA, Figure 11) was performed to determine the structural response for the defined load conditions. Metallurgical examination suggested that the failure was due to fatigue associated with a relatively high mean stress. In addition, blade vibrations and plastic strain above the tolerance limit most likely assisted in the fatigue failure. It was the contention of the investigator that the crack leading to the blade failure was initiated from the concave (high pressure) side of the thin blade edge and progressed in the transverse direction (opposite direction to airflow) along about 60 percent of the fracture surface until the blade succumbed to ductile overload (Figure 12).
Figure 11. Finite Element Analysis of Failed Axial Blade. Note the High Stress on the Blade Trailing Edge.
Figure 12. Failed Axial Blade. Crack Initiated on High Pressure, Trailing Edge of Blade. After returning the machine back to service with the spare rotor, high vibration was measured on the inlet air duct with an accelerometer at the blade pass frequency of the rotors first row of blades. Vibration in gs ranged from 1 g to more than 75 gs. Dynamic loading data were recorded from pressure sensors mounted in the inlet ducting and in the compressor balance line. High energy levels were detected at various times suggesting that
Figure 10. Axial Air Compressor Blade Failure. The Second in the Compressors 22 Year Operating History.
the blade design was marginal (Figure 13). This prompted a detailed look to see where the machine was operating on its curve. Observation of fluctuating downstream and upset conditions were absorbed by the motor driven train while the steam driven train was base loaded. This was causing the motor train to intermittently operate in a choke condition (Figure 14). Corrective actions included optimization of blade geometry, higher strength material and tighter blade manufacturing tolerances, and revised operating procedures. The authors believe that a continuous online performance monitoring system could have mitigated this situation or at least assisted in the analysis for root cause.
A scheduled overhaul of the compressor revealed fouling. Random flaking off of the fouling explained the higher vibration. Analysis of the deposits showed it to be predominantly iron oxide. An inspection of the carbon steel suction piping revealed significant corrosion on the bottom of the pipe. The first attempt to mitigate the fouling problem was to install a nonstick coating on the rotors. This did not solve the problem because the coating was eventually abraded and the fouling reoccurred. An analysis of the process gas showed it to be saturated. Furthermore, the piping from methanation to the suction of the compressor is outside and not traced or insulated. Ambient conditions can range from 40F to 110F. It was realized that particularly in the winter, the cold piping was causing condensate to form in the bottom of the piping, which reacted with the gas to form carbonic acid. The acid formed then corroded the steel pipe. The corrosion products were transferred to the compressor by the high gas velocities. A suction knockout pot on the line to the compressor has not been effective in preventing the carryover from reaching the compressor. A glycol water removal unit is located between the first and second compressor cases to knockout the water before exporting the gas. Injection of chemicals to neutralize the acid formation on the compressor suction piping was investigated and found to be very expensive. The current implementation plan is to steam trace and insulate the suction piping to keep the condensate from forming and to improve the efficiency of the suction knockout pot. Figure 15 shows a drop in efficiency prior to cleaning. Similarly, the compressor efficiency returned to normal after cleaning. Figure 16 shows buildup on the rotor. Continuous performance monitoring is assisting with tracking this situation.
Figure 13. Goodman Diagram for Failed Axial Blade.
Figure 14. Axial Compressor Performance Map Showing Normal Operation. System Demands Occasionally Drop Discharge Pressure Resulting in Very Brief Periods of Choke Condition and Eventual Blade Failure. Case History C Synthetic natural gas is compressed by two parallel case steam driven compression trains after methanation for delivery into a natural gas pipeline. The trains operate at a maximum speed of 13,700 rpm. Suction conditions are 250 psig at a suction temperature of 110F. Discharge conditions are 1440 psig max and 110F after final cooling before entering the pipeline. Each train has the capability of delivering 85 mmscfd of gas. During periods of high pipeline pressures, the capacity was noticed to be less then expected causing a reduction in plant rates. Two problems were identified. The first was the suction temperature had increased to 150F. This problem was traced back to a fouled final heat exchanger in the methanation area. The second problem showed a decrease in efficiency of the first compressor case having five stages. An increase in 1 vibration of the first compressor case was also noted.
Figure 15. Synthetic Natural Gas Compressor Showing Drop in Efficiency.
Figure 16. Synthetic Natural Gas Compressor Rotor. Note Fouling in Impellers. Monitoring Compressor Performance Helps Track this Condition and Aids Scheduling of Maintenance.
CONCLUSION
In this day and age of high energy prices, revenue losses caused by machine deficiencies, and downtime at the Beulah gasification plant can exceed $600,000 per day. Over the years we have realized the importance of continuous machine vibration, thrust, bearing temperature, and oil condition monitoring to operating reliable and safe turbomachinery. Unfortunately, this does not tell us everything that we need to know to maximize revenue and minimize maintenance expenses. By integrating machine real-time performance monitoring along with mechanical parameters (vibration data, bearing temperatures, and oil condition), we have realized that better-informed decisions can be made. Online performance monitoring is now a key part of the plants machine reliability program. Key benefits that have been seen include:
a Centrifugal Impeller Using Full Load, Full Pressure Hydrocarbon Testing, Proceedings of the Twenty-Sixth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 111-121. Gresh, M. T., 2001, Compressor Performance, Aerodynamics for the User, Woburn, Massachusetts: Butterworth-Heinemann. Kushner, F., Richard, S. J., and Strickland, R. A., 2000, Critical Review of Compressor Impeller Vibration Parameters for Failure Prevention, Proceedings of the Twenty-Ninth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 103-112. Stelter, S., 2001, The New Synfuels Energy Pioneers, Bismarck, North Dakota: Dakota Gasification Company. BIBLIOGRAPHY Kushner, F., 2004, Rotating Component Modal Analysis and Resonance Avoidance Recommendations, Proceedings of the Thirty-Third Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 143-162. Kushner, F., Walker, D., and Hohlweg, W. C., 2002, Compressor Discharge Pipe Failure Investigation with a Review of Surge, Rotating Stall, and Piping Resonance, Proceedings of the Thirty-First Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 49-60.
Knowing the machine performance immediately significantly aids

points that may have subsequently affected machine condition.
the process of troubleshooting a machine problem and minimizing downtime/loss production.
Trending of machine performance allows review of operating The effects of process changes can be evaluated immediately. Knowing machine performance along with vibration, thrust,
bearing temperature, and oil condition, scheduled maintenance and overhauls can be extended for well designed, maintained, and operated equipment.
Performance
monitoring provides valuable information when justifying an extended time between overhauls to an insurance carrier as well as minimizing insurance premiums.
REFERENCES Borer, C., Sorokes, J. M., McMahon, T., and Abraham, E. A., 1997, An Assessment of the Forces Acting Upon
Table of Contents
THE CONSEQUENCES OF COMPRESSOR OPERATION IN OVERLOAD

by James M. Sorokes
Manager of Development Engineering
Harry F. Miller
Product Manager - Marketing of Turbo Products
and Jay M. Koch

Manager of Aero/Thermo Design Engineering Dresser-Rand Company Olean, New York
James M. (Jim) Sorokes is Manager of Development Engineering, with DresserRand Company, in Olean, New York. He is involved in all aspects of product development for Dresser-Rand worldwide. He previously spent 28 years in the Aerodynamics Group, becoming the Supervisor of Aerodynamics in 1984 and being promoted to Manager of Aero/Thermo Design Engineering in 2001. During Mr. Sorokes time in the Aerodynamics Group, his primary responsibilities included the development, design, and analysis of all aerodynamic components of centrifugal compressors. His professional interests include: aerodynamic design, aeromechanical phenomenon (i.e., rotating stall), and other aspects of large centrifugal compressors. Mr. Sorokes graduated from St. Bonaventure University (1976). He is a member of AIAA, ASME, and the ASME Turbomachinery Committee. He has authored or coauthored more than 30 technical papers and has instructed seminars and tutorials at Texas A&M and Dresser-Rand. He currently holds two U.S. Patents and has two other patents pending. Harry F. Miller is the Product Manager Marketing of Turbo Products at DresserRand, in Olean, New York. His career in turbomachinery began 31 years ago with Dresser Clark, and Mr. Miller has held a variety of design engineering and marketing positions, most recently, being Manager of Development Engineering and Leader of the DATUM Development Team. His prior work experience consists of four years as a mechanical construction engineer for the Pennsylvania Power & Light Company. Mr. Miller received a BSME degree from Northeastern University, and an MBA degree from Lehigh University. His areas of expertise include turbocompressor and gas turbine design and application.
63
Jay M. Koch is Manager of Aero/Thermo Design Engineering, at Dresser-Rand, in Olean, New York. He has been employed there since 1991. Prior to joining DresserRand, he was employed by Allied Signal Aerospace. He spent 14 years in the Aerodynamics Group, before being promoted to his current position in 2005. During Mr. Kochs time in the Aerodynamics Group, his responsibilities included the development, design, and analysis of all aerodynamic components of centrifugal compressors. Additionally he was responsible for the development of software used to select and predict centrifugal compressor performance. Mr. Koch holds a B.S. degree (Aerospace Engineering) from Iowa State University. He has authored or coauthored 10 technical papers.
ABSTRACT
This work describes the potential consequences associated with operating a centrifugal compressor in overload. Nomenclature is offered to explain what is meant by overload operation, and methods that are used by original equipment manufacturers (OEMs) and end users to define overload limits are presented. The paper also describes the conditions that can lead to overload operation. Computational fluid dynamics (CFD) results are used to illustrate the forces acting on an impeller when it operates at very high flow. Finally, this paper suggests considerations that should be addressed when designing (or selecting) an impeller that could be subjected to extended overload operation.
INTRODUCTION
It is common knowledge that operating centrifugal compressors in surge can have detrimental effects on the mechanical integrity of parts. Prolonged excursions into surge have caused damage to impellers, bearings, seals, and other rotating components. The violent nature of surge events makes it necessary to install sophisticated control systems to prevent compressors from operating in surge.
While there are numerous publications in the open literature addressing the consequences of operating in surge, there are few, if any, articles on the potential dangers associated with operating a compressor in overloadthe high flow region of a performance map. In the vast majority of applications, overload operation does not represent any cause for concern. In fact, many compressors spend the majority of their life operating in the high flow end of their performance envelope. Yet, in some applications, overload operation can be just as damaging as operation in surge. This paper will discuss the potential consequences of operation in overload. The primary focus will be the potential damage that could occur in impellers, as they are typically the most expensive and difficult component to replace in a centrifugal compressor. The discussion will begin with definitions of the terms overload or deep overload. These terms mean different things to different people. Therefore, it is important to understand the meaning of these terms in the context of this paper. The focus will next move to the types of situations that can lead to operation in overload. This is followed by an overview of some real-world examples of damage caused by overload operation. Computational fluid dynamics (CFD) and finite element analysis (FEA) results will then be used to provide a clearer picture of the forces imposed on impellers when they operate well above their intended design flow rates. Suggestions are offered on factors that should be considered when designing impellers that may be subjected to prolonged overload operation. Application of the general guidelines outlined in this discussion will in no way ensure the long-term mechanical integrity of the impellers. Given the unique factors associated with each situation, it is imperative that you seek the guidance and direction from engineering. However, proper application of these guidelines may increase the impellers resistance to the forces imposed by overload operation that could lead to unwanted consequences. Finally, conclusions will be offered on the need to be cognizant of the consequences of overload operation when establishing the operating envelope for a new compressor application.
Figure 2. Setting Overload Limit. In the broadest sense, the term overload is used to reference any condition under which the compressors inlet flow exceeds its design flow rate (Figure 2). However, most compressors regularly operate at flow rates higher than design and most users do not consider this operating in overload. To them, this is normal operation. Instead, this group deems overload to be operating at any condition that exceeds the safe or recommended flow limit for the machine. This hazardous region to the extreme right on the performance map is often called deep overload. End users typically defer to the original equipment manufacturer (OEM) to establish an overload operating limit if the OEM believes such a limit is warranted. This is where the difficulties really begin because, unlike surge, compressor OEMs, in general, do not specify an overload limit. Because compressor performance drops off rapidly in the overload region, it is frequently assumed that an end user will not want to operate the compressor in that regime or will lack the driver capacity to do so. As will be seen, this can be a risky assumption to make. The majority of OEMs address the issue of overload operation by putting a clause in the operators manual stating that off-the-map operation must be avoided. In the strictest interpretation, this clause covers both the low-flow and high-flow ends of a performance map, i.e., do not operate at flow rates lower or higher than shown on the map. Still, surge control systems are installed to ensure no (or limited) operation in surge. It is somewhat uncommon to find control systems that limit overload operation. Again, this is quite often the case because the compressor driver cannot provide sufficient power to maintain prolonged operation in overload. On occasion, OEMs will provide an overload limit on the righthand side of the performance map (refer to Figure 2). This limit is set based on a variety of methods that are described in the following paragraphs. In fact, in rare instances, anti-overload control systems have been provided that operate based on algorithms very similar to an anti-surge control system, except that a discharge throttle valve is utilized instead of a recycle valve. Percent of Design Flow One of the simplest methods for establishing the overload limit is to specify the maximum amount of flow as a percentage of the design flow. Such limits are typically set based on experience and are a function of the machine Mach number (U2/A0), gas handled, number of stages, etc. It is well known that a compressor that runs at high machine Mach number or is handling high mole weight gases will have less overall flow range than a compressor that runs at low U2/A0 or handles low mole weight gases (Figure 3). Therefore, the allowable percentage increase from design flow must vary. For example, one might impose an overload limit of 120 percent of the compressor design flow rate for a high mole weight machine and allow as much as 140 percent of design flow for a low mole weight application.
DEFINITIONS
Unlike surge, there is no commonly accepted definition for overload or operation in overload. The compressor aeroperformance maps given in Figures 1 and 2 will be used to illustrate the various definitions. The maps show the head coefficient and polytropic efficiency plotted against the normalized inlet flow coefficient.
Figure 1. Definition of Overload.
Table of Contents THE CONSEQUENCES OF COMPRESSOR OPERATION IN OVERLOAD 65
CONDITIONS LEADING TO OVERLOAD OPERATION

Common circumstances that lead to prolonged operation in overload are:
Loss of parallel compressor or compression train, Performance degradation (i.e., fouling) within a compressor, An undersized compressor (either due to changes in flow
requirements for an existing compressor or misapplication of a new machine),
Alternate operating conditions (i.e., summer and winter conditions in a pipeline application), Unanticipated changes in gas characteristics (mole weight, etc.), Process upsets.
Parallel Compressors Figure 3. Variation in Flow Range with Mole Weight. Impeller Mach Number Some OEMs limit the overload capacity based on calculated values for impeller inlet relative Mach number. Calculations are performed with increasingly larger flow rates and the performance map is truncated (or cut off) when the Mach number in any stage within the compressor exceeds the specified level. Typically, the limit is a calculated Mach number of 1.0, the flow rate at which the impeller throat is choked. However, to be conservative, some OEMs set the overload limit based on an inlet relative Mach number of 0.96 or lower. It is noteworthy that all OEMs do not use the same methods to calculate inlet relative Mach numbers. Further, inlet relative Mach number can be calculated at various locations on the impeller leading edge, i.e., at the shroud, at the mean. Therefore, agreement should be reached between the end user, contractor, and OEM regarding the exact definition of the Mach number being considered. Another approach often employed by end users is to specify a maximum allowable inlet relative Mach number at the design condition for any impeller in a compressor. This method assumes that limiting the inlet relative Mach number at design will ensure a sufficient amount of overload capacity. However, this method does not ensure that the compressor is not operated in a risky portion of the performance map. Drop in Head Coefficient/Efficiency Another popular method for establishing overload limit is based on the rate of change of head coefficient or efficiency as a function of flow coefficient. This method takes advantage of the rapid drop in the compressors performance in the overload region as seen in Figures 1, 2, and 3. Often referred to as stonewall (because of its vertical nature and the fact that it limits compressor flow like a stone wall), the high flow end of the map has a highly negative slope. Some OEMs will limit the overload capacity to the flow rate corresponding to an X percent drop in head coefficient or efficiency for a 1 percent increase in flow. Common values of X fall between 10 percent and 20 percent. An alternate form of this approach limits the overload capacity to the flow rate at which the efficiency or head coefficient is a given percentage of the design flow level, i.e., 15 percent, 20 percent, or 30 percent of the design efficiency or head coefficient. Beyond this point, the performance would be so low as to be unusable in most processes. Section Summary Clearly, it is important that the OEM and end user have a common understanding of how the overload limit was established and the potential implications of violating this limit. The latter will be addressed in subsequent sections herein. Many facilities have duplicate compressors or trains of compressors and the process flow is divided equally (or nearly equally) between these compressors or trains. Each machine in the train is sized such that the compressors operate near the middle of their performance map under normal conditions. However, if for some reason (process upset, regular maintenance, etc.) one or more trains are taken offline, plant operators may attempt to make up for the missing train(s) by forcing more flow through the remaining operational trains. This causes those remaining trains to operate at significantly higher flow rates, pushing them into the overload region of their performance maps. Assuming the drivers have sufficient power (and many do not), it would be possible to operate in the overload region for extended periods of time. The performance would suffer (i.e., low efficiency) but production requirements would be maintained. Excess Driver Power It is considered normal practice by many users to rate gas turbine drivers for the highest expected ambient air temperature expected, or at least an average of the historical high temperatures for that location. This is done so that the compressor and hence the process plant may be operated at close to design output on a hot day, or throughout the hot season. With typical aeroderivative gas turbines rated for a 120F inlet, the power output will increase from 25 percent to 75 percent when the ambient drops to 30F. It is also not uncommon for steam turbines to be rated for 110 percent of the compressor rated power with minimum steam conditions, only to be capable of providing 50 percent or more power when operated at maximum steam conditions. Performance Degradation In a multistage compressor, it is possible to drive the latter stages into overload if there is a reduction in the performance of the earlier stages or if there is a decrease in the mole weight of the gas being compressed. With regard to the former, the impellers or stages in a compressor are aerodynamically matched such that when all are functioning correctly, each operates near the middle of its respective performance map. However, if the performance of the early impellers or stages in the compressor begins to degrade, those impellers or stages will not provide the expected volume reduction. Therefore, any subsequent stages will be forced to swallow more flow than expected. If the performance degradation is substantial, the latter stages will operate in overload. Again, assuming there is sufficient driver capability, it would be possible for the stages to operate in overload for prolonged periods. Again, the horsepower consumption would increase significantly but production could be maintained. The same scenario occurs if there is a change in the gas characteristics such that the volume reduction or overall flow range of the individual stages within the compressor drops. This will
cause the impellers or stages to operate at successively higher capacity relative to design until the latter stages are operating at or near choke. Undersized Compressor/Alternate Operating Conditions/Unexpected Changes At times, compressors are purposely or inadvertently undersized in the selection process. For example, an end user may anticipate a sizeable reduction in flow rate during the life of the compressor. This user may choose, in the beginning, to operate the compressor at the high-flow end of its map, knowing that in coming years the flow rate will be reduced and the compressor will operate nearer the center of its map. Occasionally, end users purchase equipment before finalizing their operating requirements. When the compressors are put into operation, they discover that the inlet conditions and/or gas mixtures are not as expected or that their production requirements exceed those originally projected. These changes can cause the compressor to operate at flow rates that are much higher than anticipated. In short, the compressor operates in the overload region of its performance map. Finally, process conditions may also change during the life of a compressor, causing the flow rate through the machine to increase. The end user may lack sufficient funding to revamp or upgrade the compressor, choosing instead to operate existing equipment in overload despite the low performance.
compressor had operated in overload for long periods. In this case, the dynamic forces because of incidence caused a portion of the impeller blades leading edge to fracture and separate from the impeller. The loss of material produced an unbalance on the rotor and the compressor had to be taken offline for repairs.
Figure 4. Schematic of Blade Leading Edge Fracture While the mechanisms that ultimately led to the impeller fractures were different (more on this in the discussion to follow), generally speaking, the root cause for the fractures was the same operation in overload.
POTENTIAL CONSEQUENCES OF OVERLOAD OPERATION

Two examples are presented to illustrate the potential consequences of overload operation. In both cases, the compressors were known to have operated for extended periods in the overload region. Note: Before proceeding further, it must be noted that end users and OEMs are often very reluctant to provide details on difficulties resulting from off-design operation. End users do not want details of their operating practices made public and OEMs typically do not want any suggestion that their products are anything but perfect. For that reason, limited details will be provided on the two sample cases. The first example is taken from a high-pressure gas reinjection compressor. The compressor experienced repeated impeller fractures. The welded impeller was a low flow coefficient design, having a very short blade height relative to the impeller diameter (low b/r). Essentially, the impeller disk was fracturing at or around the leading edge region of the impeller. In the worst case, the inner portion of the impeller disk separated from the outer portion. Following an investigation by the OEM and end user, it was reported that the impeller had been run extensively in the overload region of its performance map when a parallel train was taken offline. Dynamic forces caused by the combination of high impeller leading edge incidence, and a nonuniform pressure distribution caused by the downstream discharge volute, were sufficient to initiate cracks in the impeller, leading to the fractures. The compressor had to be taken out of service for several days to allow installation of the spare rotor. It should be noted that the forces that led to the fractures were significantly lower at the design flow condition. That is, at design flow, the dynamic forces due to incidence and the volute pressure field were not sufficient to initiate the cracking. Further details on this case can be found in Borer, et al. (1997), and Sorokes, et al. (1998). The second example is from a compressor processing heavy hydrocarbons. The subject impeller was a high-flow coefficient design, implying that the leading edge was quite tall (high b/r). Blade fractures occurred near the impellers leading edge. In a few of the blades, a portion of the blade broke away as indicated by the crosshatched area in Figure 4. Again, based on an analysis conducted by the end user and the OEM, it was found that the
DESCRIPTION OF FORCES
Although the forces resulting from operation in overload have a similar root cause to those found in surge or stall, the true nature of those forces is radically different. The violent forces associated with surge are typically very low frequency (6 Hz or less) and result from the flow reversal through the compressor when the impeller or impellers can no longer overcome the downstream static pressure. When in surge, the inability of the compressor to overcome such pressure is directly related to the increase in incidence or other losses in the compressor components, i.e., impellers, diffusers, return channels, etc. That is, as the flow rate in a compressor is reduced from design toward surge, the angle at which the flow impinges on the bladed or vaned components increases, thus increasing the incidence (or delta angle between the flow angle and the blade/vane angle) (Figure 5). At some point, the incidence angles lead to flow separation or other anomalies that cause very high losses within all of the compressor components, making it impossible for the compressor to overcome the downstream pressure. Because flow moves from the region of higher pressure to a region of lower pressure, the flow reverses direction and goes backward through the compressor. The resulting forces on the compressor internals can be destructive.
Figure 5. Impeller Incidence.
The basic driver of the forces associated with overload operation is also incidence. However, the frequency of the force mechanism tends to be high frequency. Whereas surge phenomenon has a frequency of 6 Hz or less, the forces associated with overload tend to be more on the order of blade passing frequency (1 to 3 kHz), i.e., the number of impeller blades times compressor running speed. This distinction is important as one considers the nature of the potentially damaging forces. As with surge, when the compressor is operated at flow rates above design, there is an increase in blade incidence on the impeller. Eventually, the incidence becomes high enough that, as in surge, the flow separates from the impeller blade. In or near surge, this separation occurs on the suction surface; while in overload, it occurs on the pressure surface of the blade. The flow separation causes a reduction in the effective throat (or minimum) area in the impeller, leading to a significant increase in the flow velocity. If the velocity gets high enough, shock waves will form and choke will occur. The pressure fields associated with shock waves are highly dynamic and cause excessive forces on the impeller blades and walls. These forces alone could be sufficient to damage an impeller. However, when these forces are further exacerbated by other nonuniformities within the compressor flow path, one can rapidly reach conditions that can quickly lead to impeller fractures. Two such situations are described in the following discussion. Blade/Vane Interactions It is well known that interactions occur between impellers and stationary vanes upstream (inlet guidevanes or IGVs) and/or downstream (diffuser vanes). Numerous technical papers have been written on this subject including Kushner (1980), Fisher and Inoue (1981), and Eckert (1999). Summarizing, the upstream or downstream vanes are surrounded by fields of varying pressure through which the impellers rotate causing pressure fluctuations and, therefore, varying forces on the impeller blades. The frequency of the pressure disturbance is determined by multiplying the rotational speed of the impeller by the number of stationary vanes. Of course, if the upstream component comprises multiple vane rows, the vane wakes from all of the upstream vanes must be considered because it may be possible for the wakes from the first row to propagate through the second row and reach the impeller. Problems arise when the frequency of the pulsations align with a natural frequency in the impeller, thereby exciting the impeller, resulting in a resonance condition. When the forces are of sufficient magnitude, it would be possible to initiate a fracture in the impeller. Blade/vane interactions during overload operation are of particular concern because of the high level of energy in the gas stream. In overload, the higher-than-nominal flow rate causes the velocities within the aerodynamic components to be higher than at design. Therefore, the static pressure variation at the exit of the upstream inlet guide vanes will be less uniform than it is at design flow. That is, the high core flow velocity in the IGV passages will cause regions of low static pressure, while the static pressure in the stagnated (or wake) region immediately downstream will be very high (Figure 6). If the inlet guide vanes are sufficiently close to the impeller blade leading edges (as is often the case with full inducer-style impellers), each blade will be subjected to this highly nonuniform static pressure field. Because the magnitude of the pressure variation is higher in overload than at design, the forces during overload operation may be sufficient to create problems, even though the forces at design are not.
Figure 6. CFD Results Showing IGV Wakes. The situation with downstream diffuser vanes is somewhat different. Rather than interacting with wakes, the excitation forces associated with vaned diffusers are a consequence of the pressure fields forming around the diffuser vanes because of potential flow effects (Figure 7). Again, as the impeller blades pass the diffuser vanes they pass through the lobes in the pressure distribution. The resulting variation in pressure imposes a dynamic force on the impeller that could excite an impellers natural frequency. As with the IGV wakes, the magnitude of the pressure variation surrounding the diffuser vanes will be greater during overload operation than at design. This is due in large part to the increased incidence on the diffuser vanes and probable separation of the flow from the vane pressure surface. The increased velocities resulting from the higher flow rates also increase the kinetic energy or velocity pressure (1/2V2). This causes even higher loads across the impeller blades. As a result, the dynamic forces in overload will be greater and might be sufficient to initiate a fracture whereas those at design are not.
Figure 7. CFD Results Showing Pressure Field Around LSD Vanes.
Interaction with Other Pressure Nonuniformities The forces acting on the impeller during overload operation can be further exacerbated by the presence of other pressure nonuniformities in the compressor flow path. A prime example of this is the nonuniformity caused by a discharge volute or collector. Again, several technical papers have been published that document this nonuniformity, including Hagelstein, et al. (1997), and Sorokes and Koch (2000). Essentially, an impeller upstream of a discharge volute or collector will be operating in a nonuniform circumferential pressure field as seen in Figure 8. The impeller can be somewhat shielded from the effect of the volute though the use of vaned diffusers. However, the nonuniformity persists nonetheless.
Wake Effects There is much conjecture as to whether it is possible for an impeller to excite itself into resonance. Although there is no definitive proof that this can occur, it has been theorized that the blade wakes shed by an impeller could provide the mechanism for self-excitation. Under normal operation, the wake regions are sufficiently small and of reasonably limited pressure magnitude (i.e., the difference between the static pressure in the wake region and the static pressure in the core flow is small). Therefore, there is not enough force or delta pressure to cause any difficulties. However, as in the inlet guide vane, the difference between the static pressure in the core flow and static pressure in the secondary zone and blade wakes is higher. It is possible that an impeller blade could interact with the wake shed by the preceding blade (Figure10), resulting in an excitation at blade passing or other frequency (i.e., number of impeller blades times the rotational speed).
Figure 10. Impeller Blade Wakes. Figure 8. Nonuniform Pressure Field Due to Volute. As reported in Sorokes and Koch (2000), such nonuniformities, whether caused by a volute, compressor inlet, or other driving mechanism, further increase the level of dynamic forces acting on the impeller because the circumferential pressure distribution tends to become even more skewed when operating in overload. The dynamic strains shown in Figure 9 are for an impeller upstream of a volute operating at various flow rates. As can be seen, the strains are highest in the overload region of the performance map. It should be noted that these strains were obtained during an extensive test program to identify the root cause of the impeller fracture.
ANALYTICAL METHODS
Computational fluid dynamics can provide valuable insight into the flow physics and the detrimental forces associated with operation in overload. For example, Borer, et al. (1997), used CFD to investigate possible excitation mechanisms in a problematic stage in a high-pressure reinjection compressor. Their work showed that a high static pressure load was occurring near the leading edge of the impeller while operating in overload. The static pressure distribution for the impeller is shown in Figure 11. The root cause of the pressure load was high negative incidence on the impeller blade caused by operating at a high flow rate (i.e., overload). When the impeller was operating nearer design flow, the pressure load was significantly smaller. In short, the analytical results were helpful in identifying the adverse conditions within the impeller that contributed to the fractures.
Figure 9. Variation in Impeller Dynamic Strain with Flow.
Figure 11. Impeller Static Pressure Distribution During Overload Operation.
CFD can also be used to develop pressure loads or net impeller blade forces that can be applied when conducting finite element analyses to assess the structural integrity of designs. CFD can also be used to determine how said forces vary with flow rate. Historically, CFD was limited to providing steady-state (or time averaged) pressure distributions that could be applied as pressure loads in FEA studies. However, in the past decade, advances in CFD have made it possible to generate unsteady or transient pressure distributions. These unsteady pressures can be translated into a force at a frequency and imposed as a boundary condition in an impeller natural frequency analysis. This offers designers the opportunity to assess possible excitation and determine if changes are needed to avoid a resonance problem. High-flow coefficient impellers that may be subjected to overload operation are of particular concern. For clarity, high-flow coefficient is defined herein as any impeller having a flow coefficient, , of 0.12 or greater, where or phi is an internationally accepted, nondimensional flow coefficient relating flow, speed, and diameter. By their nature, such impellers have very tall leading edges, that is, the length of the blade from hub to shroud at the leading edge is large. The excessive length makes it possible for the blade to flex or flutter if it is subjected to a sufficient dynamic force. The impeller blade leading edge is most susceptible to flexing because it is the tallest and often the thinnest portion of the blade. If the magnitude of the vibrations becomes sufficiently large, a fracture could occur. Therefore, it is imperative that designers understand the magnitude and the frequency of the forces acting on such impellers. The above issue is of even greater concern on unshrouded or open impellers. Without a cover or shroud to help hold the blades in position, the blades have significantly more freedom to flex or flutter. As a result, operation near stonewall on open impellers is strongly discouraged. As noted, operation in overload causes high levels of negative incidence on the impeller blade leading edges. This negative incidence causes unbalanced pressure forces between the two sides of the blades. If these forces become dynamic or unsteady because of the presence of upstream IGVs or some other circumferential asymmetry in the flow field, the fluctuating pressure forces could be more than sufficient to induce alternating stresses in the impeller. Assuming the excitation force does not change in frequency, and assuming the frequency aligns with an impeller natural frequency, resonance could occur and ultimately result in the loss of mechanical integrity. As an illustration, the pressures acting on the leading edge of a high-flow coefficient impeller are shown in Figures 12 and 13. The analytical results given in Figure 12 are for the impeller operating at its design flow rate while the results in Figure 13 are for overload operation. In both cases, the plot on the right is for the pressure surface of the blade, while the plot on the left is for the suction surface. Also, the pressure scales are consistent among all of the figures.
Figure 13. Static Pressure Load on High Flow Coefficient Impeller Operating at Overload. As can be seen, the variation in pressure loads is much higher when the impeller is operating in overload. One can note the wide variation in pressure on the pressure side leading edge at overload versus the considerably lower variation at design. Since the pressure on the suction side of the leading edge is fairly consistent between design and overload, it is also clear that the delta pressure suction to pressure surface is higher in overload than at design. Note further that the pressure distribution in Figure 13 bears a resemblance to the schematic of the blade leading edge fracture in Figure 4. This alone is not definitive evidence that a fracture would occur. However, were the pressure forces on the impeller unsteady (which they certainly would be), and were the unsteadiness to be at a frequency that aligns with the natural frequency of the blade leading edge (i.e., the blade leading edge mode), a blade fracture like the one shown in Figure 4 could occur.
DESIGN OR OPERATIONAL CONSIDERATIONS

It is critically important that end users and OEMs discuss the full range of conditions that a compressor will face in operation. For example, if the process engineers or end user know that there is a chance for extended operation in overload, the OEM may want to undertake more detailed analyses of the proposed equipment. These analyses may include detailed CFD and FEA analyses as described above to understand the potential ramifications of overload operation. These analyses can be used to:
Establish the maximum safe operating limit, Help the end user understand the risks of operating in that flow
regime, or
Provide
guidance to the OEM regarding how to modify component designs to minimize the risks by increasing the robustness of the design. Regarding the latter, it may be possible to change the number of vanes in adjacent stationary hardware to avoid natural frequency interference issues. Alternatively, the design or manufacturing technique used for an impeller could be changed to make it less susceptible to interference issues. For example, the OEM may choose to apply more stringent criteria for welds (i.e., weld shape requirements, more rigorous inspection techniques) or possibly resort to single-piece impeller fabrications. Of course, a natural frequency in the speed range is acceptable as long as its response does not exceed allowable limits imposed by the end user or OEM. In short, by having full knowledge of the potential modes of operation, the OEM can undertake efforts to minimize the risks. One additional example illustrates how this happened on a recent compression project. The client, a major oil and gas producer, needed more compression on an existing offshore platform to boost
Figure 12. Static Pressure Load on High Flow Coefficient Impeller Operating at Design Flow.
gas going to shore. Because of the declining pressure nature of the field, the compressors were designed with both the present and several future conditions in mind to allow for change-out of internal parts over the life of the field to accommodate the changing operating conditions. Because the application required very high horsepower drivers (>40 MW), and because the early years of operation had the highest inlet pressures, only a few stages (impellers) were required to meet the specified discharge pressure. This resulted in a fairly high horsepower per stage relative to the end users and OEMs experience for this particular size unit. In contrast, operation in future years at much lower inlet pressures would require more stages (impellers) to meet the same desired discharge pressure. Therefore, in the later years of operation, the horsepower per stage would drop into more a comfortable range. The end user also desired that if one unit had to be taken out of service, the remaining units would need to accept more inlet flow; that is, operated in overload, a flow rate 30 percent to 50 percent higher than the design flow rate. Because the stage power was high at the normal design point, the OEM was concerned about the prospect of even higher powers being absorbed in an off-design mode. When the end user consulted the OEM about the potential operation at higher flows, the OEM responded with an engineering study to determine the magnitude of the dynamic stress. The study included an innovative combination of transient CFD and FEA. The resulting analyses indicated that the original impeller geometry was acceptable but with relatively low safety margins. The OEM then determined that the calculated stresses could be significantly reduced with very minor changes to the geometry of the impeller, notably in the blade to shroud fillet weld. This substantially increased the factor of safety, and providing confidence that the impeller integrity would not be at risk due to overload operation. If the end user had not consulted the OEM, it is uncertain whether or not a problem would have occurred. However, by having an open dialog, and by being proactive in performing an analysis, the risk of a problem was much more remote. Of course, it may not be possible to eliminate all of the risks associated with overload operation. Even the most advanced analyses are but approximations of the real world. Therefore, one cannot be assured that such analyses will capture all of the potentially damaging phenomena within the compressor flow path. Further, even the most robust designs will fail if subjected to the right excitation mechanism, i.e., one that aligns with the natural frequency of the impellers. Even a solid ring would fail if subjected to the right excitation. At some point, common sense must prevail. The end user and OEM must face the reality that the safest approach is to avoid operating in the overload region of the performance map. If the compressor might break if you run there, do not run thereor be prepared to undergo regular overhauls to check for internal damage. End users can employ overload control systems to ensure that a compressor does not operate beyond some agreed upon maximum capacity. These are implemented by incorporating algorithms in the control system that limit driver operation (speed, load, etc.) or restrict the movement of control valves to keep the compressor in a safe region on its performance map. Some argue that such overload controls limit production and decrease profitability. However, when this reduction is weighed against the costs and lost production associated with equipment failure, limiting overload operation does not seem to be a bad choice.
overload has many connotations, so it is important that all parties adopt a common definition. Because one persons overload might well be anothers normal operation, a more rigorous definition must be applied when specifying compressor flow range requirements or discussing how the compressor is being operated in production. It might also be possible to adopt an industrywide standard that defines overload as operating a compressor at flow rates that exceed the maximum flow rates shown on the performance map provided by the OEM. This would put the onus on both the OEM to provide an accurate prediction of overload capability and the end user to properly assess their need to operate at such high flow rates. As seen herein, in some reported cases, prolonged operation in overload can lead to impeller fractures. Overload operation can also exacerbate structural natural frequency interference issues that may exist within a compressor flow path. Forces and/or pressure nonuniformities tend to be greater when operating at flow rates much higher than design. Such increases are caused in large part by the increased incidence levels on impeller blades or adjacent stationary vanes (i.e., vaned diffusers). Advanced analytical tools such as computational fluid dynamics or finite element analysis help quantify the magnitude of the forces associated with overload operation. Such analyses can also be used to mitigate risk. Designs can be modified to reduce the potential for harmful interferences, or operating limits can be derived so as to avoid risky portions of the performance map. However, the most sophisticated analyses and most advanced manufacturing methods cannot eliminate the risk of component failures due to overload operation. Common sense dictates that the most effective way to eliminate such risk is to avoid high risk operating conditions. Simply put, if there is increased risk of mechanical failure by running at a portion of the performance envelope, the risks must be weighed against the potential gains and the cost of maintenance or replacement of the equipment. In conclusion, though not receiving as much attention as surge, prolonged operation in overload can have very detrimental effects on a centrifugal compressor. End users and OEMs alike need to be cognizant of the potential risks associated with operating in this portion of the performance map. End users accept the risks associated with surge and, despite the extra horsepower consumed, often run their compressors on recycle so as to avoid surging the units. As noted, the forces associated with surge and overload are similar, yet the industry has not taken steps to protect equipment from overload operation. Failure to recognize the risks associated with overload operation can have a devastating impact on compression equipment, production, profitability, and engineering careers.
NOMENCLATURE
ACFM A0 b Cm D2 IGV LSD N Q r U1 U2 U2/A0 V W1 = Actual cubic feet per minute = Sonic velocity of gas at impeller inlet = Impeller blade height = Impeller meridional velocity = Impeller exit diameter = Inlet guide vane = Low solidity vaned diffuser = Rotational speed in rpm = Inlet volumetric flow in ACFM = Impeller radius = Impeller inlet peripheral velocity = Impeller exit peripheral velocity = Machine Mach number = Gas velocity = Impeller . inlet relative velocity 700Q = 3 D2 = Gas density
CONCLUSIONS
The most important conclusion to be derived from this work is that, contrary to many commonly held beliefs, operation in overload can subject a centrifugal compressor to adverse forces. In fact, in some circumstances, overload operation can be just as detrimental to component structural integrity as surge. Of course, it is critical that the end user and OEM come to an understanding on the meaning of overload operation. The term
REFERENCES
Borer, C., Sorokes, J. M., McMahon, T., and Abraham, E. A., 1997, An Assessment of the Forces Acting Upon a Centrifugal Impeller Using Full Load, Full Pressure Hydrocarbon Testing, Proceedings of the Twenty-Sixth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 111-122. Eckert, L., 1999, High Cycle Fatigue Cracks at Radial Fan Impellers Caused by Aeroelastic Self-Excited Impeller Vibrations, Part 1: Case History, Root Cause Analysis, Vibration Measurements, Proceedings of DETC99, ASME Design Engineering Technical Conference. Fisher, E. H. and Inoue, M., 1981, A Study of Diffuser/Rotor Interaction in a Centrifugal Compressor, IMechE Journal Mechanical Engineering Science, 23, (3), pp. 149-156, London, England. Hagelstein, D., Van den Braembussche, R., Keiper, R., and Rautenberg, M., 1997, Experimental Investigation of the Circumferential Pressure Distortion in Centrifugal Compressor Stages, ASME Paper No. 97-GT-50.
Kushner, F., 1980, Disc VibrationRotating Blade and Stationary Vane Excitation, ASME Journal of Mechanical Design, 102, pp. 579-584. Sorokes, J. M. and Koch, J. M., 2000, The Influence of Low Solidity Vaned Diffusers on the Static Pressure Nonuniformity Caused by a Centrifugal Compressor Discharge Volute, ASME Paper No. 00-GT-454. Sorokes, J. M., Borer, C. J., and Koch, J. M., 1998, Investigation of the Circumferential Static Pressure Nonuniformity Caused by a Centrifugal Compressor Discharge Volute, ASME Paper No. 98-GT-326.
ACKNOWLEDGEMENTS
The authors acknowledge Mr. Donald Schiffer, Mr. Paul Geise, and Mr. Edward Abraham for their support in the preparation of this paper. The authors also recognize the efforts of Mr. Jason Kopko, Mr. Robert Kunselman, and Mr. Edward Thierman in the preparation of several of the figures included herein. Finally, the authors thank Dresser-Rand Company for allowing them the opportunity to publish this paper.
Table of Contents
FULL LOAD, FULL SPEED TEST OF TURBOEXPANDER-COMPRESSOR WITH ACTIVE MAGNETIC BEARINGS
by Frank Davis
Engineering Consultant Doha, Qatar
Reza Agahi
Consultant Irvine, California
and Randy Wu
Senior Engineer GE Oil & Gas Rancho Dominguez, California
Frank Davis is an Engineering Consultant with ExxonMobil, assigned to Ras Laffan LNG Company Ltd., in Doha, Qatar. He has more than 30 years of experience in the field of rotating machinery. Mr. Davis specializes in project engineering for machinery applications, bid reviews, and machinery acceptance testing. After 20 years machinery engineering experience with Exxon and Mobil Corporations, he was a member of the Project Task Force managing the engineering and construction of four LNG plants in the State of Qatar over the past 10 years. He has developed specifications, completed bid reviews, and followed manufacturing, testing, and participated in the startup of more than 40 machinery trains including 65 MW gas turbine driven refrigeration compressors and gas expander/compressor units. Mr. Davis received a B.S. degree (Mechanical Engineering, 1966) from New Jersey Institute of Technology and is a registered Professional Engineer in the State of New Jersey. Reza Agahi is a Consultant, in Irvine, California. He recently retired after 33 years with GE Rotoflow where he last served as Director of Marketing, Sales, and Commercial Operations. Dr. Agahi has taught in universities in Southern California and has authored more than 40 articles and papers in system engineering and turbomachinery applications. He is the inventor and coinventor of several GE Rotoflow Patents. Dr. Agahi received B.S. and M.S. degrees (Mechanical Engineering, 1968) from Tehran University, and an M.S. degree (1974) and Ph. D. degree (Operations Research and Systems Engineering, 1977) from the University of Southern California.
81
Randy Chih-Chien Wu is a Senior Engineer with GE Oil & Gas Operations LLC-North America, in Rancho Dominguez, California. He began his career with Rotoflow Corporation (now part of GE Oil & Gas) as Thermodynamic Design and Test Engineer for expander and compressor products. In his 19 years of working experience, he has been involved in various disciplines in designing and developing of expander and compressor products. His current responsibilities are focusing on risk review and NPI activities as well as test and commission operation of expander and compressor product. Mr. Wu received a B.S. degree (Physics and Mechanical Engineering, 1976) from Chung Yuan University and an M.S. degree (Mechanical Engineering, 1983) from the University of Nebraska at Lincoln. He is a member of ASME.
INTRODUCTION
The application of the turboexpander in natural gas processing and the petrochemical industry had its beginning at a small gas plant in Southwest Texas where Dr. Judson S. Swearingen installed the first natural gas turboexpander (Swearingen, 1999). Turboexpander technology has developed considerably in the last 40 years. For example:
Advances
in fluid dynamics theory and computational fluid dynamics have made it possible to design a turboexpander with high isentropic efficiency and performance predictability;
Progress in rotordynamics evaluation and modern finite element Increase in demand and economies of scale have resulted in
natural gas processing and petrochemical plants becoming larger and larger (Figure 1) (Agahi, 2003).
analysis capabilities have resulted in more reliable turbomachinery.
Figure 1. Turboexpander Flow Development Since 1960. Three large size turboexpander compressor (TEC) units with active magnetic bearings (AMB) were installed in a gas plant in the early 1990s (Agahi, et al., 1994). Because of some difficulties with the operation of the inlet guide vanes (IGV) at the commissioning stage caused by hydrate formation, oil and gas companies and engineeringprocurement companies (EPCs) were somewhat skeptical about AMB technology and its reliability. Furthermore, with more offshore platforms using turboexpander compressor units, ease of commissioning and reliability were among the major concerns of end users. In order to address these concerns, the end users of turboexpander compressor units began to demand testing of the mechanical and control designs of custom designed equipment, and full load, full speed testing (FLFST) of turboexpander compressor units. The first tests were carried out in the late 1990s (Bergmann, et al., 1996). These FLFSTs used a hydrocarbon gas mixture that was intended to simulate the actual process gas as closely as possible. Full load, full speed testing of a turboexpander compressor with hydrocarbon gas proved to be somewhat difficult and too expensive. Furthermore, obtaining permits to conduct such tests is a challenge for most test facilities. To circumvent some of these issues, a mixture of nitrogen and helium was used in some turboexpander compressor FLFSTs in early 2000. In this approach, a mixture of nitrogen and helium formulated to simulate the molecular weight of the actual process gas is used in a closed test loop.
compressor application should incorporate features to address those deficiencies and that the turboexpander compressor should be tested as thoroughly as possible prior to shipment. Turboexpander compressors now employ magnetic radial and thrust bearings as a solution to the major problem of oil contamination. Thrust calculations and thrust balancing designs have improved and the load can be measured with the active magnetic bearing system. Inlet guide vane controls were updated by the addition of an electric actuator but it was not certain that the control would be as accurate as desired. The active magnetic bearing system, however, lacks redundancy and presents the new concern of a component failure resulting in a loss of levitation under loaded conditions. In order to demonstrate the ruggedness of the design, an FLFST was planned in series with the residue gas compressor for the same plant. The turboexpander compressor package for this project used 240 mm (9.45 inch) active magnetic bearings. Table 1 shows the turboexpander compressor wheel characteristics. This package was to be installed in a closed loop with other LNG train equipment such as a gas turbine (GT) driven residue gas compressor. The project GT provides the power required to drive the residue gas compressor, which in turn delivers the required flow and pressure to the turboexpander compressor to bring the latter up to the full speed, full load condition. Table 2 shows the design guaranteed conditions and Table 3 depicts the simulated conditions for the FLFST. Table 1. Turboexpander Compressor Wheel Characteristics.
BACKGROUND
A major Middle East gas production company was contemplating the development of six large liquefied natural gas (LNG) plants. The designs were to be similar, all based on the use of turboexpander compressor units for maximum production of LNG. The turboexpander compressor power was estimated to be in the range of 9.0 to 10.0 MW, assuming that some of the plants would have units in parallel operation to limit the size of the expanders to modules already demonstrated. The company had experience with similar plants and had identified a large turboexpander compressor as a major contributor to plant outages or production limitations. A thorough review of the problems experienced in the existing plants was conducted with the responsible operating and rotating equipment engineers. The objective of the review was to identify specific design features of the turboexpander compressor that required upgrading from previous designs. Several specific features of the existing turboexpander compressor were identified as responsible for the great majority of equipment outages. In order of importance, these were:
Table 2. Turboexpander Operating Conditions.
Compressor
Design/Guaranteed
Dilution of lubricating oil by process gas and subsequent loss of

oil viscosity resulting in excessive vibration and bearing wear.
Inadequate design of the thrust balancing provisions and poor

operation of the thrust balance mechanism resulting in thrust bearing overload.
Clamping of inlet guide vanes due to lack of controls resulting

in erratic unit performance. The review group concluded that a new, large turboexpander
Table of Contents FULL LOAD, FULL SPEED TEST OF TURBOEXPANDER-COMPRESSOR WITH ACTIVE MAGNETIC BEARINGS 83
Table 3. Turboexpander Compressor FLFS Operating Conditions.
Partial load operation Design/normal speed operation Normal shutdown Emergency shutdown (ESD) due to process upset Two coastdown auxiliary bearing landing tests Verification of critical functions such as ESD
shutdown system trips
and process
THE CLOSED LOOP TEST SETUP

Figure 3 shows a schematic of the test setup and Figure 4 shows the FLFST piping around the turboexpander compressor. The turboexpander compressor unit was integrated into the residue gas compressor (RGC) test loop. The project residue gas compressor was driven by its dedicated GT and delivered test fluid to the expander at 64 barg (928.2 psig) and 50C (122F). The expander extracted energy from the gas stream by expanding to 20 barg (290.1 psig) and cooling down to 21C (5.8F). The cold gas from the expander discharge was then fed to the recompressor, which used the expander power and boosted the test fluid to a discharge pressure of 31 barg (449.6 psig) before returning it to the RGC to repeat the cycle.
FULL LOAD FULL SPEED TEST OBJECTIVES

The main objectives of the FLFS test that were agreed to by the customer and expander manufacturer are as follows:
Verification of the mechanical integrity of the turboexpander

compressor with active magnetic bearings
Verification of the control functions related to the inlet guide

vanes of the turboexpander compressor unit, automatic thrust balancing, and the antisurge valve while operating at FLFS
Identification
and correction of any faults or defects of the turboexpander compressor system and repeat of the FLFS test to verify that the issues were indeed rectified
The FLFST sequence is shown in Figure 2
Figure 3. Closed Loop Process and Instruments Diagram for Full Load/Full Speed Test.
Figure 2. Full Load/Full Speed Test Sequence. The following conditions were to be monitored and recorded:
Startup Near trip speed operation

Full load operation, turboexpander compressor operates at maximum continuous speed (MCS) Figure 4. Test Loop Around Turboexpander Compressor.
EQUIPMENT
The major components and supporting equipment in this test loop are listed below:
Residue gas compressor, centrifugal compressor driven by a GT Residue gas compressor aftercooler Residue gas compressor auxiliary support system Test loop piping Flow measuring devices Pressure measuring devices Temperature measuring devices Pressure transducers, cabling, data acquisition/analysis equipment Startup seal gas supply, instrumentation air, etc. Gas analyzer and reporting system Vibration recording and analyzing equipment compatible with
active magnetic bearing system
The TEC speed was ramped up at 10 percent increments until the speed closely approached the trip speed of 12,400 rpm and remained at that speed for 15 minutes. Then the turboexpander compressor speed was reduced to a maximum continuous speed (MCS) of 11,813 rpm for two hours. The test loop equilibrium state was achieved by slowly adjusting the recycle valves of the recompressor and residue gas compressor. The turboexpander compressor speed was further reduced to the normal speed of 11,250 rpm and operated at this speed for another two hours. At the end of this test run, the turboexpander compressor speed was gradually increased to the shutdown speed and the turboexpander compressor tripped on high speed. The turboexpander compressor rotor coasted down under normal conditions, i.e., the active magnetic bearing system was in levitating mode during coastdown. Table 4 shows a sample of the FLFST parameters that were monitored. Table 4. A Sample of Test Parameters Monitored/Recorded.
Residue gas compressor inline inlet strainer (60 mesh) Noise meter
The auxiliary equipment and components were as follows:
Turboexpander
compressor package with control system, including active magnetic bearing signal interlock with test facility and startup seal gas supply
Expander bypass, Joule Thompson (JT) valve Expander inlet quick shutoff valve Compressor surge control system and recycle valve Check valve downstream of the recompressor Expander inline inlet strainer (60 mesh) Compressor inline inlet strainer (20 mesh)
The majority of the operating parameters such as flow rates, pressures, temperatures, vibration, etc., were monitored and logged by the automatic data acquisition system. Considering the practical aspects of the FLFST, some design parameters could not be simulated. Figure 5 shows process parameters that were different during the FLFST compared to the normal site conditions.
The turboexpander compressor was restarted and the speed was increased to 10,550 rpm for about 15 minutes in order for the system to reach thermal equilibrium. A special method was applied to bypass the active magnetic bearing controller and completely disable the AMB amplifiers in order to activate delevitation.
INLET GUIDE VANE SENSITIVITY AND FLOW CONTROLLABILITY

The turboexpander compressor flow is linearly proportional to the opening of its inlet guide vanes except in small opening and full open positions. By controlling the sensitivity of the inlet guide vanes to within 1 percent, i.e., deviation between process signal to inlet guide vanes and feedback signal to actuator system, it could be demonstrated that the expander flow controllability is within 1 percent of the total flow. As the trended data in Figure 6 show, the differences between the inlet guide vane input signals and the corresponding feedback signals were mostly less than1.0 percent from ramp up to the FLFST condition. It is interesting to note that the injection of additional fluid to increase the test loop pressure did not influence the inlet guide vane sensitivity or flow controllability. The inlet guide vane sensitivity remained within 1.0 percent even when the expander inlet pressure was increased to 64 barg (928.2 psig).
Figure 5. Comparison Between the Design and Test Parameters. FLFST Operation Before startup, the test loop was pressurized to 24 barg (348.1 psig). The test header pressure reached 52 barg (754.2 psig) after the residue gas compressor developed a stable pressure. Flow was admitted to the turboexpander compressor by opening the inlet guide vanes and closing the JT valve simultaneously. The recompressor antisurge valve was at the full open position at the startup. The antisurge valve began closing to load up the recompressor when the turboexpander compressor speed reached approximately 5000 rpm.
Figure 6. IGV Sensitivity Analysis.
AMB AUXILIARY BEARING LANDING TEST

Auxiliary bearings support the turboexpander compressor rotor when it is not levitated. Another function of the auxiliary bearings is to catch the rotor upon loss of the magnetic field resulting in delevitation as the machine coasts down from full load and full speed (FLFS) to a full stop. To demonstrate the functionality of the auxiliary bearings and show their ability to support the rotor upon delevitation at FLFS, landing tests were performed during turboexpander compressor shop tests. To implement this test, the turboexpander compressor speed was increased to 10,550 rpm, and then the expander was delevitated by intervening with the active magnetic bearing control system. The delevitation signal shut off the expander inlet quick shutoff valve and opened the JT valve. The active magnetic bearing controllers were bypassed by introducing jumpers in the control cabinet. As a result, all radial and axial amplifiers were disabled, and the rotor landed on the auxiliary bearings and coasted down to a complete stop. For both landings, it took approximately 2.4 seconds from the rotor landing until the quick shutoff valve shut off; it took 4.6 seconds for the turboexpander compressor to coast down to a complete stop; the rotor landed in the auxiliary bearings for a total of 7.0 seconds; and rotor whirling stopped within 4.0 to 4.2 seconds. Figures 7 and 8 provide detailed records of these tests.
Table 5. Comparison of Radial and Axial Air Gap Before and After Landing Tests.
The tear down and inspection showed that there were light touches on the compressor impeller blade tips. The rest of the rotor and its corresponding stator parts such as the expander wheel, shaft seals, sensor rings, thrust disk, and magnetic bearings were found to have no touch marks and were in excellent condition. There were light marks on the ball bearing inner rings and landing sleeves in both radial and axial surfaces but the ball bearings could roll freely.
ACTIVE MAGNETIC BEARING ROTOR VIBRATION

Before spinning the residue gas compressor for the FLFST, fine tuning of the active magnetic bearings, clearance, and tuning checks were carried out to ensure that the air gaps were consistent with the design values, transfer functions were up to date, and all the required securities were set correctly. The bearing system was equipped with antivibration rejection and automatic balancing system logic. The antivibration rejection activation deactivation limits were set at 3400 and 4600 rpm, respectively. The rotor first critical speed was estimated to be approximately 37 Hz. At higher speeds the automatic balancing system took over the control function. These two systems ensured that the active magnetic bearing rotor always rotates around its inertia center. Figure 9 shows the turboexpander compressor rotor vibration throughout the course of the FLFST including both landing tests. The higher rotor vibration was observed during ramp up, at a speed range between 6,000 to 9,000 rpm. The highest vibration reached 50 m and was mainly in the subsynchronous spectrum, from approximately 37 percent to 44 percent of the synchronous frequency, and occurred only during the startup period. The vibration at this level was considered normal compared to the alarm setting of 90 m. The tangential velocity component, i.e., exit from the inlet guide vanes, could cause swirling around the turboexpander wheel and resonate at subsynchronous frequencies. At FLFST conditions, subsynchronous displacements were almost nonexistent and overall vibration levels were about 15 m. The axial vibration was about 13 m.
Figure 7. First Landing Test.
Figure 8. Second Landing Test. Before and after the landing test, both the radial and axial air gaps between the rotor and auxiliary bearings were measured, compared, and contrasted (Table 5). The data showed that there were no changes in gap dimensions after two landing tests.
Figure 9. Rotor Total Displacement/Vibration.
The active magnetic bearing current chart, Figure 10, shows the current for each bearing during the FLFST. The relatively flat curves indicate that the rotor was quite stable at the FLFST conditions.
Figure 12. Axial Bearing Current. Figure 10. AMB Current. The bearing temperature chart, Figure 11, shows that the bearing coil temperatures were normal. The highest temperature of 70C (158F) was observed when the turboexpander compressor was operated near the trip speed. The rotor length expansion was measured at 150 m. This measurement shows that the relative distance of the rotor and housing was increasing during the test (Figure 11).
EXPANDER INLET QUICK SHUTOFF VALVE TIMING

The project quick shutoff valve was used during the FLFST. Three trips were initiated. One trip was triggered by an overspeed trip at the end of the four-hour mechanical running test, and two trips were performed during landing tests. It took about 2.4 seconds to complete the valve closing process. This duration included the signal transmission time, the valve actuator response time, and the stem traveling time. Based on the factory bench test report, the valve closing time was 0.602 seconds. Therefore, the signal transmission time in the test loop control system took approximately 1.8 seconds.
BEARING HOUSING, SEAL GAS, AND VENT GAS TEMPERATURE MONITORING

The turboexpander compressor seal gas control system was not in operation during the FLFST. The seal gas pressure was controlled manually with a bypass valve. The bearing housing vent gas temperature alarm was initially set at 55C (131F). This setting was too low for the FLFST conditions and had to be revised. The active magnetic bearing high temperature alarm was set at 110C (230F) and shut down was set at 130C (266F). Therefore, the vent gas temperature setting was revised to 95C (203F).
Figure 11. Shaft Extension.
ROTOR VIBRATION DURING TURBOEXPANDER TRIP

One of the FLFST criteria was to observe the rotor behavior during shutdown and coastdown to stop. Upon a request for a turboexpander compressor trip, the normal sequence of the events is to close the quick shutoff valve and open the antisurge valve. Following this procedure, during turboexpander shutdown and coastdown while the rotor was levitated, there were no significant changes in vibration levels, unbalance, temperatures, or bearing currents.
TEAR DOWN INSPECTION

After completion of the four-hour mechanical running and two landing tests, the turboexpander compressor was disassembled for inspection. All parts were in good condition. The auxiliary ball bearings were replaced with a new set despite being in good condition. The turboexpander variable inlet guide vane assembly was also removed and inspected. There were scratch marks on the interfacing surfaces between the guide vanes and nozzle clamping rings as well as on the nozzle cover. These parts were sent to the metallurgical laboratory for determination of the root cause. The conclusion was that debris carried by the gas stream of the test loop was trapped in the nozzle grooves and was dragged into the interfacing surfaces causing the scratch marks when the vanes moved. All scratches were removed and the surfaces were restored before the inlet guide vanes were reassembled.
ACTIVE MAGNETIC THRUST BEARING AND AUTOMATIC THRUST BALANCING SYSTEM

The turboexpander automatic thrust balancing system operated flawlessly in conjunction with the active magnetic bearing thrust bearing control system. There were no indications of axial thrust biases throughout the FLFST. The interlock logic between the active magnetic bearing and turboexpander automatic thrust balancing system was set such that when the axial current reached 14 A, the turboexpander system would take action to open or close the automatic thrust balancing valve to relieve the thrust load. The action automatically stopped when the bearing thrust current was reduced to 12 A. During the FLFST, the valve opening varied from 50 percent to 60 percent. Figure 12 shows the axial bearing current during the FLFST.
FLFST RESULTS
The turboexpander compressor operation during the FLFST was without any major problems or unexpected events. Uninterrupted FLFST continued for four hours at 12.5 MW and 11,250 rpm. The rotordynamics performance was stable and all dynamic parameters were within the design limits and as predicted. The normal trips at FLFS and ESD trips with landing at FLFS into auxiliary bearings were carried out successfully. Rotor coastdown and vibration levels for all trips were smooth and at safe levels. The inlet guide vane test for ease of operation under full pressure, FLFS, ramping
up, and closing down conditions did not show any faults, or indications of blow by or clamping problems. The response of the inlet guide vanes to process signals and controllability were demonstrated to be consistent with other controls in the plant and hence could control plant flow with the desired precision. The automatic thrust balancing systems of the bearing and turboexpander maintained the axial position of the rotor in the desired center position during the various tests and operating conditions. Gas dynamics performance test results were as expected and consistent with the predicted values. Expander isentropic and compressor polytropic efficiencies were better than the guaranteed values.
produced the same performance that was predicted at the design stage and that would have resulted from extrapolation of the open loop air test. Therefore one may conclude that the FLFST does not provide any additional information through the gas dynamics performance tests.
PRESENT CONDITION
The turboexpander compressor package was installed, commissioned, and put into normal operation in early 2006. The unit has been in normal operation since then.
CONCLUSIONS
The end user and EPC for a large LNG facility in the Middle East had specific requirements for the design, manufacturing, and FLFST of a turboexpander compressor package with an active magnetic bearing system. The turboexpander vendor worked closely and diligently with them and incorporated all the special requirements that were consistent with the turboexpander compressor design. The FLFST was carried out in a loop where the residue gas compressor for the same project supplied the required boost for this package. The FLFST was conducted successfully and with relatively few problems. All the specified FLFST criteria were fulfilled satisfactorily. The turboexpander compressor package is in normal operation at the present time.
LESSONS LEARNED
The FLFST requires detailed planning and coordination with the various facilities that are involved. The expander vendor had an earlier FLFST at a facility outside their group and this one was within their group. It turned out that more planning and attention to details were necessary in the latter FLFST because each team tended to leave something out anticipating that the other team would pick it up. Turboexpander compressor units are normally skid-mounted packages ready to be installed on a foundation. This package should be complete with all auxiliary and support systems before it is installed for the FLFST. There were delays, confusion and manual (in lieu of automatic) operation because some components were not shipped to the test site. The turboexpander compressor shutdown loops should be dedicated and have absolute minimum response time to guarantee the safety and security of the turboexpander and its processes. The noise level for this turboexpander compressor was estimated at 85 dBA with a noise jacket installed on the casings. The turboexpander, bearing housing, and compressor did not have a noise jacket during the FLFST and no background noise correction was applied. Noise levels were measured at 115 dBA, some 20 percent higher than calculated/expected. This test helped to highlight the need to review and revise formulas and algorithms used for noise level estimation. The integrity and ruggedness of the auxiliary bearings were tested and demonstrated by multiple landing tests. Inspection of rotor parts after landing tests showed that there was no damage. This could be considered as justification to delete landing tests that normally are requested during open loop air tests. Gas dynamics performance tests of the expander and compressor
REFERENCES
Aghai, R. , 2003, Turboexpander Technology Evaluation and Application in Natural Gas Processing, Proceedings of Eighty-Second Gas Processing Association Annual Convention, San Antonio, Texas. Agahi, R., Ershaghi, B., Leonard, M., Bosen, W. and Brunet, M., 1994, The First High-Power Natural Gas Turboexpander/Compressor with Magnetic Bearings and Dry Face Seals, Proceedings of the 23rd Turbomachinery Symposium, Revolve Technologies Inc., Calgary, Alberta, Canada. Bergman, D. and Nijhuis, H., 1996, Full Power, Full Pressure, Closed Loop Test of a Natural Gas Turboexpander, Proceedings of the Twenty-Fifth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 89-94. Dr. Judson S. Swearingen, 1999, ORBIT, 20, (4), pp. 68-69.
Table of Contents
PROCESS GAS APPLICATIONS WHERE API 619 SCREW COMPRESSORS REPLACED RECIPROCATING AND CENTRIFUGAL COMPRESSORS
by Takao Ohama
President
Yoshinori Kurioka
Application Engineering Manager
Hironao Tanaka
Project Manager KOBELCO EDTI Compressors, Inc. Corona, California
and Takao Koga

General Manager of Sales KOBELCO EDTI Compressors, Inc. Houston, Texas
ABSTRACT
Takao Ohama is currently the President of KOBELCO EDTI Compressors Inc., in Corona California, a subsidiary company of Kobe Steel, Ltd. He is in charge of both oil-free and oil-flooded screw compressors for process gas and industrial refrigeration and manages the company. Mr. Ohamas previous career for 25 years was as an engineer for oil-flooded gas screw compressors and managing the screw compressor engineering group when he was with Kobe Steel Ltd., Japan. He and his staff developed the high-pressure screw compressor H series in 1997, which is the first to be applied to 60 barG in the world as a series and expanded that range to 100 barG. Mr. Ohama also participated on the Task Force for the preparation of API 619 Fourth Edition. Mr. Ohama graduated with a B.S. degree (Mechanical Engineering, 1979) and an M.S. degree (Mechanical Engineering, 1981) from the Saga University, Japan. Oil-free screw compressors have been used for process gas application since the 1970s. Oil-flooded screw compressors have been used in many process related applications since the 1980s. Oil-flooded screw compressors are covered in the latest edition of API Standard 619 issued in 2004. Both oil-free and oil-flooded screw compressors have been expanding into process gas compression applications. It is therefore of interest to present the authors recent experiences and share the acquired knowledge by comparing features with reciprocating compressors and/or centrifugal compressors. High reliability, low maintenance costs, simple foundations, low operational costs, low initial costs, low consumed power at unloaded condition, and suitability for process fluctuation such as gas composition and pressure are some of the basic attributes of the rotary screw compressors. These attributes have resulted in a significant demand for such machines, primarily as an alternate to reciprocating compressors.
Yoshinori Kurioka is currently the Application Engineering Manager of KOBELCO EDTI Compressors Inc., in Corona California, a subsidiary company of Kobe Steel, Ltd. He is involved in proposals for both oil-free and oil-flooded screw compressors. Mr. Kuriokas previous assignment in his 15 year career with Kobe Steel Ltd., was as an engineer for API 619 type bare shaft oil-free screw compressors, R&D engineer for air packaged type oil-free screw compressors, and application engineer for both oil-flooded and oil-free screw compressors. He recently served on the Task Force for the revision of API 619 Fourth Edition. Mr. Kurioka graduated with a B.S. degree (Mechanical Engineering, 1989) and an M.S. degree (Mechanical Engineering, 1991) from the Tohoku University, Japan.
INTRODUCTION
The purpose of this paper is to present the experience acquired in the use of oil-flooded screw compressors in certain process gas compression applications and highlight the key points as compared to other types of compressors. In recent years rotary screw compressors have been applied at higher pressure and larger capacity than before. This paper presents the special features of screw compressors and provides data from actual applications highlighting those features.
HISTORY
In the late 1950s, a Swedish company developed the oil-flooded technique in a screw compressor and perfected the rotor profile to achieve higher volumetric and compression efficiencies. They then licensed compressor manufacturers in the USA, Europe, and Japan to manufacture these compressors and collected royalties. Since the screw compressors have characteristics of both rotary (centrifugal) compressors and positive displacement compressors (reciprocating), such machines found rapid acceptance in petrochemical and gas processing industries. In 1975, API 619 (2004) was introduced to specify a screw compressor. This first edition of API 619 (2004) looked only at oil-free screw compressors. During this period, the oil-free screw compressor was applied in
89
Hironao Tanaka is currently a Project Manager of KOBELCO EDTI Compressors Inc., in Corona California, a subsidiary company of Kobe Steel, LTD. He is involved in engineering the execution of oil-flooded and oil-free screw compressors for various process gas applications. Mr. Tanakas previous assignment in his 12 year career with Kobe Steel, Ltd., has been a system engineer and a project engineer for oil-flooded compressors. Mr. Tanaka graduated with a B.S. degree (Mechanical Engineering, 1992) and an M.S. degree (Mechanical Engineering, 1994) from Kobe University, Japan.
many unique applications such as butadiene, styrene monomer recycle gas, linear alkyl benzene, soda ash, etc. Most of these applications are subject to dust and liquids that are likely to be present in the gas stream. In many cases, water injection was used to control the compression process. In the 1980s, oil-flooded screw compressors started appearing in process gas applications. Around the same time, cogeneration started to take off with gas turbines becoming necessary in more and more applications. Also, oil-flooded screw compressors were finding their way into light gases such as helium and hydrogen. Less sensitivity to changes in molecular weight made such compressors particularly suitable for hydrogen pressure swing adsorption (PSA) compressors. On helium and hydrogen feed compressors, stringent oil carryover requirements made it necessary to introduce activated charcoal absorbers in the oil management system. Carbon dioxide compressors for the beverage industry switched to oil-flooded screw compressors with an oil removal system down to 10 parts per billion (ppb) by weight. In the 1990s, the demand for higher volume oil-flooded and oil-free screw compressors resulted in compressor manufacturers designing and building machines in large frame sizes. By the mid 1990s, high pressure oil-flooded screw compressors started to find their way into fuel gas boosters and many petrochemical and refinery applications. At the same time oil-free screw compressors were finding strong acceptance as vapor recovery compressors in both offshore as well as onshore applications.
Centrifugal compressors are suitable for large flowrates. Screw compressors are suitable for the following conditions.
Pressure ratio limitationsSince it is a positive displacement type compression, and has no valve movement, a high pressure ratio can be achieved. On oil-flooded screw compressors, there is no mechanical limitation for pressure ratio. The only concern is efficiency. Capacity controlOil-flooded screw compressors have an unloader called a slide valve and can provide stepless turndown (typically 100 percent to 15 percent) with corresponding reduction in power. Impact of molecular weight of gasesThere is almost no impact
of molecular weight of the gases upon the performance of an oilflooded screw compressor. Injected oil is a sealant and leakage is controlled. Therefore these compressors are highly efficient for even the lowest molecular weight gases.
Gases
containing dust and polymersIn oil-free screw compressors, any type of gas can be compressed. This is practical because compression is done by displacement with continuos rotation, the rotor shaft is rigid so that effect of unbalance is limited, and there are no internal valves to hinder the operation from dust and polymers. is the same as centrifugal machines and allows single machine operation without a spare in critical services.
AvailabilityHigh reliability resulting in compressor availability

GENERAL DESCRIPTION OF OIL-FREE SCREW COMPRESSORS
GENERAL DESCRIPTION OF THE THREE TYPES OF COMPRESSORS

Before introducing actual applications, one needs to understand the compression mechanism and typical mechanical limitation for centrifugal, reciprocating, and screw type compressors. Centrifugal compressors are continuous flow machines in which one or more rotating impellers accelerate the gas as it passes through the impellers, which are shrouded on the sides. The resultant velocity head is then converted into pressure. This occurs partially in the rotating element and partially in the stationary diffuser. Reciprocating compressors are positive displacement machines with a piston compressing the gas in a cylinder. As the piston moves forward it compresses the gas into a smaller space, thus raising its pressure. There are two types of reciprocating compressors, called lube type with oil injection and nonlube as oil-free. Screw compressors are also positive displacement machines but rotating twin rotors act as pistons that compress the gas in a rotor chamber (casing). Compression is done continuously by the rotation of the twin rotors. There are also two types of screw compressors: the oilflooded type with oil injection, and oil-free with no oil injection. Pressure, flowrate, and gas composition are the major factors to be considered in selecting the type of compressor. Table 1 shows comparison of three types of compressors with respect to pressures, flows, and gas compositions, etc. Table 1. Comparison Table of the Three Types of Compressors.
A cutaway drawing of a typical oil-free screw compressor is shown in Figure 1. There are two rotors inside the casing of the screw compressor. One rotor is referred to as male, and the other rotor is the female. The male rotor and the female rotor maintain a small clearance and do not contact each other. To keep phase with each other, a timing gear is furnished to drive the other rotor.
Figure 1. Typical Cutaway Drawing of Oil-Free Screw Compressor. To isolate the rotor chamber from the bearing with an oil atmosphere, seals are furnished next to the rotor lobe on each end of the machine. There are journal bearings outside the seal area, which are typically sleeve type hydrodynamic bearings. Thrust bearings are located on the outer side of the journal bearings, and tilting-pad type is typically used. The following are the major characteristics of the oil-free screw compressors: Generally, reciprocating compressors are suitable for high pressure ratios, low flow, and low megawatt (MW) applications.
Process gas is completely free of oil, there is no contamination, and therefore any gas can be handled. In oil-free screw compressors, due to
the positive displacement compression, even polymer gas or dirty gas can be compressed.
The rotor speed is higher than with oil-flooded screw because of no oil turbulence in the rotor chamber, but does not exceed any critical speed since the rotor shaft is to remain rigid. Rotor speed is typically higher than an oil-flooded screw machine so compressor frame size can be smaller than the oil-flooded type.
Discharge temperature is typically high because of compression heat. To avoid excessive heat deformation, cooling is required. Some applications utilize a process compatible fluid such as water or solvent to cool the gas directly by injection into the rotor chamber inlet.
Due to its longer rotor span for seal area, rotor clearance, and limits on discharge temperature, pressure ratios are limited for the oil-free screw compressors.
Because of its high rotational speed, noise is rather high so that
silencers on suction and discharge nozzles are typically required. Expansion and/or absorption type silencers are typically used in combination or separately. Frequency of the noise is high because its main frequencies are pocket passing frequency (rotational speed*lobe number) or its harmonics. The major noise is measured at discharge piping. In the authors experiences it is apparent that the expansion is good for several discrete frequencies. The size of the expansion type silencer can be optimized by using a simulation to target the specific frequencies, i.e., pocket passing frequency and its harmonics. To absorb this high frequency noise, internal absorption type silencers are considered to be more effective than external absorption type. Absorptive method is effective in abosorbing pulsation energy of frequencies ranging from 500 Hz to several thousand Hz. Further experience confirms the use of a combination of absorption and expansion type silencers to be more effective in noise reduction. By expansion type, 15 to 20 dB of sound pressure level inside the piping can be reduced whereas, by internal absorption, 25 dB of sound pressure level inside the piping can be reduced. In addition to the silencers, a noise enclosure enclosing just the compressor and gearbox is typically required if there is a sound requirement of 85 dBA at 1 m (3 ft) from the compressor skid edge.
and sealant in the rotor chamber. Typically, the male rotor is driven by a directly coupled two-pole or four-pole electric motor and drives the female rotor. An external gear unit is typically not used since the tip speed of the oil-flooded screw compressor is in the proper design range when driven at motor speed. Since oil is injected into the rotor chamber, the seal area between the lobe and bearing is no longer necessary. There is one mechanical seal located at the drive shaft end. There are typically sleeve type journal bearings on either end of the rotor lobes. Thrust bearings are typically tilting-pad type and are located on the outer side of the journal bearings. The oil and gas mixture is discharged through the compressor discharge nozzle into an oil separation system located downstream of the compressor. Oil separated in the oil separation system is circulated in the compressor lube system. An unloaded slide valve is located in the compressor just beneath the twin rotors and is used to adjust the inlet volume. The inlet volume of the compressed gas can be adjusted by moving the slide valve, which is actuated by a hydraulic cylinder. A typical schematic diagram for an oil-flooded screw compressor is shown in Figure 3.
GENERAL DESCRIPTION OF OIL-FLOODED SCREW COMPRESSORS

A cutaway drawing of a typical oil-flooded screw compressor is shown in Figure 2. There are two rotors inside the casing as with the oil-free screw compressors. However, here they contact each other at lobe surface via an oil film.
Figure 3. Typical Schematic Diagram for an Oil-Flooded Screw Compressor. Compressor lubricant oil is present in the process side, so the lube oil selection is very different from other types of machines. The bulk of the oil is separated in the primary oil separator, but a secondary coalescing oil separator may be used as an additional separator. Separation of oil is one of the important factors for oilflooded screw compressors. Typically, a combination of demister mesh pad and coalescing elements are used. For example, 0.1 parts per million by weight (ppm wt) level can be achieved by combination of a demister mesh pad and two stages of coalescing elements. Charcoal absorbers are occasionally used for more severe applications. Borocilicate microfiber is a typical material used in coalescing elements and submicronic particles of oil can be separated from the compressed gas. Unlike reciprocating compressors, oil from the compressor has no deterioration by piston rubbing so oil can be recirculated in the
Figure 2. Typical Cutaway Drawing of Oil-Flooded Screw Compressor. Oil is supplied not only to the bearing and seal, but also to the rotor chamber directly and oil will act as lubricant, coolant,
system as lubricant for longer life. The lube oil circulation system consists of compressor lube lines, oil cooler, oil filters, and oil pump. The oil pump may be double or single configuration. The design of a single pump system may be used when the pump is required only during startup. In such case, after the compressor starts and discharge pressure is established, oil can circulate in the system by utilizing gas differential pressure between suction and discharge. A slide valve is used to load and unload the compressor to maintain suction pressure or discharge pressure. There is a spool valve actuated by air with solenoid valves to switch over the oil lines to pressurize the slide valve cylinders to the load side or the unload side. A typical control range by slide valve is from 15 percent to 100 percent stepless by inlet volume. Below is a list of some of the major characteristics of the oil-flooded screw compressor:
Figure 5. Typical Cutaway Drawing of a Tandem Arrangement Oil-Flooded Screw Compressor.
Power consumption savings by a built-in slide valveThe slide valve as unloader adjusts the inlet volume of the compressor, and this equates as power savings. Figure 4 shows the basic principle of the slide valve mechanism. The slide valve is located just beneath the rotors and moved in the axial direction. The slide valve is moved typically by hydraulic cylinder with oil utilized from the compressor lube oil line. Moving the slide valve to the suction side attains full load, and unloading is achieved by moving the slide valve toward the discharge port. At full load position, the entire length of the rotor is utilized to draw the gas so that inlet volume of the compressor can be maximized. By moving the slide valve to the unloaded position (i.e., discharge side), the length of the compression chamber is shortened. As a result, inlet volume of the compressor is reduced. Compression is done with less inlet volume of the compressor so that theoretical brake horsepower is reduced.
Low maintenance costDue to the lube oil system the rotors and many other parts of the compressor have an oil film on their surface. The life of the rotors is long enough so that a spare set is not required. The mechanical seal is typically one per casing, so maintenance and replacement cost for the seal are typically reduced.
are integrated and packaged on a single skid. Thus, transportation and installation are completed in a short period.
Single skid arrangementThe compressor and lube oil system
No cooling water jacket/no gas bypass coolerSince oil acts as
coolant in the compression process, discharge temperature can be controlled by the oil injection flowrate so that the casing structure is made simpler by elimination of a cooling water jacket. The gas bypass cooler can also be eliminated by oil cooling.
process gas. Not only mineral-based oil, but synthetic oil has recently been used to expand the application range of oil-flooded screw compressors. Hydrotreated mineral-based oil has typically been used, but recently many are changing to synthetic oil. There are two kinds of synthetic oil: one is polyalphaolefin (PAO), and the other is polyalkylene glycol (PAG). With PAG there are several kinds of oil that differ in ratio of propylene oxide (PO) and ethylene oxide (EO). For a process with a heavy hydrocarbon, both mineral-based oil and PAO are subject to dilution; however, less dilution can be expected with PAG. There is no difference for dilution ratio by process with heavy hydrocarbon between mineralbased oil and PAO; however, less dilution can be expected for PAG. Dilution rate Mineral oil PAO PAG(PO)PAG(EOPO)PAG(EO) PAG with EO 100 percent is hygroscopic; however, it has no dilution for heavy hydrocarbon. By using PAG oil, oil-flooded screw compressors are now able to be used for heavy hydrocarbon applications as in refinery services.
Selection of oil is driven by the need to be compatible with
Figure 4. Basic Principle of the Slide Valve Mechanism.
High compression ratio limitationSince the oil acts as a coolant and sealant the limit on compression ratio is very high. Discharge temperature can be adjusted by oil flowrate, i.e., oil can be injected into the rotor chamber to absorb the compression heat in the oil-flooded screw compressors. When a very high pressure ratio is required, a tandem arrangement of two stage compressors combined in one casing is available. Typically, this tandem arrangement is used when the pressure ratio is larger than 7:1, and can be applied to ratios of more than 50:1. A typical cutaway drawing of a tandem arrangement oil-flooded screw compressor is shown in Figure 5. Since oil will act as a coolant at the intermediate stage, an external intercooler with piping for intermediate stage is unnecessary.
GENERAL COMPARISON BETWEEN DIFFERENT TYPES OF COMPRESSORS IN SOME APPLICATIONS AND RECENT SITUATIONS
Hydrogen Service Hydrogen is widely used in oil refining processes and many processes in petrochemical fields. Hydrogen is typically generated in pressure swing absorption, membrane, or electrolyzing systems. Hydrogen generated by the above methods is usually produced at atmospheric pressure and then compressed typically up to 30 barG (435 psiG) by compressors. Due to the very low molecular weight of hydrogen and high pressure ratio needed, centrifugal compressors or oil-free screw compressors are rarely used for such applications. Reciprocating compressors have been typically used in this service. Due to the advantage of low maintenance cost, oil-flooded screw compressors are increasingly being applied for
Table of Contents PROCESS GAS APPLICATIONS WHERE API 619 SCREW COMPRESSORS REPLACED RECIPROCATING AND CENTRIFUGAL COMPRESSORS 93
this application. Table 2 shows comparison for oil-flooded and reciprocating compressors in typical hydrogen service. Table 2. Typical Comparison Table for Hydrogen Service.
carryover to 1.0 ppmin some applications by adding charcoal absorbers. Less than 50 ppb carryover by weight is achieved. Vapor Recovery Unit (VRU) In most offshore platform applications, crude oil or natural gas drilling produces vapor gas as a by-product. This vapor by-product needs to be recovered for environmental reasons. As a result, vapor recovery units together with compression systems are used. The typical gas composition and operating condition is shown in Table 3. Table 3. Typical Gas Composition and Operating Condition for VRU.
As shown in Table 2, the reciprocating compressor has an advantage of total brake horsepower (BHP) due to multistage compression with an intercooling system. However, the oil-flooded screw compressor has the slide valve to save power at the unload condition. Figure 6 shows the typical package for hydrogen service with an oil-flooded screw compressor.
Gas composition of the recovered vapor can change due to well location and the age of the well. Even from the same well, the vapor gas composition and flowrate can fluctuate. Centrifugal compressors have difficulty in this application because of unsteady gas composition and flowrate. In recent years, lower costs have increased the use of oilflooded and reciprocating compressors in this application. In comparison with reciprocating compressors, oil-flooded screw compressors are more widely used due to their gas flow adjustment capabilities, which can be adjusted by the internal slide valve. However, the unpredictable gas composition sometimes contains serious amounts of sulfur, tar, or other unknown corrosive components as well as heavy hydrocarbons that are always present. Also, there are some difficulties in using oil-flooded compressors due to serious dilution of oil. Oil-free screw compressors are being increasingly used where the specific heat coefficient (k value) is rather small and the discharge temperature is lower for higher pressure ratios. A typical package with oil-free screw compressor for VRU is shown in Figure 7. As shown in the picture, the skid needs to be very compact due to restriction of space, which is also a very important factor on VRU.
Figure 6. Typical Package for Hydrogen Service Using Oil-Flooded Screw Compressor. Oil-flooded screw compressors have an advantage due to the smaller amount of installation area needed and less weight to support. In the case of a tandem arrangement, which is two stage compressors arranged in one casing, compact skid arrangements can be adopted on oil-flooded compressors. Another advantage of the oil-flooded screw compressor is longer times between maintenance. The typical maintenance period for a reciprocating compressor is one to two years, while an oil-flooded screw compressor is two to four years. Reciprocating compressor applications typically require a spare compressor, so investment and installation costs are doubled. The gas industry field requires a longer maintenance period such as two to four years. Equipment demands require continuous operation, and the oil-flooded screw can meet this demand. Oil carryover from the oil-flooded screw compressor is managed by oil coalescing systems, which can reduce
Figure 7. Typical Package for VRU Using Oil-Free Screw Compressor.
Lube oil does not come in contact with process gas within oil-free screw compressors. Therefore, there are no dilution problems. In addition, heat insulation and electronic heat tracing are required to avoid condensation of gas in oil-flooded screw compressors when the compressor is not running. During operation, process gas temperatures need to be kept higher than dew point to avoid dilution of oil in oil-flooded screw compressors. In the case of an oil-free screw compressor, there is no concern due to condensation of gas. Therefore the overall system is simple. Regarding the gas flow change, oil-flooded screw compressors have an advantage with adjustment by the internal slide valve and power savings. However, the rate of change of the gas flow is very slow, typically 20 to 30 years of operation, and generally changes over the life of the field. Oil-free screw compressors can accommodate this change by adjusting the operational speed. Replacement of gear and pinion combinations in a speed increasing gearbox makes this procedure possible. These parts are interchangeable and can be replaced and maintained. A comparison table between oil-free screw compressors and oil-flooded screw compressors is shown in Table 4. Table 4. Comparison of Screw Compressor Features for VRU Between Oil-Free and Oil-Flooded Types.
Figure 8. Typical Unload Curve When Mass Flowrate is Constant with Suction Pressure Change.
Fuel Gas Booster for Gas Turbines Recently, the efficiency of generating electrical power by gas turbines has been significantly improved. High efficiency type gas turbines are used in many power plants utilizing natural gas as fuel. Many gas turbines require higher supply pressure of the fuel at typically 30 barG to 50 barG (450 psiG to 725 psiG) and natural gas pressure coming out from the pipeline is low. To boost the fuel gas to the required pressure of the gas turbine, a fuel gas booster is required. For the fuel gas booster application, reciprocating compressors and centrifugal compressors have been used primarily. In the 1990s, high pressure oil-flooded screw compressors were developed and started to be used for fuel gas booster applications. The oil-flooded screw compressors are very suitable for these applications, since the requirement of the fuel gas booster fits very well with characteristics of oil-flooded screw compressor, i.e.:
Suction pressure fluctuations Gas turbine load fluctuation, i.e., flowrate fluctuation Unstable gas composition (typically pipeline quality natural gas) Figure 9. Typical Unload Curve When Suction Pressure is Constant with Mass Flowrate Change. Oil-flooded screw compressors can be operated at higher suction pressures by utilizing a slide valve with less brake horsepower required in the unloaded condition. Reciprocating compressors and centrifugal compressors cannot accommodate the big fluctuation of suction pressure so a suction control valve is typically required to control suction pressure close to design point at the compressor inlet. Thus, large power energy savings cannot be acquired with these machines. The flowrate is typically rated with some range for gas turbines, since consumed fuel gas flowrate is varied by
Also, this fuel gas booster application requires economical operation and the oil-flooded screw compressors with a slide valve as an unloader can provide significant power savings. Because of suction pressure fluctuations, the compressor needs to be sized according to the design point, which is the lowest suction pressure in specification. However, the actual suction pressure is typically higher than the design point so that the compressor is always operated in a partially unloaded condition. A typical unload performance curve is shown in Figures 8 and 9.
atmospheric temperature. As a result, the compressor needs to be capable of operation at lower flowrates than the rated point. The oil-flooded screw compressor can meet this demand by using the slide valve with power savings as well (refer to Figure 9). This application typically requires a suction scrubber since unexpected water or liquid may be present in the gases. Gas turbines require very precise delivery pressure of fuel gas. Discharge pressure needs to be maintained regardless of the suction pressure swing so that a spillback (bypass line) is always required for quick load and suction pressure changes. The screw compressor slide valve control system accommodates these large changes. In addition to the above, automatic operation is required with gas turbine operation so that a control panel with programming is typically required. Figure 10 shows a typical package for a fuel gas booster unit using an oil-flooded screw compressor.
As shown in Table 5, brake horsepower at the design rated point has almost no difference among three types of compressors. However, there is a large difference at normal operation point and when the suction pressure is higher than design and less flowrate. From a cost and installation point of view, the oil-flooded screw compressor has significant advantage for such applications. Desulfurization Compressor Recently, demand for desulfurization of vehicle gasoline and diesel fuel is increasing all over the world. New regulations to protect the environment have forced the oil refinery industry to develop a desulfurization process. For this process, gas compression is necessary mainly utilizing a hydrogen mixture. The oil-flooded screw compressor has been proven in this process, and demand for the screw compressor is increasing in this application. Table 6 shows a typical comparison of compressors for desulfurization process between oil-flooded screw, centrifugal, and reciprocating compressors. Table 6. Typical Comparison Table for Desulfurization Compressors.
Figure 10. Typical Package for Fuel Gas Booster Service Using Oil-Flooded-Screw Compressor. Except for large size machines, all equipment can be mounted on a single skid, including the oil separation system, suction scrubber, spillback line, and control panel. Sometimes, the compressor forward bypass line is provided when maximum suction pressure is above the discharge pressure. A typical comparison of the screw compressor features for a fuel gas booster application between oil-flooded screw, centrifugal, and reciprocating compressors is shown in Table 5. Table 5. Typical Comparison Table for Fuel Gas Booster.
Hydrogen is the main gas component and H2S is typically included in the gas stream in ppm level. Gas composition is not stable due to change of desulfurization process and nitrogen operation is required at startup. Therefore the compressor needs to have the capability of operating at various conditions of gas composition. Pressure condition is typically very low pressure. However suction pressure is higher when discharge pressure is high, which can change case by case with the process. The end users are oil refineries, so longer times between maintenance periods and high reliability are required. In the past reciprocating compressors and centrifugal compressors were typically used for this application. However, demand for the oil-flooded screw compressors has been increasing. In the 1990s oil-flooded screw compressors suitable for high suction pressure and low pressure ratio were developed. The oilflooded screw compressor can be suitable for gas composition changes due to positive displacement type of compression. The slide valve allows the compressor to handle pressure and flow changes with power savings. Other than the desulfurization compressor, there is another application in the desulfurization process called net gas booster, which requires higher pressure ratio and larger size. Since this net gas booster contains hydrogen the oil-flooded screw compressor has started to be used for this application instead of reciprocating compressors, for longer maintenance periods. A typical package using an oil-flooded screw compressor is shown in Figure 11. A noise enclosure is not typically required for oil-flooded screw
compressors due to low noise, so the accessibility to the compressor is secured, which is also a very important factor from a maintenance standpoint.
CONCLUSION
Oil-free and oil-flooded screw compressors can be applied in many applications. Some reasons for considering the screw compressor are changes in process conditions, recent progress in compressor technologies, and application range of screw compressors. There are many benefits for the customer such as high reliability, low initial cost, less maintenance cost, and power savings.
REFERENCES
API Standard 619, 2004, Rotary Type Positive Displacement Compressors for Petroleum, Petrochemical, and Natural Gas Industries, Fourth Edition, American Petroleum Institute, Washington, D.C.
BIBLIOGRAPHY
Ohama, T., Amano, Y., and Kawaguchi, N., 2000, High Pressure Oil-Flooded EH Series Screw Compressors, Kobelco Technology Review, (23). Figure 11. Typical Package for Desulfurization Service Using Oil-Flooded Screw Compressor. Application Chart To get a better understanding, Figure 12 shows an application chart where the applications in this paper fall with each type of compressor. Although screw compressor applicable range is confined to reciprocating compressor, and centrifugal compressor range, there are applications with ranges where screw compressors are used, as referred to in this paper because of the many advantages in using screw compressors. Ohama, T., Koga, T., and Kurioka, Y., 2004, High Pressure Oil-Injected Screw Gas Compressors (API 619 Design) for Heavy Duty Process Gas Applications, Proceedings of the Thirty-Third Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 49-56.
Figure 12. Application Chart with Each Type of Compressor Typical Applicable Range.
Table of Contents
WHATS CORRECT FOR MY APPLICATION A CENTRIFUGAL OR RECIPROCATING COMPRESSOR?

by Paul Gallick
Senior Applications Engineer Elliott Company Jeannette, Pennsylvania
Greg Phillippi
Director Process Marketing and Sales
and Benjamin F. Williams

Application Engineer Ariel Corporation Mount Vernon, Ohio
Paul B. Gallick is a Senior Applications Engineer for Elliott Company Ebara Group in Jeannette, Pennsylvania. He has worked for Elliott as a Compressor Product Engineer, a Lubrication and Seal Oil Systems Engineer, and currently as a Compressor Application Engineer, for more than 21 years. He is responsible for compressor selection, specification review, and other aspects related to the preparation of quotations for new and rerated compressors. Mr. Gallick has a B.S. degree (Mechanical Engineering, 1978) from the University of Pittsburgh and is a registered Professional Engineer in the Commonwealth of Pennsylvania. Greg Phillippi is the Director of Process Marketing and Sales for Ariel Corporation, in Mount Vernon, Ohio. He began his career as a design engineer with Cooper Energy Services in 1978. In 1985 he accepted a position as a Design Engineer with Ariel Corporation. From 1985 to the end of 1999 he worked as a Design Engineer and Design Engineering Manager at Ariel. In 2000, Mr. Phillippi accepted a position with ACI Services, Inc., in Cambridge, Ohio, where he was deeply involved with marketing, sales and engineering. In January 2004, he moved backed to Ariel in his present role. He has significant experience in the design and application of reciprocating compressors. Mr. Phillippi has a BSME degree (1978) from Ohio Northern University and an MBA degree (2000) from Ashland University. Benjamin F. Williams is an Applications Engineer for Ariel Corporation, in Mount Vernon, Ohio. He primarily focuses on process and international applications for Ariel Corporation, where he has worked since 1997. Prior to working at Ariel, he worked for Lone Star Compressor Corporation, in South Houston, Texas, from 1985 to 1997. During his 20+ years in the compressor industry (the majority in process compression), Mr. Williams has worked in design, application, reapplication, and service of many different compressor makes and models. He is specially trained in
113
Nuclear Propulsion Engineering and served on two nuclear submarines: the USS John Adams and the USS Nathan Hale.
ABSTRACT
This tutorial addresses the question of which compressor type is better suited to a given applicationa centrifugal or reciprocating design. The general application map will be presented and discussed, as will the advantages and disadvantages of each type of compressor. The application guidelines will be addressed from the standpoint of reliability, cost, efficiency, size, and other more general application parameters such as molecular weight, compression ratio, and flow range, etc. The intent of the tutorial will be to provide guidelines and comparative information to be used by contractors and users to determine which type of compressor will be the best fit for their particular application.
INTRODUCTION
The tutorial is organized into four sections. The first, HOW A RECIPROCATING COMPRESSOR WORKS, will be a short, very basic explanation of how a recip compressor works. The next section, HOW A CENTRIFUGAL COMPRESSOR WORKS, will do the same for a centrifugal compressor. The goal with these two sections is to serve as a primer for the rest of the tutorial. The third section, and the primary content of the tutorial, will be where the two different machines are compared and contrasted. Finally, the fourth section is entitled CASE STUDIES. Here 12 different sets of application conditions (four sets of inlet and outlet pressures each at three different gas mole weights) are used to compare the performance of the two machines.
HOW A RECIPROCATING COMPRESSOR WORKS

How a reciprocating compressor works will be explained by discussing pressure versus time (Figure 1) and pressure versus volume (Figure 2) diagrams.
Figure 1. Pressure Versus Time Diagram.
Figure 2. Pressure Versus Volume Diagram. The P-T diagram (Figure 1) is a plot of the pressure of the gas in the compression chamber versus the angle of crankshaft rotation, which is essentially time because crank angle and time are directly related. Understanding the P-T diagram will help in understanding the P-V diagram, which underlies all the theory of reciprocating gas compressor operation. The P-V diagram (Figure 2) is a plot of the pressure of the gas trapped in the compression chamber versus the volume of gas trapped in the compression chamber. The volume of gas trapped in the compression chamber is not linearly related to crank angle or time, so the two diagrams have different shapes and purposes. A reciprocating compressor is a positive displacement machine meaning a certain volume of gas is drawn in to the compression chamber where it is trapped, compressed, and released.
Figure 3. Aerodynamic Parts of a Centrifugal Compressor.
HOW A CENTRIFUGAL COMPRESSOR WORKS

A centrifugal compressor is a dynamic type of compressor where the pressure rise is accomplished by transfer of dynamic (motion-related) energy from the rotor to the gas.
The basic aerodynamic components of a centrifugal compressor

are the impellers, diffusers, and return channels. (Figure 3) in the impeller (Figure 4).
Velocity (kinetic energy) is imparted from moving blade to gas Velocity

(kinetic energy) is converted to pressure (potential energy) in the diffuser. Figure 4. Impeller Discharge Gas Velocity.
For a given pressure ratio, more head is required to compress a Increasing speed imparts higher kinetic energy, which converts
to higher pressure (potential energy) (Figure 6).
low molecular weight gas than for a higher molecular weight gas (Figure 5).
The
head/pressure ratio curve is essentially constant with changing volume flow at fixed speed (Figure 7).
Centrifugal compressor performance can be estimated using the
fan laws. According to the fan laws, capacity is proportional to speed, head is proportional to the square of the speed, and power is proportional to the cube of the speed. Since these laws assume ideal gases with constant k and Z values, they apply with reasonable accuracy to single-stage compressors or multistage compressors with low pressure ratios. For a multi-stage centrifugal compressor with a high pressure ratio, the laws still apply directionally, but the accuracy of performance estimates using the fan laws is insufficient for most calculations.
Figure 5. Gas Mole Weight Effect on Head.
Table of Contents WHATS CORRECT FOR MY APPLICATION A CENTRIFUGAL OR RECIPROCATING COMPRESSOR? 115
is 300F (149C). Air compressors may run at discharge temperatures in excess of 400F (204C). design itself is typically 400 to 450F (204 to 232C). Higher temperatures are possible but require special designs such as center supported diaphragms, less efficient seal materials, and high temperature O-rings and sealants. The process may also have discharge temperature limitations due to fouling, temperature limits of downstream components, and process efficiency. Minimum Suction (Inlet) Temperature
CentrifugalMaximum temperature based on the compressor
Figure 6. Effect of Speed on Head.
ReciprocatingThe common compressor cylinder materials, cast gray iron and cast ductile iron, are acceptable for use at temperatures as low as 40F (40C) which typically occur in refrigeration applications. The lowest suction temperatures requested typically are in liquefied natural gas boil-off gas applications. These inlet temperatures can be as low as 260F (162C). Compressor cylinders used for this application require very special materials and are not offered by all manufacturers. CentrifugalStandard centrifugal compressor materials are typically suitable for 20 to 50F (19 to 46C). Refrigeration compressors in ethylene service typically have inlet temperatures as low as 155F (104C), which require special low temperature compressor materials. Similar to reciprocating compressors, the lowest temperature requirement for centrifugal compressors is typically found in liquefied natural gas (LNG) boil-off gas applications. Centrifugal compressor designs and materials for this service can accommodate minimum temperatures down to 275F (171C). Special low temperature stainless steels are typically used for this service. Special low temperature seals and O-rings are also required.
Maximum Flow
Figure 7. Effect of Speed Range on Overall Flow Range.
CENTRIFUGAL VERSUS RECIPROCATING COMPARISON

The following text is organized by topic with comments for that topic relating to reciprocating and centrifugal compressors. Maximum Discharge (Outlet) Pressure
ReciprocatingThe typical reciprocating compressor is used for discharge pressures up to 12,000 psi (828 bar). Special compressors (called hypercompressors) are used in low density polyethylene production and discharge at pressures up to 50,000 psi (3500 bar).
ReciprocatingReciprocating compressors are positive displacement type compressors. Capacity is limited by cylinder size, the number of throws available, and the available driver speeds. A throw is a location on the crankcase where acompressor cylinder can be attached. CentrifugalCentrifugal compressors can be sized for an inlet flow of 400,000 acfm (680,000 m3/hr) in a single body. The maximum flow through a centrifugal compressor is limited by the choke point, which is the point at which the flow through some part of the compressor nears a velocity of Mach 1.
Minimum Flow
CentrifugalDischarge pressures to 1450 psi (100 bar) for horizontally split compressors. Discharge pressures up to 15,000 psi (1034 bar) for radially split (barrel) compressors.
Minimum Suction (Inlet) Pressure be applied with suction pressures at atmospheric or even a slight vacuum. In vacuum applications, precautions must be taken to prevent atmospheric air from leaking into the cylinder through the piston rod packing.
ReciprocatingSimilar to the maximum flow, the minimum flow

in a reciprocating compressor is limited by cylinder size, stroke, and speed. Very small reciprocating compressors are available.
ReciprocatingCan CentrifugalInlet
CentrifugalCentrifugal compressors can be sized for flow as
pressures to atmospheric or below. For subatmospheric inlet conditions, special seal and buffering designs are employed to keep atmospheric air from being drawn into the compressor.
low as a few hundred acfm. Unlike a reciprocating compressor where minimum flow is solely a function of compressor geometry and speed, the minimum flow for a centrifugal compressor is limited by an aerodynamic condition known as surge, which is a function of compressor geometry, speed, aerodynamic gas conditions, and system resistance.
Maximum Discharge (Outlet) Temperature
Flow Range
ReciprocatingDischarge temperature limits will depend on the application and the seal element materials selected. In hydrogen rich applications, API 618 (1995) limits discharge temperatures to 275F (135C). For natural gas service the maximum discharge temperature limit is 350F (175C); although in most cases a more practical limit
ReciprocatingReciprocating compressors have the ability to change flow (throughput) through speed control, the addition of fixed clearance to a cylinder (fixed or variable volume clearance pockets), cylinder end deactivation, and system recycle. Typical flow range might be 100 percent down to as low as 20 percent, and even lower.
The application will determine what type of capacity control method is required and used. On low compression ratio applications (compression ratio less than 1.6, such as natural gas pipeline gas transmission) adding fixed clearance will have little if any effect on flow. These applications may use speed control or cylinder end deactivation. In other applications, with higher compression ratios, it is quite common to use clearance pockets and cylinder end deactivation to regulate flow.
may be even longer). This depends so much on the specifics of the application and the cleanliness of the gas. A major overhaul (typically defined as completely going through the machine to the point of replacing the bearings) may be required only every 10 years or longer. compressor has to be designed for at least five years of uninterrupted service. In clean service, a centrifugal compressor can operate continuously for 10 years or longer. Maintenance requirements are typically limited to replacing bearing pads and seal wearing parts. Compressed Gas Molecular Weight
CentrifugalPer API 617 (2002), Seventh Edition, a centrifugal
CentrifugalAs discussed above, the flow range of a centrifugal compressor is set by the surge and choke points. Typical turndown for a fixed speed, multistage centrifugal compressor is approximately 20 to 30 percent. With speed variation or adjustable inlet guide vanes, the turndown can be increased to 40 to 50 percent or more (Figure 9).
Weight
ReciprocatingThe
weight of a reciprocating compressor obviously varies with size, which varies with speed, stroke, and rod load rating. It is safe to say that on a mass per power basis the centrifugal compressor will be lighter.
ReciprocatingA reciprocating compressor has no limit with regard to molecular weight. Very light and very heavy gases are compressed equally well. Over the range of molecular weight different application configurations may be required. For example, very low molecular weight gases may present some seal challenges and very high molecular weight gases pose issues with efficiency. But nonetheless, the recip handles the whole range quite well. CentrifugalCompression
ratio is highly dependent on molecular weight. Head is developed by increasing gas velocity to create kinetic energy and then converting the kinetic energy to pressure in the diffuser. The amount of kinetic energy is a function of gas velocity and mass or molecular weight. Centrifugal compressors are used for a broad range of molecular weight including low molecular weight applications such as hydrogen recycle and high molecular weight applications using refrigerant gases with molecular weights over 100.
weight of a centrifugal compressor varies depending on compressor size (refer to Size below) and, to a smaller degree, on materials of construction. The driver, baseplate, and auxiliary systems contribute significantly to the weight of a centrifugal compressor package. Size
CentrifugalThe
ReciprocatingReciprocating compressors come in a wide variety of sizes. The size and weight of a reciprocating compressor is directly related to stroke, speed, and rod load rating. Stroke, in turn, can be related to the driver speed. In general, the higher the driver rotating speed, the shorter the stroke and therefore the smaller the compressor. Conversely, a slower speed compressor will typically have a longer stroke and be physically larger. CentrifugalThe size of a centrifugal compressor is mainly a function of flow capacity (sets the diameter) and number of stages (sets the length). Casing outer diameters range from as small as 20 inches (500 mm) to as large as 150 inches (3800 mm).
Reliability
Compression Ratio
ReciprocatingThe maximum compression ratio that a reciprocating can handle in one stage is limited mostly by compressed gas discharge temperature. The piston rod load generated by the compression ratio may also be a limit. Typical compression ratios are 1.2 to 4.0. CentrifugalCompression ratio is a function of gas molecular weight, compressibility, stage geometry, compressor speed, and the number of compressor stages. For a specific gas, the limits to compression ratio are the mechanical and rotordynamic limitations on speed and the number of stages that can be accommodated in a single body. Discharge temperatures resulting from high compression ratios can usually be controlled by intercooling.
Materials
ReciprocatingReciprocating compressors will probably never be as reliable as centrifugal compressors. The recip has many more parts and more rubbing seals (pressure packing, piston rings, and rider rings) that wear and require more frequent replacement than any seal or other part in a centrifugal. In addition a recip has compressor valves, which though very simple mechanical devices (simple spring-loaded check valves), require considerable maintenance. Another significant factor affecting the reliability of the recip is the cleanliness of the process gas. Wear life of the seals and the valves will be considerably longer with a process gas that is free of liquid and solid debris.
is typically 98 to 99 percent.
ReciprocatingReciprocating
compressors are made of very common materials such as gray iron, ductile iron, carbon steel, alloy steel, and stainless steel, in cast, forging, or bar stock form. Some compressor pistons and covers may be made of aluminum. For corrosive applications it is common to see stainless steel, typically 17-4PH or a 400 series, used for piston rods and compressor valve seats and guards.
CentrifugalReliability/availability of centrifugal compressors

Typical Maintenance Intervals
ReciprocatingMaintenance
intervals for a reciprocating compressor vary significantly with the application and follow along with the comments made in the reliability section. Compressor valve and seal element intervals might be as short as a few months and as long as three to five years (some applications
CentrifugalMaterials for major components such as casings, nozzles, shafts, impellers, etc., are primarily carbon, alloy, and/or stainless steels. Components may be cast, forged, or fabricated. Some cast or nodular iron may be used for stationary components. Material selection is primarily dependent on temperature, stress (pressure, torque), and gas composition (corrosive, erosive, etc).
Multiservice Capability
ReciprocatingIt is very easy to have a multiservice reciprocating
compressor. The number of different services on a given
compressor crankcase (frame) is only limited by the number of throws available and the number of stages required for each service. Eight, 10, and even 12 throw frames are not uncommon.
Lead Time
CentrifugalWhile it is possible to have a multisection centrifugal compressor with different services/gases in each section, this is not typical.
Efficiency
ReciprocatingToday, reciprocating compressor lead time is quite long due to the increased demand from the natural gas industry. Lead time for a bare compressor will vary from 14 to 40 weeks depending on size and manufacturer. Quite often, electric motor driven compressor lead time is driven by the lead time of the motoragain depending on size (horsepower). CentrifugalTypical lead times for a centrifugal compressor train are in the range of 35 to 75 weeks. Lead time is most significantly affected by the original equipment manufacturer (OEM)/subvendor shop loading, availability of any special materials required (low temperature, corrosion resistant, etc.), special/unique design requirements, and testing/inspection requirements.
Installation Time and Complexity
ReciprocatingReciprocating
compressors have a very characteristic adiabatic efficiency curve (Figure 8). As compression ratio drops, adiabatic efficiency drops. Efficiency changes with molecular weight. Efficiency will also vary with several other factors, most significantly the compressor cylinders ratio of valve flow area to main bore diameter and piston speed.
ReciprocatingInstallation time for a reciprocating compressor varies significantly with size and location, and whether or not the compressor is packaged. Packaged compressors are common today up to 5000 hp (3.4 MW) of a high speed short stroke design. Installation time for these might vary from a few days to a couple of weeks. Larger slow speed long stroke compressors assembled at site might require three to four weeks to install.
time varies widely depending on the size of the compressor. The number of main casing nozzle connections and the type of driver also affect installation time. The location can be a factor as well. Remote or offshore locations can add to the installation time. The compressor and driver are typically packaged on a base plate complete with oil piping and wiring to junction boxes. Process equipment such as coolers and scrubbers and process control valves are typically installed at site. Auxiliary systems such as lube oil consoles, control panels, and seal buffer systems may also be installed separately. Piping and wiring from these auxiliary systems and process equipment to the compressor train are typically done at site. Installation time for a typical motor/gear driven compressor package is two to three weeks. For a very large compressor or a gas turbine driven compressor the installation time could be three to six weeks.
CentrifugalSimilar to a reciprocating compressor, the installation
Figure 8. Reciprocating Compressor Efficiency.
CentrifugalPolytropic efficiency is typically used for centrifugal compressors rather than adiabatic. Adiabatic is commonly used for air compressors. Typical polytropic efficiencies range from 70 percent to 85 percent. Efficiencies approaching 90 percent are possible. In a centrifugal compressor, efficiency is primarily affected by the internal leakage and mechanical losses.
Cost: Capital and Operating
ReciprocatingGenerally a reciprocating compressor will have a lower capital cost but a higher operating cost compared to a centrifugal. The lower capital cost is a direct result of the machine consisting of smaller parts that cost less and are easier to manufacture. Higher operating cost results from the recip containing more wearing parts requiring more maintenance and downtimemost specifically compressor valves, which do not exist in a centrifugal. CentrifugalThe capital cost of a centrifugal compressor is typically higher than a reciprocating compressor operating at the same conditions. This is primarily due to the fact that centrifugal compressors require parts with more complex geometry such as impellers and diaphragms. However, a centrifugal compressor has fewer wearing parts, resulting in lower operating costs in terms of replacement parts, repairs, and downtime.
Minimum/Maximum Power
CASE STUDIES
APPENDIX A contains tables where the performance data for reciprocating and centrifugal compressors are compared for four different sets of pressure conditions, each for three different mole weight gases, for a total of 12 case study points. The purpose of these case studies is to compare performance and to help explain why a certain design compressor fits a certain application better than the otherfrom a performance perspective. Figure 9 is a chart showing discharge pressure versus inlet flow intending to compare and contrast where each type of compressor fits best in this map. The case study points have been selected from this map.
ReciprocatingReciprocating compressors vary in size from the very small, under 10 hp (7.5 kW), to very large at 12,000 hp (9.0 MW). Even higher horsepower compressors are available from some manufacturers for some very specialized applications, like very high pressure ethylene compression (hyper compressors).
CentrifugalPower developed is dependent on the mass flow of the gas compressed, the head required, and the efficiency. The power required to drive a centrifugal compressor can be as low as 100 hp (75 kW) and as high as 130,000 hp (97 MW) or more.
Figure 9. NGPSA Compressor Coverage Chart. (Courtesy, NGPSA Engineering Data Book, 1994)
APPENDIX A
Table A-1. Case Study Performance Data, A.
Case Study Number 1A Low Flow, Low Pressure, Medium Molecular Weight Capacity = 600 acfm (1019.4 m3/hr) Suction Pressure = 5 psig (0.34 BarG) Suction Temperature = 80o F (27o C) Discharge Pressure = 45 psig (3.1 BarG) Molecular Weight = 18.83 (S.G. = 0.65) Power, hp (kW) Flow, scfm (Nm3/hr) Power per flow, hp/scfm (kW/Nm /hr) Number of Stages Equivalent Head, ft (m)
3
Reciprocating 79 (59) 774 (1,244) 0.102 (0.047) 1 54,869 (16724)
Centrifugal 150 (112) 774 (1,244) 0.193 (0.090) 7 59,230 (18,053)
Case Study Number 1B Low Flow, Low Pressure, Low Molecular Weight Capacity = 600 acfm (1019.4 m3/hr) Suction Pressure = 5 psig (0.34 BarG) Suction Temperature = 80o F (27o C) Discharge Pressure = 45 psig (3.1 BarG) Molecular Weight = 2.02 (S.G. = 0.07) Power, hp (kW) Flow, scfm (Nm3/hr) Power per flow, hp/scfm (kW/Nm /hr) Number of Stages Equivalent Head, ft (m)
3
Reciprocating 78 (58) 773 (1243) 0.101 (0.047) 2 540,746 (164,819)
Centrifugal This case is too small for a centrifugal compressor.
Case Study Number 1C Low Flow, Low Pressure, High Molecular Weight Capacity = 600 acfm (1019.4 m3/hr) Suction Pressure = 5 psig (0.34 BarG) Suction Temperature = 80o F (27o C) Discharge Pressure = 45 psig (3.1 BarG) Molecular Weight = 28.05 Power, hp (kW) Flow, scfm (Nm3/hr) Power per flow, hp/scfm (kW/Nm /hr) Number of Stages Equivalent Head, ft (m)
3
Reciprocating 85 (63) 774 (1,245) 0.110 (0.050) 2 36,702 (11,187)
Centrifugal 151 (113) 774 (1,245) 0.192 (0.089) 5 38,578 (11,759)
Case Study Number 2A, High Flow, Medium Pressure, Medium Molecular Weight Capacity = 10,000 acfm (16,990.1 m3/hr) Suction Pressure = 200 psig (13.8 BarG) Suction Temperature = 80o F (27o C) Discharge Pressure = 1000 psig (68.95 BarG) Molecular Weight = 18.83
Reciprocating
Centrifugal
Table A-2. Case Study Performance Data, B.
Power, hp (kW) Flow, scfm (Nm3/hr) Power per flow, hp/scfm (kW/Nm /hr) Number of Stages Equivalent Head, ft (m)
3
18,909 (14,100) 145,830 (234,522) 0.130 (0.060) 2 78,822 (24,025)
21850 (16.306) 145,830 (234,522) 0.150 (0.070) 6 81,400 (24,811)
Case Study Number 2B, High Flow, Medium Pressure, Low Molecular Weight Capacity = 10,000 acfm (16,990.1 m3/hr) Suction Pressure = 200 psig (13.8 BarG) Suction Temperature = 80o F (27o C) Discharge Pressure = 1000 psig (68.95 BarG) Molecular Weight = 2.02 Power, hp (kW) Flow, scfm (Nm3/hr) Power per flow, hp/scfm (kW/Nm3/hr) Number of Stages Equivalent Head, ft (m)
Reciprocating 14,295 (10,660) 145,830 (225,430) 0.098 (0.047) 2 813,384
Centrifugal This case can not be accomplished with a reasonable number of centrifugal compressor bodies/stages.
Case Study Number 2C, High Flow, Medium Pressure, High Molecular Weight Capacity = 10,000 acfm (16,990.1 m3/hr) Suction Pressure = 200 psig (13.8 BarG) Suction Temperature = 80o F (27o C) Discharge Pressure = 1000 psig (68.95 BarG) Molecular Weight = 28.05 Power, hp (kW) Flow, scfm (Nm /hr) Power per flow, hp/scfm (kW/Nm3/hr) Number of Stages Equivalent Head, ft (m)
3
Reciprocating 21,102 (10,660) 154,014 (225,430) 0.137 (0.047) 2 53,211 (16,219)
Centrifugal 20,960 (15,642) 154,014 (225,430) 0.136 (0.063) 4 49,700 (15,149)
Case Study Number 3A, Medium Flow, Medium Pressure, Medium Molecular Weight Capacity = 3,000 acfm (5,097 Nm3/hr) Suction Pressure = 200 psig (13.79 BarG) Suction Temperature = 80o F (27o C) Discharge Pressure = 600 psig (41.37 BarG) Molecular Weight = 18.83 Power, hp (kW) Flow, scfm (Nm /hr) HP per flow, hp/scfm (kW/Nm3/hr) Number of Stages Equivalent Head, ft (m)
3
Reciprocating 3,945 (2,948) 43,867 (70,537) 0.090 (0.042) 2 50,545 (15,406)
Centrifugal 4,252 (3,173) 43,867 (70,537) 0.097 (0.045) 6 51,700 (15,758)
Table A-3 .Case Study Performance Data, C.
Case Study Number 3B, Medium Flow, Medium Pressure, Low Molecular Weight Capacity = 3000 acfm (5,097 Nm3/hr) Suction Pressure = 200 psig (13.8 BarG) Suction Temperature = 80o F (27o C) Discharge Pressure = 1000 psig (69 BarG) Molecular Weight = 2.02 Power, hp (kW) Flow, scfm (Nm3/hr) Power per flow, hp/scfm (kW/Nm /hr) Number of Stages Equivalent Head, ft (m)
3
Reciprocating 3,775 (2,815) 41,861 (67,610) 0.090 (0.041) 2 511,799 (155,996)
Centrifugal This case can not be accomplished with a reasonable number of centrifugal compressor bodies/stages.
Case Study Number 3C, Medium Flow, Medium Pressure, High Molecular Weight Capacity = 3000 acfm (5,097 Nm3/hr) Suction Pressure 200 psig (13.8 BarG) Suction Temperature = 80o F (27o C) Discharge Pressure = 1000 psig (69 BarG) Molecular Weight = 28.05 Power, hp (kW) Flow, scfm (Nm3/hr) HP per flow, hp/scfm (kW/Nm /hr) Number of Stages Equivalent Head, ft (m)
3
Reciprocating 3,927 (2,928) 45,477 (73,125) 0.086 (0.040) 1 33,937 (10,344)
Centrifugal 4,062 (3,031) 45,477 (73,125) 0.089 (0.041) 4 31,161 (9,498)
Case Study Number 4A, Medium Flow, High Pressure, Medium Molecular Weight Capacity = 2000 acfm (3,398 Nm3/hr) Suction Pressure = 200 psig (13.8 BarG) Suction Temperature = 80o F (27o C) Discharge Pressure = 3500 psig (241.4 BarG) Molecular Weight = 18.83 Power, hp (kW) Flow, scfm (Nm3/hr) Power per flow, hp/scfm (kW/Nm /hr) Number of Stages Equivalent Head, ft (m)
3
Reciprocating 7,012 (5,229) 29,027 (46,675) 0.241 (0.180) 3 163,227 (49,752)
Centrifugal 9,732 (7,263) 29,027 (46,675) 0.335 (0.156) 16 (2 bodies) 147,763 (45,038)
Case Study Number 4B, Medium Flow, High Pressure, Low Molecular Weight Capacity = 2000 acfm (3,398 Nm3/hr) Suction Pressure = 200 psig (13.8 BarG) Suction Temperature = 80o F (27o C) Discharge Pressure = 3500 psig (241.4 BarG) Molecular Weight = 2.02
Reciprocating
Centrifugal
Table A-4. Case Study Performance Data, D. Power, hp (kW) Flow, scfm (Nm /hr) Power per flow, hp/scfm (kW/Nm3/hr) Number of Stages Equivalent Head, ft (m)
3
6,844 (5,134) 27,908 (44,880) 0.245 (0.114) 4 1,788,446
This case can not be accomplished with a reasonable number of centrifugal compressor bodies/stages.
Case Study Number 4C, Medium Flow, High Pressure, High Molecular Weight Capacity = 2000 acfm (3,398 Nm3/hr) Suction Pressure = 200 psig (13.8 BarG) Suction Temperature = 80o F (27o C) Discharge Pressure = 3500 psig (241.4 BarG) Molecular Weight = 28.05 Power, hp (kW) Flow, scfm (Nm /hr) Power per flow, hp/scfm (kW/Nm3/hr) Number of Stages Equivalent Head, ft (m)
3
Reciprocating 5,890 (4,392) 30,318 (48,750) 0.194 (0.090) 3 105,164 (32,054)
Centrifugal 7,811 (5,829) 30,318 (48,750) 0.258 (0.120) 10 76,779 (23,408)
REFERENCES
API Standard 617, 2002, Axial and Centrifugal Compressors and Expander-Compressors for Petroleum, Chemical and Gas Industry Services, Seventh Edition, American Petroleum Institute, Washington, D.C. API Standard 618, 1995, Reciprocating Compressors for Petroleum, Chemical, and Gas Industry Services, Fourth Edition, American Petroleum Institute, Washington, D.C. NGPSA Engineering Data Book, 1994, 1, Revised 10th Edition, Compiled and Edited in Cooperation with the Gas Processors Association, Copyright 1987 Gas Processors Association.
Brown, R. N. ,1986, CompressorsSelection and Sizing, Houston, Texas: Gulf Publishing Company. Paluselli, D. A, Basic Aerodynamics of Centrifugal Compressor, Elliott Company, Jeannette, Pennsylvania.
ACKNOWLEDGEMENT
The authors wish to acknowledge Mr. John F. Vanderhoff, formerly senior application engineer with Elliott Company and presently with Westinghouse in Monroeville, Pennsylvania, who contributed a significant amount of effort to this tutorial while employed at Elliott Company.
BIBLIOGRAPHY
Basic Thermodynamics of Reciprocating Compression, Short Course Presented at the 2005 Gas Machinery Conference, October 2005.
Table of Contents
SURGE AVOIDANCE FOR COMPRESSOR SYSTEMS

by Robert C. White
Principal Engineer
and Rainer Kurz

Manager, Systems Analysis Solar Turbines Inc. San Diego, California
INTRODUCTION
Robert C. White is a Principal Engineer for Solar Turbines Inc., in San Diego, California. He is responsible for compressor and gas turbine performance predictions and application studies. In his former position he lead the development of advanced surge avoidance and compressor controls at Solar Turbines. Mr. White holds 12 U.S. Patents for turbomachinery related developments. He has contributed to several papers, tutorials, and publications in the field of Turbomachinery. Rainer Kurz is Manager of Systems Analysis and Field Testing for Solar Turbines Inc., in San Diego, California. His organization is responsible for conducting application studies, gas compressor and gas turbine performance predictions, and site performance testing. He joined Solar Turbines in 1993, and has authored more than 30 publications in the field of turbomachinery and fluid dynamics. Dr. Kurz attended the University of the German Armed Forces, in Hamburg, Germany, where he received the degree of a Dipl.-Ing., and, in 1991, the degree of a Dr.-Ing. He was elected as an ASME Fellow in 2003. Operation of centrifugal gas compressors can be defined by three operating parameters: speed, head, and flow. Centrifugal compressors have a maximum head that can be achieved at a given speed. At that peak head there is a corresponding flow. This is a stability limit. Operation of the compressor is stable provided the head is lower (less resistance in series with the compressor) and the flow is greater than these values. That is, the system is stable, as long as reductions in head result in increases in flow. Surge occurs when the peak head capability of a compressor is reached and flow is further reduced. Depending on the dynamic behavior of the compression system, system surge can occur at somewhat higher or, seldom, lower flows than the peak head capability. This is a particular issue in systems with low frequency pulsations (Kurz, et al., 2006). When the compressor can no longer meet the head imposed by the suction and discharge condition (which are imposed by the compression system), flow reverses. When a compressor approaches its surge limit, some of its components (diffusers, impeller) may start to operate in stall. Stall occurs when the gas flow starts to separate from a flow surface (Figure 1). Changing the operating point of a compressor always involves a change in incidence angles for the aerodynamic components. Just as with an airfoil (Figure 1), increasing the incidence angle will eventually lead to stall. Stall in turbomachines often appears as rotating stall, when localized regions of separated flow move along the diffuser at speeds below the rotational speed of the impeller (Day, 1991). Surge is the ultimate result of system instability.
ABSTRACT
This tutorial explains the surge phenomenon as well as the precursors of surge and the damage that surge can cause to the equipment. Then, methods to keep the compressor system from surge, i.e., modern surge control systems, are discussed. This includes the necessary instruments, algorithms used, as well as the piping layout required. Sizing considerations for valves and other system components, in particular methods to correctly estimate allowable upstream pipe volumes, are described. Surge control systems for single body and multibody or multisection compressors will be explained, and attention will be given to the integration of the surge control system with other aspects of the station and unit control. In a look forward, new methods for surge detection and detection of system changes prior to surge are covered.
123
Figure 1. Progression of Stall. (NACA4412 airfoil at increasing angle of attack, based on data by Nakayama, 1988)
Attempting to increase the resistance in series with the compressor beyond that which drives the compressor to peak head will result in unstable operation. Flow will decrease and subsequently the head capability of the compressor will also decrease. As the head capability decreases, flow further decreases. Once the compressor can no longer meet the external head, flow reverses. Surge is what happens after the stability limit of the compression system is passed. Not only is this detrimental to meeting the process objectives, the resulting axial and radial movement of the rotor can cause damage, sometimes severe, to the compressor. Surge can be avoided by ensuring the flow through the compressor is not reduced below the flow at peak head. The surge avoidance system prevents surge by modulating a surge control (bypass) valve around the compressor. A typical system consists of pressure and temperature transmitters on the compressor suction and discharge lines, a flow differential pressure transmitter across the compressor flowmeter, an algorithm in the control system, and a surge control valve with corresponding accessories. A surge avoidance system determines the compressor operating point using the pressure, temperature, and flow data provided by the instrumentation. The system compares the compressor operating point to the compressors surge limit. The difference between the operating point and the surge limit is the control error. A control algorithm (PID) acts upon this difference, or error, to develop a control signal to the recycle valve. When opened, a portion of the gas from the discharge side of the compressor is routed back to the suction side, and head across the compressor is prevented from increasing further. When the operating point reflects more flow than the required protection margin flow, the surge control valve moves toward the closed position and the compressor resumes normal operation. There are five essentials for successful surge avoidance: 1. A precise surge limit modelIt must predict the surge limit over the applicable range of gas conditions and characteristics. 2. An appropriate control algorithmIt must ensure surge avoidance without unnecessarily upsetting the process. 3. The right instrumentationInstruments must be selected to meet the requirements for speed, range, and accuracy. 4. Recycle valve correctly selected for the compressorValves must fit the compressor. They must be capable of large and rapid, as well small and slow, changes in capacity. 5. Recycle valve correctly selected for the system volumesThe valve must be fast enough and large enough to ensure the surge limit is not reached during a shutdown. The piping system is the dominant factor in the overall system response. It must be analyzed and understood. Large volumes will preclude the implementation of a single valve surge avoidance system.
reach a final working pressure of 7 barg, a turbocompressor needed as many as 10 or 11 stages. A single modern impeller can produce a pressure ratio as high as 8:1. As the performance of compressors increased, so did the potential for damage from surge. Not only the damage to compressors needed to be avoided but avoiding upsets to the process became essential. To address these ever growing needs, surge avoidance systems evolved. Early surge controls were pneumatic. They monitored the pressure differential (DP) across the compressor versus the DP across a flow measuring element through a balance beam that controlled the pneumatic signal to a recycle valve. With higher performance, higher stressed compressors, and more challenging applications, better surge controls were required. Initially electronic surge controls were models of the pneumatic ones. They were faster, less complicated, more reliable, and required less maintenance than their pneumatic predecessors. With the advent of the microprocessor, surge avoidance systems became more algorithmically intense, surge limits were modeled via polynomials, and asymmetrical control schemes came into use. Modern surge controls ensure surge avoidance while maximizing the operating range of the compressor.
THE SURGE LIMIT MODEL

In order to avoid surge it must be known where the compressor will surge. The more accurately this is predicted, the more of the compressors operating range that is available to the user (Figure 2). A compressors operation is defined by three parameters: head, flow, and speed. The relationship between the compressors operating point and surge can be defined by any two of the three.
BACKGROUND
The first turbocompressors were manufactured at the turn of the 1900s. They were originally developed by steam turbine manufacturers and were widely used for ventilation purposes in deep shaft mining, in particular the coal industry. At that time, the method of producing an impeller relied upon fabrication. It would be decades before technology would allow highly efficient turbocompressors to be made. It was not until late in World War II that sufficient money was pumped into technology to allow the production of efficient high-speed compressors. In 1947 to 1948, Ingersoll-Rand and Clark designed the first centrifugal compressors for natural gas transmission. In September 1952, El Paso Natural Gas became the first company to use large gas-turbine-driven centrifugal compressors for natural gas transmission. The pressure ratio across an early riveted impeller would have only been on the order of 1.2:1. This would have meant that to Figure 2. Surge Avoidance System Schematic. The first two models on the left of Figure 3 involve speed. The speed of the compressor at an operating condition is strongly influenced by changes in gas composition, because the machine Mach number will change. The head versus flow relationship on the right provides a means for modeling the surge limit without being affected by gas conditions or characteristics. The parameters of the surge limit model on the right can be measured in terms of head across the compressor and head across the flowmeter.
Hp = K
PD PS 1
T SG Z
Table of Contents SURGE AVOIDANCE FOR COMPRESSOR SYSTEMS 125
Surge Margin
Head Rise to Surge
Turndown
Q
H = Head
N2
Q = Flow N = Speed
Q2
F Figure
3. Surge Limit in Different Systems of Reference.
This is the basic equation for head. In the head across the compressor and the head across the flowmeter monitoring the flow through the compressor, there are common terms. These common terms (units, gas temperature, specific gravity, and compressibility) are equal in both equations and can be cancelled. This results in a simplified model that is referred to as reduced head versus reduced flow.
What does a surge avoidance system do most of the time? Hopefully nothing! Then, with very little margin it must act aggressively, probably requiring gains higher than could be maintained stable, to protect the compressor. To avoid instability the gains are reduced to close the valve. Once surge has been avoided, the control system should bring the process back online slowly and smoothly to avoid further upsets. The need for extremely high gains is driven by the following: surge avoidance systems are normally built up out of commonly available process control plant components. As such these components are designed for ruggedness, reliability, and low maintenance. In general they are not focused on speed of data acquisition. Information about changing process conditions is often 1/10 of a second old. As will be seen in later sections significant advances in surge control valve action have been made recently. However, the response of the valve is typically the dominant lag in the system.
INSTRUMENTATION
To avoid surge, the control needs to know where the compressor is operating in relation to surge in real time. Again, how close the protection margin can be placed to surge depends on how accurately and how quickly the change in flow is reported to the control. Correctly selected instrumentation is essential. The system must have accurate measurements of the suction and discharge pressures and temperatures, and the rate of flow. Flow is the most important parameter as it will move the fastest and farthest as the surge limit is approached. Ideally, the flow transmitter should be an order of magnitude faster than the process. Unfortunately, compared to pressure and temperature transmitters, flow transmitters tend to be slow. Even the best surge avoidance control will allow a compressor to surge, if it is connected to a slow transmitter. Flow-Measuring Devices Most commercially available flow-measuring devices are accurate enough for surge avoidance; however, it is the transmitter that slows things down. A differential pressure transmitters response time is inversely proportional to its range; thus, the stronger the signal, the faster the response. Devices that develop high DP signals are desirable. Those with low signal levels tend to have low signal-to-noise ratios. Transmitters for low DP signal ranges typically have slow response times. Devices that create an abrupt restriction or expansion to the gas, such as orifices, cause turbulence and, subsequently, create noise. It is preferable to place the flow-measuring device on the suction side of the compressor. Typically, variations in pressures, temperatures, and turbulence of the gas are less upstream of the compressor. Also, the device must be inside the innermost recycle loop (refer to Figure 2). At a minimum, failure of the device will cause the compressor set to be shut down until the device can be replaced. If the failure results in pieces being ingested by the compressor, it can cause an expensive overhaul. For this reason, devices that are cantilevered into the gas stream are not recommended. Low cost flow-measuring devices do not always result in cost savings in the long run. Low permanent pressure loss (PPL) devices are often recommended; however, their benefits may be marginal. The lost power cost impact of operating a device can be calculated. For example, a flowmeter developing a 100 inch H2O signal and a 50 percent PPL flowing 100 mmscfd (50 lb/sec) is equivalent to about 20 hp. As noted, strong signal devices are highly preferred. Pitot types (annubars and verabars) have a relatively low signal level, around 25 inches H2O. In the middle are orifices and venturis with a moderate signal of around 100 inches H2O. Compressor suctionto-eye provides a strong signal (around 700 inches H2O) with the added benefit of not causing any additional pressure loss.
H REDUCED =
PD PS 1
THE CONTROL ALGORITHM

A surge avoidance control needs to be able to react appropriately to changes in power or the process. There are two very different situations that the system must respond to. If the operating point slowly crosses the protection line, that is, at the same rate it has been moving left for the past several hours, opening of the recycle valve should be small and slow. The interdiction of the surge avoidance control should be unnoticeable. It should be as though the compressor had infinite turndown. Conversely, if the operating point races across the compressor map, the recycle valve should begin opening before the operating point crosses the protection line. Reaction of the control should be aggressive to protect the compressor. In this case we are less concerned about the process, as it has already been impacted. A sudden change in the system produces a control response. This is a standard control test. Ping it and see how it rings. Figure 4 reflects reactions of variously tuned controls. Low gains produce a slow response. A critically damped control produces an aggressive response but settles down quickly. If the gains are too high the system will oscillate.
Error Signal
Gain Too Low Too Slow
Critical Damping Optimum
Gain Too High Too Fast - Unstable
High and Low Gains Smooth Response
Figure 4. Reaction of a Control System to an Error Signal.
Suction-to-Eye Flow Measuring Suction-to-eye uses the inlet shroud or inlet volute of the compressor as a flow-measuring device. This feature is now available on many compressors. The design requirements for the inlet volute and the flow measuring device have several things in common. Performance of the first stage impeller and the device is dependent on the uniform direction and velocity of the flow presented to it. Critical to the operation of suction-to-eye flow measurement is the placement of the eye port. As the impeller approaches surge an area of recirculation begins to develop at the outer perimeter of the inlet to the impeller. If the eye port is placed too close to the impellers outer perimeter the relationship of the DP to flow will be affected. Fortunately the meter factor (C) typically remains nearly the same for the same surge margin. Hence, selecting the meter factor at the desired surge protection margin will contribute to effective surge avoidance. In a typical pipeline application (600 psi suction pressure) suctionto-eye will develop 25 psid (692 inches H2O). This is nearly seven times the differential of an orifice plate. Typically the signal to noise ratio is low and there is no additional permanent pressure loss. For surge avoidance the suction-to-eye method is strongly recommended. Compressor Instrumentation Optimal performance of any control system is dependent on the speed, accuracy, and resolution of the instrumented process conditions. To achieve optimal performance the instruments should have performance specifications an order of magnitude better than the requirements for the system. Typical gas compressor systems have a first-time constant of about one second; hence, no instrument should have a first-time constant of greater than 100 ms. The surge control system is expected to discriminate between single digit percentages of surge margin; hence, measurement of the process parameters should be accurate to 0.1 percent. The final control elements (recycle valves) probably can resolve 1 percent changes in their command signals; hence, the process variables should be resolved to at least 0.1 percent (10 bits) of their normal operating range. Over-ranging transmitters degrade resolution.
Equal percentage valves, and in particular noise-attenuating ball valves, are recommended for surge avoidance systems with a single surge control valve. They perform like smaller valves when nearly closed and bigger valves when close to fully open. Figure 6 is a comparison of two types of equal percentage valve. For a given valve size, the noise-attenuating ball valve is often twice the cost of the globe valve, but it provides approximately three times the flow coefficient (Cv) or capacity. Also, it is more reliable as it is less susceptible to fouling and improper maintenance.
Figure 6. Ball and Globe Valves Compared. Employing a valve with an equal percentage characteristic may provide the capacity needed to avoid surge during a shutdown while maintaining enough resolution at less than 50 percent capacity to provide good control at partial recycle. With an equal percentage characteristic the valve typically has greater resolution than a single linear valve selected to fit the compressor. Multiple Valves If the volumes on either side of the compressor are large, a multiple valve approach may be needed. If an integrated approach is used, the total valve capacity will be reduced. Probably the most common is the hot and cold recycle configuration (Figure 7). Usually the cooled (outer) valve is modulating and the hot (inner) valve is a quick opening on-off type. Generally the two valves are sized independently. If the cooled valve has a solenoid, its capacity can be considered with that of the shutdown valve; subsequently the shutdown valve can be smaller.
THE SURGE CONTROL VALVE

Characteristics Earlier it was discussed how the control should react differently to gradual and rapid approaches to surge. Likewise, the valve must address these two very different requirements. For the gradual approach, it should behave like a small valve and produce smooth throttling. For the rapid approach case, it should act like a large fast valve to handle sudden major changes. There are three general valve characteristics (Figure 5): quick opening, where most of the valves capacity is reached early in its travel; linear, where capacity is equal to travel; and equal percentage, where most of the capacity is made available toward the end of the valves travel. All three types of valve have been used in various configurations as recycle valves.
Figure 7. Hot and Cold Recycle Valve Arrangement. An alternate to this configuration is having a second cooled valve in parallel with the first (Figure 8). This arrangement provides some measure of redundancy. In the control the two valves are operated in cascade. That is, they have different set points, say 9 percent and 10 percent surge margin. Under normal movements of the operating conditions only the 10 percent surge margin valve (primary valve) will open. If movement is fast enough to push the operating point down to 9 percent, the second valve (secondary valve) will open. If the primary valve becomes fouled and no longer positions properly, the control can place it in
Figure 5. Valve Types.
the secondary position and the secondary becomes the primary valve. This change can be made without taking the compressor offline.
Figure 8. Parallel Recycle Valves. The advantages of the two parallel valves do not come without a price. In normal operation 2 percent to 5 percent of the pressure rise across the compressor will be lost across the cooler. In the shutdown scenario the required flow through the cooler to avoid surge may be two or three times the normal flow. This will result in four to nine times the pressure drop across the cooler. This additional pressure drop may increase the needed recycle valve capacity significantly. Recycle valves need to be fast and accurately positionable. They also need to be properly sized for both the compressor and the piping system. A valve well suited for modulating recycle around the compressor may not be suitable for a shutdown. (Refer to the REVIEW OF SYSTEM VOLUMES section below.) For some two-valve applications, single purpose valves may be suitable, one for controlled recycling and one for shutdown. A linear characteristic valve is appropriate for the controlled recycling and a quick opening characteristic globe or ball valve for shutdown. For the applications where the compressor speed lines are fairly flat (little increase in head for a decrease in flow) from the design conditions to surge, extra fast depressurization may be required. To achieve this, two quick opening valves may be employed. In this case a single 6 inch linear characteristic valve is replaced by two 4 inch quick opening valves. The two 4 inch valves should have slightly less flow capacity (Cv) but they will open nearly 45 milliseconds faster. For linear valves 50 percent travel equals 50 percent capacity. For quick opening valves, capacity approximately equals the square root of travel. As such the two 4 inch valves will have 70.7 percent of their fully open capacity at 50 percent open. Comparing the two arrangements 250 ms after the shutdown is initiated, the two 4 inch quick opening valves will have 56 percent more flow capacity than the single 6 inch linear valve. For throttling, the valves are operated in cascade or split range. For most controlled recycling only one valve is opened. Although the valves have a quick opening characteristic the valves are smaller thus the capacity per percent travel is less. The two quick opening valves operated in cascade or split range will have the same Cv as the 6 inch linear at 25 percent travel. Valve Actuation As previously discussed, there are two operational scenarios for the surge avoidance system: modulating (minimum flow control) and rapid depressurization for shutdown. By inserting a three-way solenoid valve into the positioners output, the valve can be made to open with either a proportional (4-20 mA) signal for modulating control, or a discrete (24 VDC) signal for total fast opening. The primary difference between a surge control valve and a standard control valve is in its actuation system. The preferred actuator for surge avoidance is spring return, fail open. This design is simple, reliable, and ensures the compressor is protected in the event of a power failure. Both spring and
diaphragm and spring and piston actuators are used. The spring and diaphragm actuator is most commonly used on globe valves. The spring and piston actuator is more often used on ball valves. The more powerful spring and piston actuators are required on rotary valves due to the greater forces required to accelerate the mass of the ball. Some ball valves are not suitable for surge control applications because their shafts and attachments to the ball are not strong enough to transmit the torque required to open these valves at the required speeds. Surge control valves need to be able to open very quickly. As such their actuators will have strong springs, very large air passages, and shock absorbers at their end of travel. This must be considered when sourcing recycle valves for surge avoidance. The accessory unique to a sound surge control valve assembly is the single sided booster or exhaust booster. This is essentially a differential pressure relief device. Opening the booster vents the actuator pressure to atmosphere. The threshold for opening is about 0.5 psid. There is a small restriction (needle valve) between the control pressure from the positioner via the three-way solenoid valve and the top of the booster. Small slow reductions in pressure (opening the valve) do not cause the booster to open. Large fast reductions in pressure developing more than 0.5 psid across the restriction, cause the booster to open. If the solenoid valve is de-energized, the top of the booster is vented to atmosphere and the booster fully opens. Standard industry quick-exhausts are not recommended for this application. They have a high threshold for opening (typically 2 to 4 psid) and an equally high threshold for reclosing. Although they may work well for fully opening the valve, they will not work well with the positioner. Positioners should be selected for high capacity and quick response to changes in their control signals. Most of the major valve manufacturers have released second and third generation smart positioners that are suitably fast for this application. Figures 9 and 10 show globe and ball valves with their preferred instrumentation configurations.
Needle Valve & Check Valve Exhaust Booster
24VDC Three-way 24VDC Solenoid Valve
Position Transmitter 4 - 20 mA 4 - 20mA Limit Switch Closed Limit Switch Open 4 - 20mA Instrument Air Supply Pressure Regulator (Airset 35 - 50 psig)
Electro- pneumatic Positioner Yoke Mounted 4 - 20 mA 6 - 30 psig
F 9. Globe Valve Assembly. Figure
Position Transmitter 4 - 20 mA 4 - 20mA Closed Limit Switch Open Limit Switch
Electro-pneumatic Positioner Yoke Mounted 4 - 20 mA 6 - 30 psig Three-way 24VDC Solenoid Valve
24 VDC
4 - 20 mA 80 - 100 psig
Exhaust Booster
Needle Valve & Check Valve
Figure 12. Valve Matched to Compressor.
REVIEW OF SYSTEM VOLUMES

Design of the piping and valves, together with the selection and the placement of instruments, will significantly affect the performance of an antisurge control system. This should be addressed during the planning stage of a project because the correction of design flaws can be very costly once the equipment is installed and in operation. As described above, the control system monitors the compressor operating parameters, compares them to the surge limit, and opens the recycle valve as necessary to maintain the flow through the compressor at a desired margin from surge. In the event of an emergency shutdown or ESD, where the fuel to the gas turbine is shut off instantly, the surge valve opens immediately, essentially at the same time the fuel valve is closing. The worst case scenario for a surge control system is an emergency shutdown (ESD), particularly if the compressor is already operating close to surge when the engine shutdown occurs. If an ESD is initiated, the fuel supply is shut off immediately and the compressor will decelerate rapidly under the influence of the fluid forces counteracted by the inertia of the rotor system. Figure 13, which displays data based on test data and theoretical considerations, indicates a 30 percent drop on compressor speed within the first second after shutdown. A 30 percent loss in speed equates to approximately a loss in head of about 50 percent. The valve must, therefore, reduce the head across the compressor by about half in the same time as the compressor loses 30 percent of its speed.
Figure 10. Ball Valve Assembly. Recycle Valve Sizing Tool The recycle valve needs to be sized based on the expected operating conditions of the compressor. A valve-sizing program can facilitate matching a recycle valve to a compressor. The compressor data are entered into the tool in its normal form (pressures, temperatures, heads, speeds, and flows). Various operating conditions for a specific application are then entered, such as the minimum and maximum operating speeds, pipe operating pressures, temperatures, relief valve settings, and cooler data, if applicable. The tool calculates the equivalent valve capacities or Cvs from that data. Typically the surge limit of a compressor equates to a single valve capacity or Cv (Figure 11). The valve can be selected based on valve Cg, Cv, and Xt tables from surge control valve suppliers. As previously described, a single surge control valve application will have an equal percentage characteristic. Once a valve is selected several performance lines of a specific opening can be developed and overlaid on the compressor map. The equal percentage characteristic valve should be at about two-thirds travel at the surge conditions. The valve evaluation in Figure 12 shows such a valve with its flow characteristic when 60 percent, 70 percent, and 100 percent open, superimposed on the compressor map.
Figure 13. Simplified System and Transient Characteristic. The larger the volumes are in the system, the longer it will take to equalize the pressures. Obviously, the larger the valve, the better its potential to avoid surge. However, the larger the valve, the poorer its controllability at partial recycle. The faster the valve can be opened, the more flow can pass through it. There are, however, limits to the valve opening speed, dictated by the need to control intermediate positions of the valve, as well as by practical limits to the power of the actuator. The situation may be improved by using a valve that is only boosted to open, thus combining high opening speed for surge avoidance with the capability to avoid oscillations by slow closing.
Figure 11. Almost Constant Cv at the Surge Limit.

0.5
If the discharge volume is too large and the recycle valve cannot be designed to avoid surge, a short recycle loop (hot recycle valve) may be considered (Figure 7), where the recycle loop does not include the aftercooler. While the behavior of the piping system can be predicted quite accurately, the question about the rate of deceleration for the compressor remains. It is possible to calculate the power consumption for a number of potential steady-state operating points. The operating points are imposed by the pressure in the discharge volume, which dictates the head of the compressor. For a given speed, this determines the flow that the compressor feeds into the discharge. In a simple system, the boundaries for the gas volume on the discharge side are established by the discharge check valve, compressor, and recycle valve (Figure 2). The volume on the suction side is usually orders of magnitude larger than the discharge volume and, therefore, can be considered infinite. Thus, to simplify the analysis of a system, the suction pressure can be considered constant. This is not a general rule, but is used to simplify the following considerations. This yields the simplified system, consisting of a volume filled by a compressor and emptied through a valve (Figure 14).
p p1 1 Qstd = 1360 Fp cv Y 2 p SG T Z 2 2 2
and
Qv = Qstd
std ( p2 , T2 , Z2 )
The compressibility Z2 can be calculated with the Redlich-Kwong equation of state. Equations (3), (4), and (5) mean that the discharge pressure change depends on the capability of the valve to release flow at a higher rate than the flow coming from the compressor. It also shows that the pressure reduction for a given valve will be slower for larger pipe volumes (V). Kurz and White (2004) have shown the validity of the simplified model. The discharge pressure p2 in Equation (3) is a function of the compressor operating point, expressed by:
k
p2 k 1 h(Q, N ) SG k 1 = 1 + 287 ZT1 p1 k
h N
2
Q Q = + + N N
Alternatively, a lookup table, showing the head-flow relationship for the compressor, can be used. The behavior of the compressor during ESD is governed by two effects. The inertia of the system consisting of the compressor, coupling, and power turbine (and gearbox where applicable) is counteracted by the torque (T) transferred into the fluid by the compressor (mechanical losses are neglected). The balance of forces thus yields:
T = 2 J
dN dt
Knowing the inertia (J) of the system and measuring the speed variation with time during rundown yields the torque and, thus, the power transferred to the gas: Figure 14. Train Deceleration and Valve Opening. The basic dynamic behavior of the system is that of a fixed volume where the flow through the valve is a function of the pressure differential over the valve. In a surge avoidance system, a certain amount of the valves flow capacity will be consumed to recycle the flow through the compressor. Only the remaining capacity is available for depressurizing the discharge volume. In such a system, mass and momentum balance have to be maintained (Sentz, 1980; Kurz and White, 2004). From this complete model, some simplifications can be derived, based on the type of questions that need to be answered. For relatively short pipes, with limited volume (such as the systems desired for recycle lines), the pressure at the valve and the pressure at compressor discharge will not be considerably different. Also, due to the short duration of any event, the heat transfer can also be neglected. Therefore, mass and momentum conservation are reduced to:
P = T N 2 = (2) J N
dN dt
If the rundown would follow through similar operating points, then P~N3, which would lead to a rundown behavior of:
dN k N2 k = N 2dN = dt + c N (t ) = 2 dt J (2)2 J (2) 1 t N1
k 2 J ( 2 )
t =0
dp2 k p2 = [Q Qv ] dt V
The rate of flow through the valve is calculated with the standard ISA method (ANSI/ISA, 1995) (Qstd is the standard flow. Fp is the piping geometry factor. It is usually not known and can be assumed to be 1. The pressure is assumed to be constant in the entire pipe volume. It is thus the same just upstream of the valve and at the discharge pressure of the compressor):
Regarding the proportionality factor (k) for power and speed, this factor is fairly constant, no matter where on the operating map the rundown event starts. Thus, the rate of deceleration, which is approximately determined by the inertia and the proportionality factor, is fairly independent of the operating point of the compressor when the shutdown occurred, i.e., the time constant (dN/dt [t = 0]) for the rundown event is proportional to k/J. However, the higher the surge margin is at the moment of the trip, the more head increase can be achieved by the compressor at constant speed. While the behavior of the piping system can thus be predicted quite accurately, the question about the rate of deceleration for the compressor remains. It is possible to calculate the power consumption for a number of potential steady-state operating points. The operating points are imposed by the pressure in the discharge volume, which dictates the head of the compressor. For a given speed, this determines the flow that the compressor feeds into the discharge.
The model described above, which accounts for the primary physical features of the discharge system, can be used to determine whether the combination of discharge volume and valve size can prevent the compressor from surge during an ESD. It allows the two important design parameters to be easily varied to avoid surge during ESD. The surge valve size and opening speed can be increased for a given discharge volume or the maximum allowable discharge volume for a given configuration of valves and compressor characteristic can be limited. The second method, which has the advantage of being more transparent for the station design, is used here. The simplified model calculates the maximum discharge volume where the head across the compressor can be reduced by half in one second, based on the assumption that this reflects the speed decay during an ESD as outlined above. Therefore, the calculation of the instant compressor speed is replaced by a fixed, presumed to be known, deceleration rate. The assumption is made that the power turbine and compressor will lose about 20 to 30 percent speed in the first second of deceleration. This is, for example, confirmed by data from Kurz and White (2004) showing a 30 percent speed reduction of a gas turbine driven compressor set, and, where the gas turbine driven configurations lost about 20 to 25 percent speed in the first second, while the electric motor driven configuration lost 30 percent speed in the first second. As a result of the loss of 25 percent speed, the head the compressor can produce at the surge line is about 56 percent lower than at the initial speed, if the fan law is applied. A further assumption is made about the operating point to be the design point at the instant of the ESD. Any ESD is initiated by the control system. Various delays in the system are caused by the time for the fuel valve to shut completely, the time until the hot pressurized gas supply to the power turbine seizes, and the opening time of the recycle valve. ESD data show it is a valid assumption that the surge control valve reaches full open simultaneously with the beginning of deceleration of the power turbine/compressor. This is the starting time (T0) for the model. Usually, the suction volume (no check valve) is more than three orders of magnitude greater than the discharge volume and is therefore considered at a constant pressure. The general idea is now to consider only the mass flow into the piping volume (from the compressor) and the mass flow leaving this volume through the recycle valve. Since the gas mass in the piping volume determines the density and, thus, the pressure in the gas, we can for any instant see whether the head required to deliver gas at the pressure in the pipe volume exceeds the maximum head that the compressor can produce at this instant. Only if the compressor is always capable of making more head than required can surge be avoided. A further conceptual simplification can be made by splitting the flow coefficient of the recycle valve (cv) into a part that is necessary to release the flow at the steady-state operating point of the compressor (cv,ss) and the part that is actually available to reduce the pressure in the piping volume (cv,avail). The first stream and, thus, cv,ss of the valve necessary to cover it are known. Also known is the cv rating of the valve. Thus, the flow portion that can effectively reduce the backpressure is determined by the difference:
test data. The method described can easily be expanded to situations where a relatively small suction volume leads to a fast increase in suction pressure. Application The model described above can be used in a software simulation program to rapidly evaluate whether a selected valve is sized correctly for the piping volume. The model iteratively determines the maximum allowable discharge volume for a given valve configuration. This is important, because the valve size can be determined early in the project phase. With a known valve configuration, the station designer can be provided with the maximum volume of piping and coolers between the compressor and the check valve, that allows the system to avoid surge during an ESD. The calculation requires specification of the head-flow-speed relationship of the compressor, and the definition of the surge line as a function of either compressor speed, compressor head, or compressor flow. Further, the valve needs to be described by its maximum capacity (Cv), as well as by its capacity as a function of valve travel and the opening behavior, including the delay. The discharge check valve is assumed to be closed as soon as the recycle flow exceeds the compressor flow, i.e., once the depressurization begins. The calculation procedure is started by initiating the deceleration of the train and the valve opening. For each time step, the compressor head and flow, based on speed and system pressures, and the flow through the valve, based on system pressures and valve opening, are calculated. The mass of the gas trapped between the recycle valve and compressor discharge is subsequently determined, yielding a new discharge pressure. If surge occurs, i.e., if the flow drops below the flow at the surge line, the backwards flow through the compressor is assumed to increase with time in surge, with a recovery once the required head drops 1 percent below the head at surge. The modeling of the backwards flow is not critical, and is only made to avoid numerical instabilities, because the only information that is expected from the model is whether or not the compressor will surge for the given configuration. Figure 13 shows the deceleration of the engines power turbine, and thus the driven compressor, following an ESD. Also shown is the response of the recycle valve. Figures 15, 16, and 17 show typical results of these simulations. In Figure 15, the discharge volume is small enough, and while the actual flow of the compressor approaches the minimum allowable flow (surge flow) at about 500 ms after the initiation of the ESD, so surge can be avoided. In Figure 16 the compressor surges about 700 ms after the initiation of the ESD. For this configuration, either the valve size has to be increased, or the discharge volume has to be reduced, to avoid compressor surge during an ESD. In Figure 17, the system is severely underdesigned and will require significant changes including the possible addition of another valve in a hot bypass mode.
cv ,avail = cv cv , ss
The model is run at constant temperature. Most of the compressor systems modeled contain aftercoolers. The thermal capacity of the cooler and the piping are much larger than the thermal capacity of the gas; thus, the gas temperature changes are negligible within the first second. The flow calculated above in each step of the iteration is then subtracted from the gas contained in the discharge and a new pressure in the pipe volume is calculated. The calculation yields the maximum allowable piping volume for the set parameters that will not cause surge at ESD. Kurz and White (2004) validated this approach, using actual
Figure 15. Actual Flow and Flow at the Surge Line During ESD Surge Avoided.
Figure 19. Aftercooler Inside Recycle Loop. Figure 16. Actual Flow and Flow at the Surge Line During ESD Compressor - Surge at 0.7 Sec. Small discharge volume, fast recycle response If there is significant heat in the suction header may improve compressor performance Requires an additional cooler
Precooler and aftercooling (Figure 20)
Figure 20. Precooler and Post Cooling. Figure 17. Actual Flow and Flow at the Surge Line During ESD Multiple Surges.
Cooled recycle loop (Figure 21)
RECYCLE ARRANGEMENTS FOR SPECIFIC APPLICATIONS

The arrangement of recycle loops impacts the operational flexibility, as well as the startup and transient behavior of a compressor station. In the following section, various arrangement concepts are described, together with their basic advantages and disadvantages. Recycle Configurations for Single Compressors Small discharge volume, fast recycle response Although some partial recycle can be maintained, 100 percent recycle cannot.
Small discharge volume, fast recycle response No inline pressure loss Requires an additional cooler, although smaller than a precooler
Basic recycle system (Figure 18)
Figure 21. Recycle Cooling. Provides modulating surge control valve and shutdown valve ideally suited for their purpose More components, more cost
Hot recycle valve and cooled recycle valve (Figure 22)
Fig. 18: The Basic Recycle System (Hot recycle)

Figure 18. The Basic Recycle System (Hot Recycle).
Aftercooler inside recycle loop (Figure 19)
100 percent recycle possible Additional discharge volume impacts recycle response.
Figure 22. Hot and Cooled Recycle Valves.
Parallel recycle valves (Figure 23)
Provide good modulating surge control and fast shutdown valves Provide some level of redundancy More components, more cost
Figure 26. Low Ratio Compressors in Series: Recycle for Individual Compressor.
Figure 23. Parallel Recycle Valves. Recycle Configurations for Multiple Compressors
Hot unit valves and cooled overall recycle valve (Figure 24)
Provides good modulating surge control and fast shutdown valves Provides some level of redundancy More components, more cost
Figure 27. Low Ratio Compressors in Series: Common Recycle Header.
RECYCLE CONNECTION CONSIDERATIONS

Connecting Upstream or Downstream of Scrubbers Connecting upstream of the suction scrubber will prevent debris left in piping from entering the compressor and will add volume in the recycle path to extend time before overheat. Connecting downstream of the suction scrubber will avoid the pressure drop across the scrubber thus pressure rise at the compressor suction will be faster and higher. Startup Considerations
Figure 24. Hot Unit Valves Cooled Overall Recycle Valve. valve for second compressor and overall recycle valve (Figure 25) Provides good modulating surge control and fast shutdown valves Provided some level of redundancy More components, more cost
Compressor 2 significantly larger than compressor 1: recycle
C1
C2
Figure 25. Recycle Valve for Second Larger Compressor and Overall Recycle Valve. Low ratio (low discharge temperature) individual compressors connected in series (Figures 26 and 27) In either configuration any compressor can be started and put online with the recycle valve closed. As the compressor begins to make head its discharge check valve will open and its bypass check valve will close. In the configuration with a common recycle header (Figure 27), multiple parallel recycle valves are necessary. This is due to the fact that the pressure ratio over the recycle valve is reduced if one or more of the compressors upstream are shut down. This in turn requires added flow capacity in the recycle valve. This configuration also requires a number of additional block valves.
The design of the antisurge and recycle system also impacts the startup of the station. Particular attention has to be given to the capability to start up the station without having to abort the start due to conditions where allowable operating conditions are exceeded. Problems may arise from the fact that the compressor may spend a certain amount of time recycling gas, until sufficient discharge pressure is produced to open the discharge check valve (Figure 3) and gas is flowing into the pipeline. Virtually all of the mechanical energy absorbed by the compressor is converted into heat in the discharged gas. In an uncooled recycle system, this heat is recycled into the compressor suction and then more energy added to it. A cubic foot of natural gas at 600 psi weighs about 2 lb (depending on composition). The specific heat of natural gas is about 0.5 Btu/lb (again depending on composition). One Btu/sec equals 1.416 hp. If the recycle system contains 1000 cubic feet, there is a ton of gas in it. One thousand four hundred sixteen hp will raise the temperature of the gas about 1 degree per second. This approximates what happens with 100 percent recycle. At 100 percent recycle, eventually this will lead to overheating at the compressor discharge. The problem usually occurs when there is a long period between the initial rotation of the compressor and overcoming the pressure downstream of the check valve. Low pressure-ratio compressors often do not require aftercoolers. There are several strategies that can be employed to avoid overheating the uncooled compressor during startup:
Accelerate quickly Delay hot gas re-entering the compressor Dilute hot gas re-entering the compressor Throttled recycle
Compressors without cooling must be accelerated and placed online quickly to avoid overheating. Uncooled compressor sets
cannot be started and accelerated to idle. They must be accelerated quickly through the point where the discharge check valve opens and the recycle valve closes. If acceleration slows when the discharge pressure is met, and recycle valve closes slowly, a shutdown may still occur. Often standard start sequences are very conservative and can be shortened to reduce the time it takes to get a compressor online. Extending the length of the recycle line downstream of the recycle valve increases the total volume of gas in the recycle system. This reduces the heat buildup rate by delaying when the hot gas from the compressor discharge reaches the suction. Some heat will be radiated through the pipe walls. If the outlet is far upstream into a flowing suction header, dilution will occur. Figures 28 and 29 outline a solution to a rather difficult starting problem for a compressor station without aftercooling capacity. In Figure 28, to start the first unit is relatively easy, because there is virtually no pressure differential across the main line check valve, and therefore the unit check valve will open almost immediately, allowing the flow of compressed gas into the pipeline. However, if one additional unit is to be started, the station already operates at a considerable pressure ratio, and therefore the unit check-valve will not open until the pressure ratio of the starting unit exceeds the station pressure ratio. Ordinarily the unit would invariably shut down on high temperature before this can be achieved. By routing the recycle line into the common station header (Figure 29), the heat from the unit coming online is mixed with the station suction flow. This equalizes the inlet temperature of all compressors; higher for the compressors already online, lower for the compressor coming online. With this arrangement overheating of a compressor coming online is nearly always avoided.
Figure 30 shows the problem of a conventional system that includes 3000 ft of 24 inch pipe without aftercooling. The temperature in the recycle line starts to rise and, assuming a shutdown set point of 350F, the compressor would shut down after about 20 minutes.
Figure 30. Temperature Build Up. Figures 31 and 32 outline the startup event with the revised system. The power turbine and the compressor start to rotate once the gas producer provides sufficient power. Subsequently, the gas temperature rises, but, because the discharge pressure required to open the check valve is reached fast enough, overheating can be avoided. The temperature rise in the recycle loop during startup is shown as a function of power turbine and compressor speed (Figure 31), gas producer speed (Figure 32), and time (in minutes) (Figure 33). The power turbine starts to turn at about 75 percent gas producer speed, at which point the temperature starts to rise. After the discharge check valve opens (at 0.2 minutes after the compressor starts to rotate, 95 percent gas producer speed and 70 percent power turbine speed), the temperature drops rapidly.
HP
HP
HP
Figure 31. Temperature (F) Versus NPT (Percent).
Figure 28. Original Station Layout.
HP
Figure 32. Temperature (F) Versus NGP (Percent).
HP
HP
Figure 29. Improved Station Layout.
Figure 33. Temperature (F) Versus Time (Min.).
Further analysis of the startup problem indicates the advantage of throttling the recycle valve, rather than starting the unit with the recycle valve fully open. With the valve sizing tool described previously it can be determined exactly what valve opening will be required to maintain a specific surge margin at steady-state operation. As the compressor is accelerating, flow is increasing. The pressure in the discharge is lower and the pressure in the suction is higher than they would be if the compressor at this speed were steady-state. This is due to the effect of the suction and discharge volumes. This also causes the flow to be higher and subsequently the surge margin will be higher. As such, if the valve is set at a fixed position to obtain a fixed small surge margin, the actual surge margin will be higher during acceleration. To use this strategy safely the control must be able to sense a loss of acceleration (flame out) and if detected open all recycle valves immediately. As the volumes up and downstream of the compressor cause the surge margin to be higher during acceleration they make surge avoidance more challenging with loss of speed. Figure 12 illustrates this: at 70 percent open setting, the startup of the compressor is significantly closer to the surge line than at 100 percent open setting. For any given speed, the power requirement of the compressor is lower when it is closer to surge than when it is farther in choke. Therefore, for a given amount of available power, the start is quicker if the compressor operates closer to surge. If the rate of acceleration is quicker, the heat input into the system is lower. Actively modulating the surge during startup is virtually impossible as the parameters defining the surge limit of the compressor are too low to be practically measured. Returning to Figure 5 the surge limit of a compressor matches well with a fixed travel (constant Cv) line for a recycle valve. As such, a compressor can be started with a fixed recycle valve position.
Q SG SM T t V Y Z ,, avail compr op p surge std ss v 1 2
= Volumetric flow = Specific gravity = Surge margin (percent) = Temperature = Time = Volume = Expansion factor = Compressibility factor = Constants = Density
Subscripts = Available = Compressor = Operating point = Polytropic = At surge = At standard conditions = Steady-state = Valve = Compressor inlet = Compressor discharge
REFERENCES
ANSI/ISA S75.01, 1995, Flow Equations for Sizing Control Valves. Arnulfi, G. L., Giannatasio, P., Micheli, D., and Pinamonti, P., 2000, An Innovative Control of Surge in Industrial Compression Systems, ASME 2000-GT-352. Bakken, L. E., Bjorge, T., Bradley, T. M., and Smith, N., 2002, Validation of Compressor Transient Behavior, ASME GT-2002-30279. Blanchini, F., Giannatasio, P., Micheli, D., and Pinamonti, P., 2001, Experimental Evaluation of a High-Gain Control for Compressor Surge Suppression, ASME 2001-GT-0570. Day, I. J., 1991, Axial Compressor Performance During Surge, Proceedings of the 10th International Symposium on Air Breathing Engines, Nottingham, United Kingdom, pp. 927-934. Epstein, A. H., Ffowcs Williams, J. E., and Greitzer, E. M., 1994, Active Suppression of Compressor Instabilities, AIAA-86-1994. Kurz, R. and White, R. C., 2004, Surge Avoidance in Gas Compression Systems, TransASME Jturbo, 126, pp. 501-506. Kurz, R., McKee, R., and Brun, K., 2006, Pulsations in Centrifugal Compressor Installations, ASME GT2006-90070. McKee, R. J. and Deffenbaugh, D., 2003, Factors that Affect Surge Precursors in Centrifugal Compressors, Proceedings of GMRC Gas Machinery Conference, Salt Lake City, Utah. Nakayama, Y., 1988, Visualized Flow, Oxford, United Kingdom: Pergamon Press. Sentz, R. H., 1980, The Analysis of Surge, Proceedings of the Ninth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 57-61.
OUTLOOK
There are ongoing efforts to improve surge avoidance systems. One line of efforts attempts to increase the stability margin of a compressor by active (Epstein, et al., 1994; Blanchini, et al., 2001) or passive means (Arnulfi, et al., 2000). Other efforts try to increase the accuracy of determining the surge margin (McKee and Deffenbaugh, 2003) by detecting the certain precursors of surge. However, most of the ideas will remain valid even if some of the new methods, currently in an experimental stage, are introduced. This is due to the fact that surge avoidance is a systems issue and meaningful gains can be made by better understanding the interaction between the compressor, the antisurge devices (control system, valves), and the station piping layout (coolers, scrubbers, check valves).
CONCLUSION
This paper has addressed the key factors that must be considered in the design of surge avoidance systems. The most important point is the realization that surge avoidance must be viewed in terms of the total system and not as an isolated item looking only at the compressor itself.
NOMENCLATURE
A cv C Fp h Hcooler J k K Kv L N p = Flow area = Flow coefficient (cv = Q/ SG/p) = Compressible valve coefficient = Piping geometry factor = Head = Gas cooler heat transfer = Inertia = Isentropic exponent = Constant = Valve coefficient = Pipe length = Speed (1/s) = Pressure
Table of Contents
GUIDELINES FOR SPECIFYING AND EVALUATING THE RERATING AND REAPPLICATION OF STEAM TURBINES
by Donald H. Mansfield
Manager of Bladed Rotor Application Engineering
Douglas W. Boyce
Senior Application Engineer, Global Services
William G. Pacelli
Manager, Product Engineering, YR Steam Turbines Elliott Company Jeannette, Pennsylvania
ABSTRACT
Donald H. Mansfield is Manager of Bladed Rotor Application Engineering for Elliott Company, Ebara Group, in Jeannette, Pennsylvania. He has more than 33 years experience in the Compressor and Turbine Application Engineering, Project Management, and Marketing groups within Elliott Company. Currently, he is responsible for steam turbine, gas expander, and axial compressor selection standards, specification reviews, and all aspects related to proposal preparation for new and rerated equipment. Mr. Mansfield received a B.S. degree (Mechanical Engineering, 1970) from the University of Pittsburgh. Douglas W. Boyce is a Senior Application Engineer for Elliott Company, Ebara Group, in Jeannette, Pennsylvania, in the Global Services operation. He has been in the application engineering function at Elliott Company for his entire tenure of 38 years. This experience is equally divided in the companys steam turbine, compressor, and vacuum technology product lines. Present responsibility is in engineering of reapplied compressors and steam turbines. Mr. Boyce received a B.S. degree (Mechanical Engineering, 1969) from Lehigh University. He is a registered professional engineer in the State of Texas. William G. Pacelli is the Manager of YR Steam Turbine Product Engineering for Elliott Company, Ebara Group in Jeannette Pennsylvania. He has 24 years experience in turbine design and service engineering. Currently he is responsible for YR steam turbine product engineering, design, and development. Mr. Pacelli received a B.S. degree (Mechanical Engineering, 1982) from the University of Pittsburgh. He is a registered professional engineer in the Commonwealth of Pennsylvania.
135
Steam turbines may be rerated, reapplied, or modified to meet several specific goals. This tutorial will review possible reasons for these changes, including optimizing performance, improving reliability, reducing maintenance requirements, solving operating problems, extending equipment life, and, finally, replacement of the turbine, either in whole or part, due to a catastrophic failure, normal wear, or problems found during an inspection. Whatever the project, it normally requires clear goals specified by the user and buyer. The vendor can then review these goals and determine what can realistically be accomplished within the constraints of the existing equipment, the project budget, and the time allowed. Most of the equipment limitations will be discussed in this tutorial as well as the presentation of a case study. For many of these projects, the vendor will not have access to the turbine since it will still be operating at the users facility. Therefore, closer than normal coordination between the user and the vendor is required to ensure smooth and timely completion of the project.
INTRODUCTION
The first part of this tutorial will define many of the reasons for rerating or reapplying a steam turbine, as well as what hardware or software may be available to accomplish the task. This is followed by a listing of the information needed from the user and an explanation of the limitations faced by the vendorspecifically those that are not usually encountered with new equipment. Finally, a case study of an uprate of a mechanical drive turbine is presented.
PERFORMANCE CHANGES
The basic power and/or speed rating of a steam turbine may change for many reasons. The most common one is an increase (or decrease) in the power required by the driven machine due to a plant expansion or debottlenecking. Other reasons include a reapplication of the turbine to drive a different machine, a search for increased efficiency, a change in the plant steam balance, or a change in steam pressure or temperature. Power An increase in power usually requires more steam flow area inside the turbine, which may or may not be possible within the physical limits of the existing casing. A decrease in power is almost always possible simply by blocking off some flow area, but maintaining efficiency usually requires a more sophisticated
solution. Reapplication of a turbine often is the most difficult problem for the vendor since the power, speed, and steam conditions can all be considerably different from those for which the turbine was initially designed. This is discussed in more detail later in this paper. Efficiency More efficient blading may be available for older turbines. Considering the cost of energy today, it may make sense to invest in new blades (both rotating and stationary) if the gain in efficiency is large enough. However, higher efficiency blading often requires more axial spacing along the rotor, and there may not be enough room in the existing casing. There are simpler ways to improve or maintain efficiency. Leakage through labyrinth seals can be reduced by up to 80 percent by integrating brush type seals with the usual stationary labyrinths as shown in Figure 1.
Labyrinth teeth can be damaged by rubs, especially during startup or coastdown when the turbine rotor passes through a lateral critical speed. Rubs may also occur at startup due to different rates of thermal expansion between the seal and the rotor. The rub opens the clearance and reduces efficiency. Retractable packing is a possible solution if your turbine has this problem. The labyrinth ring is circumferentially divided into segments that are spring-loaded to hold them apart and therefore give a very generous clearance. Once the turbine starts and steam pressure builds up on the outside diameter of the seal, it overcomes the spring pressure and it closes the seal to normal clearance. When the turbine trips, steam pressure is reduced and the spring again opens the clearance for coastdown. Almost all turbine stages have seals between each diaphragm and the shaft, but many turbines do not have tip seals between the rotor blade tips and the casing/diaphragm. If room permits, tip seals can be added to increase efficiency. Tip seals are more effective when used in higher pressure stages and stages with greater reaction. Tip seals are shown in Figure 3.
Figure 1. Labyrinth-Brush Seal. The brush seal consists of bristles that are angled slightly with shaft rotation. They can tolerate some deflection and still spring back to their original position. They can be incorporated between the labyrinth teeth as well as at the ends. For higher pressures, brush seals may require pressure balancing to avoid excessive downstream deflection. Note that due to the angled bristles, some brush seals may not tolerate reverse rotation. Figure 2 shows brush seals in a heavy metal retainer.
Figure 3. Tip Seals. Any of the seals mentioned above can combine the features of the brush seal and retractable seal.
RELIABILITY AND MAINTENANCE

The trend in most industries is for longer runs and shorter turnarounds. There is also a lot of pressure to eliminate unplanned shutdowns and reduce required maintenance. Electronic Controls Perhaps the biggest change in steam turbines in the past 20 years has been the conversion of the speed control and trip systems from totally mechanical to totally electronic. Many existing turbines can benefit from a change to electronic controls. Figure 4 shows a typical mechanical governor system for a straight-through turbine. The governor is powered off the turbine shaft by a worm and wheel drive. The governor itself has many internal moving parts including flyweights, springs, an oil pump, accumulator pistons, and several valves. These governors are quite reliable, but with all those moving, contacting parts, wear and maintenance are inevitable.
Figure 2. Brush Seal in Metal Retainer.
Table of Contents GUIDELINES FOR SPECIFYING AND EVALUATING THE RERATING AND REAPPLICATION OF STEAM TURBINES 137
electronic ones are not powerful enough to move the valve racks. We can go a step further, however, and eliminate the prepilot and pilot valves, in addition to the linkages, cams, and rollers shown as the restoring linkage in Figures 4 and 5. Figure 6 shows that the servo motor can be fed by a way valve. The way valve takes the oil directly from the control oil header and directs it to the proper side of the servo motor. The electronic governor sends a control signal to the actuator coil that is an integral part of the way valve. The restoring linkage is replaced by one or more linear variable differential transformers (LVDTs) that provide feedback to the control system. The way valve may have dual actuator coils for redundancy.
Figure 4. Mechanical Governor Straight Through Turbine. An electronic governor replaces the worm gear with a toothed wheel. Multiple noncontacting magnetic speed pickups read off the toothed wheel and provide the signal to the electronic governor. The power cylinder in Figure 4 is replaced with an electronic or pneumatic actuator. The pilot valve and restoring linkage remain intact. Many electronic governors are also more versatile in that they can be programmed to use parameters other than speed alone to control the turbine. Most electronic governors can also interface with the users distributed control system (DCS). Things get much more complicated for mechanical controls when they are applied to an extraction turbine, as shown in Figure 5. Here the power cylinder controls the added extraction pressure regulator, which includes even more linkages and springs as well as a pneumatic signal to monitor extraction steam pressure. The extraction pressure regulator sends signals to either prepilot cylinder on either or both servos. This system can maintain a set turbine speed and an extraction pressure level. However, the linkages require maintenance, the linkages and springs wear, and if there is a change to extraction pressure, the linkages and springs need adjustment.
Figure 6. Way Valve Retrofit. Turbines are still tripped by dumping the oil that holds the trip and throttle valve open. Figure 7 shows the arrangement formerly used on most turbines. The overspeed trip was initiated by a springloaded trip pin or a weighted Bellville spring that struck a mechanical lever that actuated a dump valve. The solenoid dump valve shown was for remote tripping purposes.
Figure 5. Mechanical Governor Controls for an Extraction Turbine. An electronic governor designed for extraction turbines works as described above for straight-through turbines, with the addition that it receives an electronic signal for extraction steam pressure and outputs signals to the electronic (or pneumatic) actuators that replace the prepilot cylinders on each servo motor. Again other signals (such as a compressor discharge pressure signal) can be fed to and processed by the electronic governor. Here a change to the extraction pressure is a simple set point change. Electronic governors are not failproof, but they are available in redundant and triple modular redundant formats, so failure of electronic components will cause the governor to switch to backup components while the failed components are replaced. Most multivalve turbines still require hydraulic servo motors since
Figure 7. Mechanical Overspeed Trip. Electronic trips, which are standard on most new turbines, use three noncontacting magnetic speed pickups that read off a toothed wheel. This is the same method that sends the electronic governor its signal. The electronic overspeed trip control box uses two out of three voting logic to trip the turbine and eliminate spurious trips. Two solenoid dump valves are piped in parallel to provide redundancy. It is sometimes difficult to find enough room on an
existing turbine to mount the necessary speed pickups, but beyond that it is fairly simple to add an electronic overspeed trip. Monitoring Systems There are many steam turbines that have been in operation for 30 or more years that do not have the level of instrumentation that is considered standard today. Radial and axial vibration probes, as well as bearing temperature instrumentation, may be fitted to existing turbines. Readouts from these instruments can be fed to sophisticated monitoring equipment that not only can display the current readings, but also can determine trends that may uncover a potential problem before it causes an unexpected outage. The instruments could indicate that a change to a different bearing (tilt shoe, spherical seat) or orientation, such as load between pad, would be beneficial. The main stumbling block to retrofitting vibration probes is finding the space in the bearing housings to mount them with a corresponding free area on the shaft to read from. The shaft area may also have to be treated to reduce electrical runout to an acceptable level. Bearing temperature instrumentation will normally require some machining of the bearing retainers and housings to allow the wires to exit the turbine. The user should be aware that it is fairly common for machines that have been operating satisfactorily for many years to show high vibration readings once probes are installed. The excessive vibration may have always been there, but it just has not been to the point where it has caused a problem. In such cases, the vibration level is normally still not a problem. Although it exceeds current standards, if the vibration has not caused any operating or maintenance problems in the past, there is no reason to believe that it will be a problem in the future. With the probes installed, however, it is now possible to monitor and trend the vibration levels and to avoid potential future problems.
matter. This is an emergency and a quick solution is essential. A new turbine is usually out of the question because the lead time is too long. Depending upon the extent of the damage, most original equipment manufacturers (OEMs) and shops that specialize in turbines can accomplish fairly major repairs in a short period of time by utilizing welding, plating, and other proven repair techniques. Blades and other specialized parts might be available in a reasonable amount of time. Catastrophic damage will require a search for a suitable replacement. Again, most OEMs and turbine specialty shops keep track of equipment that is available on the used equipment market. The vendor may even have used turbines of his own to reapply. A suitable turbine may be overhauled with minor changes and be back in service in a fraction of the time required for a new unit. If no suitable turbine can be found on the used market, it may even be possible to reapply another of the users turbines (possibly in a less critical service) to replace the damaged turbine. It may then be feasible to replace that turbine with either a new or used unit.
STARTING THE PROCESS

Define the Objective Although not considered a vital part of the process, possibly the most critical part of the project is to define the objective of the project. Many times projects are initiated without a clear objective. This lack of a defined goal often leads to failure of the entire project. The process starts by an analysis of the driven machine to determine the actual power requirement and speed and evaluation of steam conditions at all the expected operational points, including excess power margin. Steam pressure drops in the supply, exhaust, and extraction piping must be accurately calculated. On extraction units, the extraction pressure and flow must also be defined. In the initial purchase, many turbine standards were considered, including corporate, American Petroleum Institute (API), National Electrical Manufacturers Association (NEMA), industrial, and local governmental standards. It is up to the engineering staff directing the changes to determine which, if any, of these specifications are required to be applied. Some of the original specifications or specifications that have been enacted since the units original commissioning may need to be applied. One note of caution: older turbines must be carefully analyzed, as many older units cannot meet current standards or may only do so at considerable added lead time and expense. Construction Rating Turbines undergoing a change in inlet pressure, temperature, and/or exhaust pressure require that maximum casing pressure and temperature rating be checked. This includes a review of the casing and flange ratings to ensure the casing is suitable for the new conditions. Multistage turbines with changes in pressure and flow require a verification of the first stage maximum allowable pressure. In the case of a multistage extraction turbine, a change in flow or extraction pressure will also require that intermediate casing and extraction flange be reviewed. A change in inlet temperature will require a verification of the casing material. Typical limits are given in Table 1. Table 1. Casing Material Temperature Limits.
Casing Material ASTM A278 Class 40 ASTM A216 Grade WCB ASTM A217 Grade WC1 ASTM A217 Grade WC6 ASTM A217 Grade WC9 Maximum Temperature (Deg F) 600 775 825 900 1000
OPERATING PROBLEMS
Many operating problems may be solved by some of the items that have already been discussed. Monitoring systems can uncover problems such as a bowed rotor or fouling. There are also much better tools to determine rotordynamic characteristics than there were 30 years ago. The seals that were discussed can eliminate water contamination of the oil and excessive gland leakage. The electronic controls can cure some process control problems or speed fluctuations. Speed fluctuation can also be caused by improperly sized governor valves, or valve and seat wear. Performance problems have to be analyzed on an individual basis. Again, there are more sophisticated tools available now than there were just a few years ago, such as computational fluid dynamics (CFD).
LIFE EXTENSION
There are several options available to extend the life of an existing turbine. Most types of rotor damage can now be repaired by machining off the damaged area, rebuilding it with weld, and remachining the rotor. Some casing damage may be repaired in a like manner. Better materials are often available to solve erosion problems. Blades also can be protected by a stellite overlay in critical areas. Some coatings are also available that may help with erosion, corrosion, or fouling problems. A bearing upgrade or something as simple as an at-speed balance could also extend the life of a turbine by eliminating a vibration problem.
REPLACEMENT
If the user knows in advance that they will be replacing their turbine, they have many options. A new turbine is certainly an option. Or there may be used turbines on the market that can be refurbished to meet the requirements. The user may even have another turbine that they would like to reapply to the existing service. If the turbine is damaged in a wreck, it is an entirely different
There are many other materials that have been used over the years, but these are the major casing materials currently in use. Before exceeding these limits, a thorough review of the metallurgy of the existing casing must be done. Once the casing rating has been determined, the original hydrotest pressures must be determined. This can be done by contacting the original OEM, reviewing the original construction documentation (i.e., hydrotest certificates, data sheets, correspondence), or reviewing the turbine casing for markings. The hydrotest pressure must meet the new conditions based on the latest industry standards. The API guidelines refer to ASME Boiler & Pressure Vessel Code - Section VIII (2004). In some cases, the casing will need to be rehydrotested to the new conditions. This may need to be done to one or several sections or the entire turbine casing. A decision must be made as to whether NEMA limits need to be taken into account. For new construction, most OEMs verify the casing design will ensure the casing will meet the maximum design pressure and temperature plus 120 percent of the rated pressure combined with a temperature increase of 50F. This verification is also included in the evaluation of the flange rating. Flange Sizing If during the modification of the turbine, the flow increases or the specific volume decreases, the size of the inlet flange must be reviewed. Typical maximum value for the inlet velocity is 175 ft/sec. Equation (1) determines the inlet velocity.
Figure 9. Inlet Casing with Dual Inlet Flanges. The trip and throttle (T&T) valve must also be reviewed, but this will be discussed in a later section. As a worst case option, the steam velocity can be allowed to exceed the 175 ft/sec limit as long as the correct pressure drop is taken into account and additional acoustic protection is provided as the noise level will increase. The extraction and exhaust line sizes are also areas that need to be considered. The maximum value for an extraction line is 250 ft/sec. The maximum for a noncondensing exhaust is 250 ft/sec and for a condensing exhaust it is 450 ft/sec. Typically there are not many options to upgrade an extraction or exhaust connection. The most practical solution is to increase the exhaust header size as close to the turbine casing as soon as possible. Again the appropriate pressure drop calculations must be done to determine the pressure at the flange. It will be an iterative procedure with good communication between the engineering staff involved in the rerate and the engineering staff with site responsibility. Nozzle Ring Capacity After the external issues have been decided, the first internal component requiring review is the nozzle ring or nozzle block. This is the inlet to the control stage that ultimately controls the amount of steam a given turbine will be able to pass. If during the original manufacture of the turbine, the nozzle ring had additional area available, the turbine may be uprated by adding additional nozzles. Depending on the design, this can be as easy as installing a new nozzle ring. Some small turbines have nozzles machined into the steam end and modifying these turbines is difficult, if not impossible. It may be cost effective to purchase a replacement turbine if this is the case. If the change in flow is significant, the nozzle may have to be replaced with a nozzle ring that has an increased nozzle height. This will normally require increased blade height on the turbine rotor. The maximum nozzle height will be dictated by the height of the opening in the steam end of the turbine. The maximum number of nozzles in each bank or segment of the steam end is dictated by the original design parameters. Many times, the maximum nozzle area in a steam end is matched to the volumetric area of the inlet of the turbine. The same issues affect the extraction nozzle ring on an extraction turbine although the extraction diaphragm in some instances can be changed or modified to allow for additional nozzle area. Steam Path Analysis Once the inlet nozzling has been reviewed, the remainder of the steam path must be reviewed. An increase in flow may require an increase in diaphragm nozzle height and rotor blade height in some or all stages. As stage flow increases, the heat drop across the stage will increase with the associated decrease in velocity ratio. The velocity ratio is defined as the ratio of the blade at the pitch diameter to the steam jet velocity, as defined in Equation (2).
V =.0509 M / d 2
where: V M d = Velocity in flange (ft/sec) = Mass flow (lb/hr) = Diameter of the inlet (inch) = Specific volume of steam (cu ft/lb)
In smaller turbines, the inlet flange with the steam chest and control valving can easily be changed. On larger turbines, increasing the inlet size may be very difficult. Some turbines have the ability to add an additional inlet on the existing steam chest by welding on a flange followed by a local stress relief. Figure 8 shows a flange undergoing modification from an 8 inch inlet to a 10 inch inlet.
Figure 8. Increasing a Flange in Size. Some turbines may have a blank flange that can be easily removed and piped for additional inlet area. Refer to Figure 9 for a typical steam end with dual inlet capability.
Vo =
DN 224 h
where: Vo D N h
= Velocity ratio = Pitch diameter of rotating blades = Shaft rotational speed = Enthalpy drop across the stage
In order to increase the efficiency, the diaphragm and blade path areas must be carefully matched. Each stage design has an optimum velocity ratio with the equivalent maximum efficiency. During the modification, some of the stages will require changes for mechanical or other reasons. Some of the stages will be acceptable with less than optimal efficiency. Decisions must compare the loss in efficiency versus the cost of modifying each stage. Rotor Blade Loading With the change of the power, flow, or speed of a turbine, each rotating row of blading requires review of its mechanical properties. This includes a review of the speed limits and analysis of both Goodman and Campbell diagrams. The blade mechanical speed limits are related to the disk and blade geometry, stress values for material, and shroud design. The Campbell diagram graphically compares the blade natural frequency to the turbine speed range. The natural frequencies of the blading can be obtained from the OEM or determined experimentally. If the speed range is not changed in the modification, the Campbell diagrams will not change except for any blading that has been changed. Figure 10 illustrates a typical Campbell diagram.
for a Goodman number. There may be several values depending on the location of the blade within the steam path; typically partial arc admission and the moisture transition stages will have higher limits. A significant change in the stress levels of the blading may require replacement or redesigning of the blading. In some rare cases, no blading design can be found and low Goodman numbers may need to be accepted.
Figure 11. Goodman Diagram. Thrust Bearing Loading In many cases, an increase in flow and speed may increase the thrust values. Depending on the age of the turbine in question, a replacement thrust bearing may be required. There are higher capacity thrust bearings available that will fit in the existing cavity. In the most extreme cases, a bearing housing with a larger thrust bearing may be required. Governor Valve Capacity During the design cycle, the capacity of the governor valve must be checked to ensure the correct flow area is present. The valve area must be carefully matched to the area of the nozzle ring. In most cases the nozzle area should be the limiting point of the flow in a valve bank. Setting the valve as the limit will not provide the most efficient conversion of pressure into velocity. If the valve(s) and seat(s) require an increase in size, most casings can accept a larger size valve and seat. Some single valve turbines are able to have the steam chest size increased to the next larger size. Shaft End Torque Limit The Goodman diagram will evaluate the combined effects of alternating stresses and steady-state stresses in the turbine root and base of vane locations. This diagram contains a Goodman line, which in theory represents the minimum combination of steady-state stress and alternating stress, above which a blade fatigue failure could occur. Figure 11 is a typical Goodman diagram. The OEM will normally have minimum allowable values With an increase in power or a decrease in speed, the shaft end stress will need to be reviewed. The basic equation required to calculate shaft stress is given in Equation (3).
Figure 10. Campbell Diagram.
321000 P Nd 3
where: P N d
= Shaft shear stress = Power (hp) = Shaft speed (rpm) = Shaft diameter (inch)
The limit of the shaft shear stress is a function of the material and the heat treatment as well as the shaft end design. Typical turbine shaft ends are single or double keyed, NEMA taper, hydraulically fit, or integral. Each of these has different allowable stresses depending on the stress risers within the design. Typical turbine shaft materials are given in Table 2. Table 2. Turbine Shaft Materials.
Commercial Designation AISI C-1040 AISI 4140 AISI 4340 ASTM A470 Class 4, 7, 8 Material Medium Carbon Steel Chrome Moly Alloy Steel Nickel Chrome Moly Alloy Steel Nickel Chrome Moly Vanadium Alloy Steel Forging Typical Maximum shear stress (PSI) 5000 11500 12500 11500, 12500,12000
The loading of the driven machine may also have an effect on these values. For example, a generator drive may reduce the allowable stress further to account for short circuit loading. Speed Range Changes Although this has been briefly discussed in other sections, care must be taken when making significant changes to the speed range. The most important is the blading speed limits (refer to the previous section, Rotor Blade Loading), but other areas require analysis such as lateral critical speeds, torsional critical speeds, coupling speed limits, etc. Lateral critical speeds are a function of the rotor design, bearing design, and bearing support design. Typically the original critical speed is denoted on the nameplate for the turbine and also in the vendor literature. If no changes are being done to the rotor and operational data agree with the noted critical speed, this value normally will not change. Should changes be made to the rotor or bearing system, a new lateral analysis will need to be done to confirm the speed of the lateral critical speed. A typical report may cost between $3000 and $30,000 depending on the complexity of the rotor and what existing data can be reused. Torsional critical speeds will be required. Each body in the string must be modeled as well as each coupling and gear set. Direct drive units are not normally an issue, but strings with gears require the torsional to be checked. These may not be as readily available as the lateral critical speed. Normally this documentation is supplied by the vendor in the form of a report with engineering documentation. The cost to complete a report will vary from $3000 to $10,000 per body. The turbine governor, which will need to be modified or reprogrammed to meet the new speeds as well as the trip mechanism, could either be mechanical or electronic. Other driven systems will require review such as a gear (if used) or any shaft driven oil pumps. Auxiliary Equipment Review Each turbine has auxiliary equipment that will require some review during any modification. These may include the T&T valve, other steam block valves, leakoff and sealing steam system, surface condenser system, relief valve sizing, lubrication and control oil systems, valve actuation system, and supervisory instrumentation. The T&T valve may be acceptable as is for the change in the flow, but may be upgraded for reliability or the addition of a manual
exerciser. Some T&T valve OEMs are no longer actively pursuing this market, so an upgrade may be more expensive than a complete new valve of a current design. For extraction turbines, the nonreturn valve should also be looked at to determine acceptability of long-term continued operation. Other steam valves in the system should be reviewed for proper sizing and good mechanical operation. The leakoff and sealing steam system must be reviewed to determine if additional leakoff flow will be experienced with the new conditions. API 612 (1995) and 611 (1997) both require 300 percent capacity in the leakoff system. If a change has been made to the first stage pressure or the exhaust pressure, this additional capacity may be difficult to achieve without major rework. The gland condenser will require a review along with the ejector or vacuum pump to ensure these are of adequate capacity. From a piping point of view, it may be a good decision to replace some or all of the leakoff and drain piping if significant corrosion or buildup is noted on the inner diameter (ID) of the piping from the turbine. On a condensing turbine, the surface condenser system must be reviewed to ensure the proper capacity is available with the current cooling water temperature and available cooling water flow. The hotwell capacity and pump sizing must be verified to ensure the correct hotwell level can be maintained. Instrumentation connected with the condenser system should be reviewed and upgraded at this time. When modifying a backpressure turbine, the relief valve sizing must be reviewed once the final flow capacity is determined to ensure the exhaust casing is protected from overpressurization. With a condensing turbine, the relief valve or rupture disc sizing must be reviewed to ensure the exhaust casing is prevented from going over the maximum pressure rating, which is normally 5 psig. The lubrication and control oil system must be carefully checked to ensure the new required oil flow and the proper cooling capacity are available. If the control oil system is being modified, check to ensure the proper oil pressure and flow are available.
CASE STUDY
This example illustrates the rerate procedure for a large mechanical-drive steam turbine in a process plant. The turbine of interest drives a compressor train in ethylene charge gas service. The compressors were to be rerated as part of a plant expansion, and the as-built turbine power rating was insufficient for the planned compressor rerate. Necessary turbine performance for the driven equipment modifications is shown in Table 3. Table 3. Summary of Performance Changes.
Parameter
Change (Rerate vs. Original) +25% -1% No Change No Change As Required
Rated Power Rated Speed Steam Inlet Pressure Steam Inlet Temperature Exhaust Pressure
There is no extraction steam requirement in this application. The scope of hardware change to affect this magnitude of power increase was beyond that of an efficiency upgrade alone. Steam flow could roughly be expected to increase by the same 25 percent as rated power. The flow limit of each steam path section was evaluated in order to use the existing casing. Using Equation (4), inlet steam flow, G, is:
S . R. P = G
where S.R. is the steam rate in lb/hp-hr and the rated power, P, is in horsepower. For this rerate, power is 25 percent higher than the original design. Estimated inlet steam inlet flow is:
. = 348,563 lb / hr 7.50 37,180 hp 125

For this particular turbine, the original design specifications and turbine layout were available for reference. As a first step, the engineer prepared a turbine summary shown in Table 4. Table 4. Turbine Summary.
Frame size Rated Power Rated Speed Maximum Continuous Speed Inlet Steam Pressure Inlet Steam Temperature Exhaust Pressure Steam Rate Inlet Flange Size Exhaust Flange Size 2NV-11 37,180 hp 4200 rpm 4200 rpm 600 psig 750F 2 in. Hg Abs. 7.50 lb/hp-hr 10 CL900FF 87 inches 110 inches rectangular (100 inches diameter equivalent) 92% Curtis control stage 11 Rateau stages, all impulse type, 32 inch to 54 inch pitch diameter 2 inch average on the Curtis stage increasing to 16 inches 12.5 inches 7.0 inches 7 inch diameter, tilt pad, at inlet end 8 inch diameter, tilt pad, at exhaust end 112.5 square inches Labyrinth Oil relay
buckets only on the Curtis control stage. The same high strength 900 lb nozzle and bucket profiles were retained. Existing staging downstream of the Curtis stage would be retained or replaced as indicated by shop inspection. Calculations showed that the existing 12-stage steam path could not pass the required additional flow. As flow in the model increases, stage pressure at each successive nozzle row also increases. This takes place in an actual turbine because nozzle area is fixed. In this case, pressure at the first stage nozzle exit rose to the point at which even new larger nozzles became choked. In general, flow can increase through a certain size nozzle until the resulting pressure drop declines to a critical ratio of approximately 0.6. For ratios higher than 0.6, nozzle flow cannot increase and the nozzle is described as choked. The only available means of reducing the first stage nozzle exit pressure was to remove nozzles and buckets, starting with the second stage. The engineer determined that the required power could only be achieved by removing the existing second and third stages. Pressures at each remaining downstream nozzle row were higher than in the original power rating. The diaphragm of the new second stage (original fourth stage) was replaced with a reinforced weldment. Removal of the second and third stages relieved the first stage nozzle exit pressure to an acceptable ratio with inlet pressure. Diaphragms and rotor discs of these stages were replaced with a flow tunnel to smooth the flow in the empty area of the casing. Performance with the new 10-stage steam path is shown in Table 6. Table 6. Calculated Rerate Performance.
Exhaust Quality Staging Description
Parameter Rated Power Rated Speed Rated Steam Flow Inlet Steam Pressure Inlet Steam Temperature Exhaust Pressure Exhaust Quality
Calculated 46,350 hp 4150 rpm 336,210 lb/hr 600 psig 750F 4 in. Hg Abs. 91%
Rotating Blade Heights
Shaft Diameter Between Rotor Disks Nominal Shaft End Diameter Journal Bearings
Thrust Bearing Main Casing Seals Governor type
The rerate stage selection is compared with the original arrangement in Table 7. Note that stages downstream of the original third stage are unchanged. Table 7. Rerate Staging Comparison with Original Staging.
The engineer next qualified the turbine casing for this rerate by calculating velocities at the inlet and exhaust flanges. Exhaust pressure for this rerate is permitted to vary, so the engineer assumed a pressure of 4 in HgA. Table 5 shows a comparison of flange velocities with NEMA maximum velocity. Table 5. Flange Velocity Estimate. Flange Calculated Velocity 196 289 NEMA Table 8-1 Maximum 175 450
Casing Stage No. 1 2 3 4 5 6 7 8 9 10 11 12
10 Main Inlet 87 110 Exhaust
Inlet velocity exceeding NEMA maximum did not disqualify the turbine casing at this phase of the rerate. However, this area of the turbine would have to be carefully checked after actual performance was calculated. The engineer proceeded to the next step of selecting a preliminary steam path for the rerate. However, axial space available on the rotor and casing dimensions allowed larger geometry nozzle and
Original Design Nozzle Ht. 1.125 in 0.813 in 0.938 in 1.125 in 1.500 in 1.875 in 2.375 in 2.450 in 3.600 in 4.170 in 6.500 in 14.76 in
Rerate Design nozzle ht. 1.250 in ----1.125 in 1.500 in 1.875 in 2.375 in 2.450 in 3.600 in 4.170 in 6.500 in 14.76 in
To confirm suitability of this turbine steam path, the engineer conducted a series of limit checks, starting with flange velocities, shown in Table 8. Table 8. Rerate Flange Velocity.
The new conditions required a minimum shaft diameter of 6.8 inches. Since the existing shaft end diameter was 7.0 inches, no new shaft or subarc weld buildup of the existing shaft end was required. CONCLUSION
Flange
Calculated Velocity 189 279
10 Main Inlet 87 110 Exhaust
NEMA Table 8-1 Maximum 175 450
NEMA SM 23 (1991) maximum inlet velocity may be regarded as a suggested limit. Exceeding this velocity did not disqualify the turbine casing. To provide sufficient flow area in the steam chest and reduce losses, all governor valves were replaced. Blade stress was acceptable for all rows of buckets. Governor speed range did not change, so blade frequency was of concern only on the new Curtis bucket rows. Evaluation showed that there was no interference with natural modes for these buckets within the operating speed range. The engineer evaluated the turbine shaft end diameter at the rerate power of 46,350 hp. From Equation (6), we have minimum shaft end diameter, d, equal to:
Rerating or reapplying steam turbines can save considerable time and money compared to buying new units. It is usually possible to get more power and/or speed out of existing units by changing items in the steam path. Efficiency can often be improved at the same time. Older steam turbines can be brought up to present day turbine standards by retrofitting items such as more efficient seals, modern bearings, electronic controls and, monitoring/trending systems. Turbine life can be extended by changing to better materials, adding coatings, and repairing existing damage. These courses of action, however, have more physical constraints and normally require closer coordination between the buyer, user, and vendor than in the purchase of a new machine. Attention to detail in these matters will result in a successful project.
REFERENCES
ASME Boiler & Pressure Vessel Code - Section VIII - Pressure Vessels, 2004, American Society of Mechanical Engineers, New York, New York. API Standard 611, 1997, General-Purpose Steam Turbines for Petroleum, Chemical, and Gas Industry Services, Fourth Edition, American Petroleum Institute, Washington, D.C. API Standard 612, 1995, Special-Purpose Steam Turbines for Petroleum, Chemical, and Gas Industry Services, Fourth Edition, American Petroleum Institute, Washington, D.C. NEMA Standard Publication SM-23, 1991, Steam Turbines for Mechanical Drive Service, National Electrical Manufacturers Association, Washington, D.C.
d =3
321,000(P) (S )(N )
Maximum allowable torsional shear stress for the existing shaft material was 11,500 psi. The equation for diameter then became:
d =3
321,000(46,350) (11,500)(4150)
d = 6.8 inches
Table of Contents
API SPECIFICATION REVIEW FOR GAS TURBINE DRIVEN TURBOCOMPRESSORS

by Klaus Brun
Program Manager
and Jeff Moore

Principal Engineer Southwest Research Institute San Antonio, Texas
Dr. Klaus Brun is currently Program Manager at Southwest Research Institute, in San Antonio, Texas. His experience includes positions in business development, project management, engineering, and marketing at Solar Turbines, General Electric, and ALSTOM Power. Dr. Brun is the inventor of the Single Wheel Radial Flow Gas Turbine, the Semi-Active Compressor Plate Valve, and co-inventor of the Planetary Gear Mounted Auxiliary Power Turbine. He has authored over 40 papers on turbomachinery, given numerous technical lectures and tutorials, and published a textbook on Gas Turbine Theory. He is the Chair of the ASME-IGTI Oil & Gas Applications Committee, a member of the Gas Turbine Users Symposium Advisory Committee, and a past member of the Electric Power and Coal-Gen Steering Committees. Dr. Brun received his Ph.D. and M.Sc. degrees (Mechanical and Aerospace Engineering, 1995, 1992) from the University of Virginia, and a B.Sc. degree (Aerospace Engineering, 1990) from the University of Florida. J. Jeffrey Moore is a Principal Engineer at Southwest Research Institute, in San Antonio, Texas. His professional experience over the last 15 years includes engineering and management responsibilities at Solar Turbines, Inc., Dresser-Rand, and Southwest Research Institute. His interests include rotordynamics, seals and bearings, finite element analysis, controls, and aerodynamics. He has authored more than 10 technical papers in the area of rotordynamics and aerodynamics and has given numerous tutorials and lectures. Dr. Moore received his B.S., M.S., and Ph.D. degrees (Mechanical Engineering, 1991, 1993, 1999) from Texas A&M University.
(2000), 671 (1998), and 677 (1997) as they apply to gas turbine driven compressors. Critical sections of the codes are discussed in detail with a special focus on their technical interpretation and relevance for the purchasing scope-of-supply comparison, machine testing, and field operation. Technical compliance issues for package, core engine, instrumentation, and driven compressor are addressed individually. As API 616 (1998) forms the backbone for most oil and gas combustion turbine acquisitions, it is covered in more detail. Some recommendations for acceptance of manufacturer exceptions to API and the technical/commercial implications will be provided. A brief discussion of the relevance NFPA 70 (2002) to gas turbine driven compressor sets in oil and gas applications will also be included. This tutorial course is intended for purchasing, operating, and engineering staff of turbomachinery user companies.
INTRODUCTION
American Petroleum Institute (API) specifications are generally applied to oil and gas turbomachinery applications rather than to large industrial power generation. Oil and gas applications of gas turbines have requirements that are inherently different from those of the electric power industry. Namely, oil and gas applications, and customers require:
High availability/reliability Ruggedness High power/weight ratio Efficiency often not critical
While industrial power generation customers have different critical requirements, namely:
Cost of electricity Efficiency Cost of operation and maintenance (O&M)

Because of these inherent market differences, oil and gas customers often insist on compliance with API codes and are willing to accept the resultant higher turbomachinery package costs.
ABSTRACT
This tutorial provides an overview of the latest edition of American Petroleum Institute (API) Code API 616 (1998) and also provides a brief summary of API 614 (1999), 617 (2002), 670
145
OIL AND GAS TURBOMACHINERY APPLICATIONS

Many applications within the oil and gas industry require the usage of turbomachinery equipment for compression, pumping, and generation of electricity. These applications are generally
divided into three separate areas from the production of oil to the sales of the refined product. These are upstream, midstream, and downstream. Upstream refers to the oil and gas production and gathering process, midstream is the oil and gas transportation process, and downstream is the refining and distribution of oil and gas products. For each area and defintiion, different processes that require turbomachinery are listed below. Upstream
Self-generationPower generation to meet needs of oil field or platform Enhanced oil recovery (EOR)Advanced technologies to improve oil recovery Gas liftInjecting gas into the production well to help lift the oil WaterfloodInjection of water into the reservoir to increase reservoir pressure and improve production Gas reinjectionReinjection of natural gas into the reservoir to increase the reservoir pressure Export compressionInitial boosting of natural gas pressure from field into pipeline (a.k.a. header compression) Gas gatheringCollecting natural gas from multiple wells Gas plant and gas boostProcessing of gas to pipeline quality (i.e., removal of sulfur, water, and CO2 components) Gas storage/withdrawalInjection of gas into underground structure for later use: summer storage, winter withdrawal
Midstream
Filter (simple, duplex) Pumps (main, pre/post, backup) Accessory gear Fire/gas detection system Starter/helper drive Pneumatic, hydraulic, variable speed alternating current (AC) motor Controls and instrumentation (on-skid, off-skid) Seal gas/oil system (compressors) Outside the Package
Enclosure and fire protection system Inlet system
Air-filter (self-cleaning, barrier, inertial, demister, screen) Silencer Exhaust system Silencer Stack Lube oil cooler (water, air) Fuel filter/control valve skid Motor control center Switchgear, neutral ground resistor Inlet fogger/cooler Within API there are standards that address each one of the above package elements; however, the focus of most API specifications is on packaging and ancillaries, rather than on gas turbine core components.
Pipeline compressionCompression station on pipeline to pump natural gas typically 800 to 1200 psi compression Oil pipeline pumpingPumping of crude or refined oil LNG plantLiquefied natural gas (LNG) plant; NG is cooled and compressed for transportation in liquid form LNG terminalLoading and unloading of liquid natural gas onto LNG tanker (vessel)
Downstream
GAS TURBINE PACKAGE SYSTEMS

As turbocompressors are complex mechanical devices, they require electronic and mechanic controls as well as instrumentations. The control system of a gas turbine driven compressor must, as a minimum, provide the following functions through the use of a human machine interface (HMI): Equipment startup, shutdown, and protective sequencing Stable equipment operation Alarm, shutdown logic Backup (relay) shutdown Driven load regulation Fuel/speed control Process control Surge control Communication (supervisory control and data acquisition, SCADA) interface API sets minimum functionality, redundancy, alarms, and shutdown requirements of the HMI. The control system also interfaces gas turbine instrumentation for sensing and emergency alarms/shutdowns. Typical instrumentation that is provided with a turbocompressor package can be divided into sensors for control instruments and alarm/shutdown instruments. Control sensing provides feedback for supervisory control of the machine while alarm/shutdown sensors (or switches) are primarily intended to protect the machine from damaging failures. However, there are a number of instruments on the package that serve both functions. For example, compressor discharge temperature measurements can be used both for control and high temperature shutdown. A list of typical instrumentation, classified as sensing and control versus alarm and shutdowns devices, is shown below. Sensing and Control Instrumentation
RefineryProcessing of crude oil into its various sellable components Integrated gasification combined cycleAdvanced process to convert heavy oils and pet-coke into synthesis gas and hydrogen Methanol plantPlant that produces methyl alcohol from methane (natural gas) Fischer-Tropsch liquidArtificial gasoline produced from coal and/or other lower cost hydrocarbons Fractional distillation columnDistillation tower to separate crude oil into its gasoline, diesel, heating, oil, naphtha, etc. CrackingBreaking large hydrocarbon chains to smaller chains (chocker, visbreaking, thermal, catalytic) BlendingMixing of hydrocarbons to obtain sellable refinery products DistributionDelivery of oil and gas products to end-users
TURBOCOMPRESSOR COMPONENTS AND LAYOUT
Most major elements of a typical turbocompressor can be classified as either inside the package or outside the package. Inside the package generally refers to all equipment inside the gas turbine enclosure, while outside the package refers to equipment connecting to the package, such as ancillaries and auxiliaries. The main turbocompressors equipment, classified by this definition, is listed below. Inside the Package
Machinery monitoring and protection
Fuel system and spark igniter
Natural gas (control valves) Liquid (pumps, valves) Bearing lube oil system Tank (overhead, integral)
Temperature sensors (resistance temperature detectors [RTDs] and thermocouples) Lube oil drains Lube oil cooler in/out
Table of Contents API SPECIFICATION REVIEW FOR GAS TURBINE DRIVEN TURBOCOMPRESSORS 147
Lube oil header Fuel gas Bearing metal (radial and axial thrust bearings) Gas path (T1, T2, T4, T5-spread) Driven compressor (Ts, Td) Driven generator stator coils Pressure sensors Gas path (P1, P2, P4) Fuel gas Air inlet filter Delta P Lube oil header Lube oil filter Delta P Driven compressor (Ps, Pd) Flow sensors Surge control (driven compressor) Position sensors (proximity probes) Measures relative movement between housing and shaft Gas turbine (GT) bearings X, Y, Z Gearbox bearings X, Y, Z Driven equipment X, Y, Z Velocity probes or accelerometers Measures absolute motion Gas turbine case Main gearbox case Auxiliary gearbox case Driven equipment case Critical Alarm and Shutdown Switches
Fire fighting CO2 system Emergency power supply (diesel generators, tanks) Instrument and seal gas air compressor Slug catcher, scrubbers, filters Flare and/or blowdown Station and unit flowmeters Pig launcher Gas treatment plant (dehydration, CO2 removal, sour gas treatment)
CODES AND STANDARDS
Many codes and standards are used by turbocompressor manufacturers and users to specify, select, and characterize their equipment. Codes are generally used for the following purposes: Example: Vibrations, materials, service connections, package design Intent to facilitate, manufacture, and procurement No code designs a gas turbine Codes are by nature rather general. Any particular application may require modifications. For turbomachinery applications in the oil and gas sector, the most commonly used standards are:
Codes provide design guidelines and definitions
Lube oil level, temperature, and pressure Lube/seal oil filter Delta P Lube oil drain temperature Casing vibration Proximity probe vibration Axial thrust position Fuel gas temperature and pressure Gas and fire detectors Inlet filter Delta P Magnetic speed pickup (shaft speed) Exhaust temperature T-5 spread Flame-out (flame detector) Bearing metal temperatures Compressor discharge pressure Seal gas or oil pressure and temperature (driven compressor) Seal vent leakage (dry gas seal) Buffer air Delta P Synchronization/grid frequency (driven generator) Manual emergency shutdown (ESD) button
COMPRESSOR STATION LAYOUT
Most turbocompressors are installed in gas compression applications, most often along major gas pipelines. Since turbocompressor codes and standards sometimes affect equipment outside the package, the turbocompressor is tightly integrated into the station process. The major equipment at a pipeline compression station is listed below:
API 616 (1998)Gas turbines API 617 (2002)Centrifugal compressors API 614 (1999)Lube oil system API 670 (2000)Machinery protection API 613 (2003)Load and accessory gear API 677 (1997)Accessory drive gear API 671 (1998)Flexible couplings NFPA 70 (2002)Electric code ASME PTC-22 (1997)Gas turbine testing ASME PTC-10 (1997)Compressor testing ASME B133Gas turbines API RP 686 (1996)Machinery installation
Some other international codes that are also applicable are:
ISO codes (ISO 3977, 1995-2002Gas turbines) IEC/CENELEC (electrical and fire systems) Local government codes European Union environmental and health compliance User specific specifications
UpstreamMost companies follow API specifications. MidstreamSome companies utilize API, but interpretation is generally handled with more flexibility. DownstreamAPI is a critical specification for most downstream process applications.
Of the above codes, API 616 (1998) and 617 (2002) are the most critical codes when evaluating or purchasing a turbocompressor package. However, API 614 (1999) and 670 (2000) are strongly referenced in both API 616 (1998) and 617 (2002) and should, thus, also be critically reviewed when purchasing a turbocompressor.
Turbocompressors and ancillaries/auxiliaries Surge recycle loop with surge control valves Gas coolers (intermediate, discharge) Suction scrubbers, discharge scrubbers Station isolation and bypass valves Suction and discharge valves Compressor loading and bypass valves Unit and station control room Fuel gas heaters Fuel gas filter and control skid SCADA system Fire fighting water system
WHAT IS API?
The American Petroleum Institute is the primary trade organization for the U.S. petroleum industry. API has over 400 member companies that cover all aspects of the oil and gas production. API is an accredited American National Standards Organization (ANSI) and started developing industry specific codes in 1924. Currently, API publishes about 500 standards, which are widely referenced by the Environmental Protection Agency ( EPA), Occupational Safety & Health Administration (OSHA), Bureau of Land Management (BLM), American Society of Mechanical Engineers (ASME), and other codes and
regulations. APIs philosophy for developing codes is based on the following principals:
Fill out as much info as is availableeven a partially filled out sheet is better than no sheet.
API 616GAS TURBINE
The following section describes a brief review of some critical points of API 616 (1998) with engineering opinions, interpretations, and comments (in italics). Some common manufacturer exceptions are also discussed. This discussion (in italics) presents only an engineering opinion of the authors, and the opinions are clearly debatable and open to other interpretations. A careful review of API specification applicability should be performed for every gas turbine purchase based on the specifics of the application. The reader should note that API 616 (1998) does not apply to aeroderivative gas turbines (i.e., it is intended for industrial type gas turbines only). Section 1.0Alternative Designs Section 1.0 reemphasizes that alternative designs are allowed. This is also briefly discussed in the API 616 (1998) Foreword. It is important to note that when alternative designs are proposed, the manufacturer must explicitly state the deviation and explain why the alternative design is superior to the API standard. These explanations should focus on safety, reliability, and efficiency of the equipment, while cost is generally not an acceptable justification for an API deviation. Section 2.0References All reference standards included in this section are automatically included in the standard. As this is difficult to review and be accountable for all referenced specifications, all manufacturers of machinery take general exception to this requirement. Section 3.0Definitions Section 3.0 provides a valuable discussion of common nomenclature and basic definitions. Some of the more critical items in this section are:
Improve safety Improve environmental performance Reduce engineering costs Improve equipment interchangeability Improve product quality Lower equipment cost Allow for exceptions within reason
However, as with most other codes, API specifications often lag technology developments, especially in the rapidly changing gas turbine and compressor markets.
API STANDARDS AS PURCHASING SPECIFICATIONS

API standards are often used as a convenient purchasing document. They provide the means for a customer to normalize the quotations by forcing all machinery vendors to quote on similar scopes. API codes also provide a common language and set of rules between vendor and customer to limit misunderstandings (e.g., definitions of efficiencies, test procedures, vendor data, data sheets). All API codes clearly state in their foreword that exceptions are allowed, if they lead to an improved or safer technical offer. For gas turbine applications in the oil and gas sector, API 616 (1998) is the foundation for almost all purchase specifications. API 617 (2002) is used for most centrifugal compressor applications. Also, API references National Fire Protection Agency (NFPA) NFPA 70 (2002) electric code for hazardous locations. This specification is fundamental in most machinery standards and is, thus, a critical requirement for oil and gas applications. Namely, oil and gas turbocompressors must at least meet Class 1, Division 2 (Zone 2), Group D. Some Applications require Division 1 (Zone 1).
API STANDARD TOPICS

The API 616 (1998) and 617 (2002) codes can be divided by a set of standard topics. These are:
Definitions
ISO rating, normal operating point, maximum continuous speed, trip speed, etc. Mechanical integrity Blade natural frequencies, critical speeds vibration levels, balancing requirements, alarms and shutdowns Design requirements and features Materials, welding, accessories, controls, instrumentation, inlet/exhaust systems, fuel systems Inspection, testing, and preparation for shipment Minimum testing, inspection, and certification Documentation and drawing requirements Often overlooked, but critically important are the API data sheets as they clearly form the technical basis of a typical proposal. Within these data sheets, the application specific issues are addressed, such as customer site, operating conditions, basic equipment selection, and equipment minimum integrity requirements. Thus, for a purchaser of a turbocompressor it is important to always fill out (as a minimum) the data sheets for API 616 (1998, gas turbine), API 617 (2002, compressor), API 614 Appendix D (1999, lube oil system), and API 670 Appendix A (2000, machinery protection). When filling out these data sheets a couple of industry accepted norms should be remembered:
Cross-out requirements that are not required or cannot be complied with. Include notes for critical technical comments. By including them on the data sheets these comments are elevated to a contractual technical requirement.
3.17ISO conditions are defined as T = 15C, P = 1.0133 bar, RH = 60 percent. 3.19Maximum allowable speed: speed at which unit can safely operate continuously per manufacturer 3.22Maximum continuous speed: 105 percent highest design speed 3.26Normal operating point: Usually the performance guarantee (heat rate, power) point for a gas turbine. This is defined by speed, fuel, and site conditions. 3.38Rated speed: power turbine speed at which site rated power is achieved 3.42Site rated conditions: The worst site condition at which still site rated power can be achieved. This must be provided by the user as a requirement. 3.45Site rated power: The maximum power achievable at site rated conditions. This must be provided by the manufacturer. Note: Unless specifically stated by purchaser, this is not a guarantee point. Thus, it is recommended that the user explicitly states that this must be a guarantee point. 3.50Turbine trip speed: speed where controls shut unit down from the fuel gas supply to the gas turbine 3.51Unit responsibility: Prime package contractor that has contractual responsibility to the purchaser. Either GT or compressor manufacturer can be prime, but in more complex projects the compressor vendor is usually selected prime (gas turbine is off-the-shelfcompressor is customized). As the compressor is usually only about 30 percent of the cost of the turbocompressor package, this arrangement often leads to commercial complexities and project financing requirements for the compressor vendor.
Section 4.0Basic Design This section in API 616 (1998) covers the basic design of the gas turbine itself. Included in this section are pressure casing requirements, combustor design, casing connections, shafts, bearings, seals, dynamic requirements, and material quality standards. This section is very comprehensive. However, as most gas turbine manufacturers are generally unwilling to make customized changes to their core engine, this is also the section in which most critical exceptions can be expected. A number of relevant items of this section are discussed below: of uninterrupted service. Hot section inspection must be performed every 8000 hours. Many gas turbines on the market cannot reach three years of uninterrupted service and/or require combustion inspections at 4000 hours. However, as this is a design requirement, vendor interpretations of this vary widely. 4.1.2The gas turbine vendor has unit responsibility unless otherwise specified. 4.1.5Speed range operating requirements are: Two-shaft gas turbines 50 to 105 percent rated speed, single-shaft gas turbines 80 to 105 percent rated speed. This refers to the output shaft only. 4.1.14On-skid electric must meet NEC NFPA 70 (2002). This is a critical requirement that generally applies throughout all oil and gas plants and has many significant packaging implications, which will be discussed later. 4.1.16Most gas turbine internals shall be exchangeable at site. This requirement disqualifies most aeroderivative gas turbines as they are usually overhauled by core-engine exchange only. 4.1.17Unit shall meet performance acceptance criteria, both in the factory and at site. Most manufacturers take exception to site performance guarantee unless a field test is specified and performed. 4.1.18Vendor shall review customers installation drawings. However, this does not imply that vendor certifies or warrants the customers installation design. 4.1.21Gas turbine must meet site rated power with no negative tolerances. This does allow for additional test uncertainties. 4.2.1Casing hoop stresses must meet ASME Section VIII (2004). A standard requirement throughout API. This defines hydro test procedures. 4.2.7Openings for borescope inspection must be provided for entire rotor without disassembly. Borescope inspection of the first compressor stages can be through the inlet and power turbine through the exhaust. Few manufacturers provide borescope points that allow access to all compressor stages. 4.2.9Field balancing (if required) without removal of casing is acceptable. 4.3Combustor design: As this section aims to apply to all types of combustors (can, can-annular, annular, dry low emission [DLE], diffusion) it is somewhat general. 4.3.1Combustors must have two igniters or igniters and crossfire tubes (in each combustor for multiple cans). This requirement is not clear for annular combustors and most can design combustors have a single ignitor in each can but do have cross-fire tubes. 4.3.2A temperature sensor is required in each combustor. This is not feasible in most gas turbine designs, as a thermocouple will not survive extended periods of operation at gas turbine firing temperatures. Most manufacturers provide T4 (gas generator turbine outlet) for two-shaft engines and/or T5 (exhaust temperature) for single-shaft engines. Firing temperature (T3) is usually calculated and displayed on HMI. 4.3.7Fuel Wobbe Index range must be indicated in the proposal. Most small gas turbines can handle 10 percent variation, even with DLE combustion. 4.4Casing connections: Section outlines good design practices. Most manufacturers have few exceptions in this section or have reasonable justifications in their deviations.
4.1.1Equipment must be designed for 20 years and three years
4.5.1.2Shafts shall be single piece heat-treated steel. Stacked rotors with a tie-bolt are not allowed. This is difficult to meet for most manufacturers and is not critical if a stacked rotor has been operationally proven and meets API dynamic requirements. Stacked rotors also have significant advantages for repair and maintenance. 4.5.2.1Rotor shall be designed for overspeed up to 110 percent of trip speed. This does not imply that the rotor must be tested to this speed. 4.5.2.2All rotor components must withstand instantaneous loss of 100 percent shaft load. Namely, coupling failure should not lead to catastrophic power turbine failure. 4.6Seals: Renewable seals are required at all close clearance points. Some manufacturers utilize internal lip-seals. This requirement makes the most sense for interstage seals. 4.7Rotordynamics: This extensive section describes minimum rotordynamic requirements and basic operational definitions. 4.7.1Critical speeds: When the frequency of a periodic forcing phenomenon corresponds to a natural frequency of that system, the system may be in a state of resonance. When the amplification factor is greater than or equal to 2.5, that frequency is called a critical speed. Most gas turbines operate above their first critical speed. API provides required separation margin between the operating speed range and the critical speeds. The most common source of excitation is unbalance but other sources include rubs, blade pass, gear mesh, and acoustic. The flexibility and potential resonances of the supporting structure should be considered, especially for three-point mount baseplates. 4.7.2Lateral analysis: This section recognizes that gas turbines are designed as standard products. Once the design is established, no modifications should be required on a particular application. An exception to this is if the drive coupling mass is significantly different from the design coupling. A new lateral analysis is only required on new prototype engines. 4.7.3Torsional analysis: Torsional analysis is required, but many manufacturers take exception to this for gas turbine drives. API requires that intersections between the torsional natural frequencies and one-times (1) the running speed must be separated by at least 10 percent. API also states that intersections at two and higher orders of running speed shall preferably be avoided, but most manufacturers take exception to this. 4.7.4Vibration and balancing: A progressive balancing procedure where no more than two components are added to the rotor at a time between corrections is required by API. However, since most gas turbines rotors are built-up designs with tie-bolts, this procedure is not possible. For these built-up rotors, component balancing should be performed and careful attention to rotor-runout should be made after assembly. Blade mass sorting routines help to minimize the amount of correction required. Sufficient speed during low speed balance should be made to ensure that the blades are seated in the roots or fur-trees. During assembly balance, one should distribute the correction planes to several locations along the rotor, since the exact location of unbalance is not known. A residual unbalance check is encouraged by API, but given the resolution of modern balance machines, this check is not required. High speed balancing of the entire rotor is acceptable but is discouraged. Clearly, rotors that require high speed balancing to achieve vibration limits during test will likely require field balancing. The required API balance limits are reasonable and should be adhered to on both the component and assembly level. Care should be made to minimize electromechanical runout by the proximity probes. API permits subtraction of the run-vector from the vibration level at running speed, but few customers accept this practice. 4.8Bearings: applies to all gas turbine bearings. 4.8.1.1Hydrodynamic radial and thrust bearings are preferred. These should be thrust-tilt pad, radial-tilt pad, or sleeve bearings.
4.8.2.5If rolling element bearings are used they must meet
50,000 hours of continuous operation. Few industrial gas turbines utilize rolling element bearings and aeroderivative engines are not applicable to API 616 (1998). 4.8.3.3The bearing shells shall be horizontally split. Many original equipment manufacturers (OEMs) take exception to this requirement. 4.8.4.2Hydrodynamic thrust bearings shall be selected at no more than 50 percent of the ultimate load rating at site power. This requirement should not be taken exception to. 4.8.5.2Bearing housings: Replaceable labyrinth type buffer seal required. Lip-seals are not acceptablethis cannot be met by some manufacturers. 4.8.5.3Two radial proximity probes in all radial bearings (X, Y), two axial proximity probes in the thrust bearing, and a key phasor should be provided (when space allows). Many manufacturers will only provide a single thrust bearing proximity probe. 4.9 Lubrication: specifies lubrication requirements and refers to API 614 (1999). 4.9.2Synthetic lube oil is acceptable in the lube oil system. However, most users prefer mineral oils for cost reasons. 4.9.5Integrated or separate lube GT/compressor oil systems are both acceptable. 4.9.7Lube oil system must meet API 614 (1999). Most modern gas turbines have the lube oil tank integrated into the skid but an overhead tank is also acceptable. 4.10Materials: This section describes basic material, casting, welding, and forging requirements and essentially states that good design rules should be followed. Low grade carbon steel is not permitted. All materials of construction must be stated in the vendor proposal (and most manufacturers do not comply with this). Purchaser may require 100 percent radiography, magnetic particle inspection, and/or liquid penetrant inspection on all welds. 4.11.3A stainless steel (SS) nameplate to include rated performance and conditions, critical speeds, maximum continuous speed, overspeed trip, and fuel type. Few manufacturers provide this relatively simple and valuable requirement adequately, unless the purchaser insists. Section 5.0Accessories Section 5 deals with all accessories that are required to operate a gas turbine (i.e., mostly package items, instrumentation, controls, and inlet/exhaust systems). Section 5 also includes critical sections on gas and liquid fuel systems. Some of the main points of this section are:
5.2.2Vendor to supply all couplings and guards for load and
5.1Starting and helper drivers: Starter motors decouple from the turbine shaft after startup while helper motors stay connected. 5.1.1.1Electric, pneumatic, electrohydraulic, gas engine, and steam turbine are all acceptable starter/helper motors. Most modern gas turbines utilize AC starter motor with variable frequency drives (VFD). 5.1.1.7Starter motors must be supplied with all necessary gears, couplings, and clutches. 5.1.2.1Starters shall be rated 110 percent of starting torque required. Vendor to supply torque curves. Most AC starter motors are severely underrated for the application and overheat after multiple starts. 5.1.3Turning gear: Device to slowly rotate to avoid shaft deformation. Must be separate device from starter motor. This is generally not required on smaller turbines, and, if required, most manufacturers utilize the starter motor. 5.2.1Vendor to supply accessory gear for starting, auxiliary equipment (e.g., lubricating oil pump). Vendor to also supply separate load gear, if required. All gears shall meet API 613 (2003). Accessory gears shall be rated for 110 percent of required load. Most gas turbine lube oil systems utilize engine driven main pump, undersized AC pre-post pump, and undersized direct current (DC) backup pump.
accessory shafts. Couplings to be sized for maximum continuous torque and meet API 671 (1998). 5.3Gas turbine to be supplied on single mounting plate (baseplate). This is typically interpreted to allow for skid mounting. Skid to be designed to limit worst-case shaft alignment change to 50 micrometers. For some offshore applications this is difficult to meet without gimbled three-point mount. Supporting the baseplate at three pounds can result in baseplate vibration modes in the operating speed range of the GT or driven equipment. This foundation flexibility may also adversely affect the critical speeds. Careful finite element analysis should be performed, although it is not required by API 616 (1998). 5.3.2A single piece baseplate is preferred. The baseplate shall extend under driven equipment. As the compressor and gas turbine are often supplied by different vendors, this requirement is often not practical; tightly bolted gas turbine and compressor skids are common and generally function well when properly engineered. 5.3.2.4Skid to permit field leveling. 5.4.1.2All controls and instrumentation shall be suitable for outdoor installation. This is usually not necessary or practical for inside enclosure devices and devices that are in a control room. 5.4.1.3Gas turbine instruments and controls shall meet API 670 (2000). 5.4.1.4GT control system must provide for safe startup, operation, and shutdown of gas turbine. There is an implicit guarantee in this statement that the gas turbine must function/operate properly. 5.4.1.6Unit shall continue to operate for specified time after AC failure. This is only possible with redundant control system and not practical for some applications as safe relay shutdown on backup batteries is preferred (i.e., if the unit operation is not critically important, it is often safer from a system perspective to shut down on an AC failure than to continue operating). 5.4.2.1Automatic start to require only single operator action. Although most modern control systems have a single button start-stop operation, in reality the operator has to usually acknowledge and reset a host of alarms during the startup process. 5.4.2.2Control system to completely purge unit prior to startup. This is a critical safety requirement to avoid explosions in a restart attempt. A purge crank of five to 10 minutes is typical. 5.4.3.4Load control to limit driven equipment speed to 105 percent rated speed. 5.4.4.1Alarm/shutdown system to be provided to protect unit and operator. Both normal and emergency (fast) shutdown (i.e., cut off fuel supply) must be available. 5.4.4.3Fuel control must include separate shutoff and vent valves, separate from fuel control valve, with local and remote trip. A control valve that also acts as a shutdown valve is not permitted and should not be accepted by purchasers for safety reasons. 5.4.4.8.1Alarm and shutdown switches shall be separate. This is often not practical for gas turbine internal sensors (e.g., flame detector, axial thrust position). 5.4.4.8.3Alarm and trip settings should not adjustable from outside housing. This is not relevant with modern control systems. Most alarm and shutdown levels are set from the unit control system. 5.4.4.9All instruments (other than shutdown switches) shall be replaceable without shutting unit down. This is often not practical on alarm/shutdown switches from a single sensor (refer to 5.4.4.8.1). 5.4.5.1.1Off-skid or on-skid control system to be supplied with unit. As most oil and gas facilities have a control room available that does not have NFPA 70 (2002) hazardous area requirements, it is more common to utilize off-skid control system. 5.4.5.1.1Any on-skid control system must meet hazardous area requirements (NFPA 70, 2002). 5.4.6.2On-skid electric systems must meet NFPA 70 (2002).
5.4.7Instrumentation: The API 616 (1998) data sheets must be
filled out by purchaser to indicate the required instrumentation. Manufacturers will often cross out lines to indicate noncompliance. The vendor must include in his proposal a full list of instrumentation. API 670 (2000) governs basic requirements for temperature and vibration sensors. 5.5.3.1Air inlet and exhaust system including inlet filter, inlet/exhaust silencers, ducting, and joints must be provided by vendor. As this is application specific, some vendors will take exception to this. 5.5.3.3Inlet system shall be designed for maximum 4 inch H2O pressure drop at site rated power. In very humid or moist environments, it is difficult to meet this requirement with wet barrier filters, especially during startup. 5.5.3.4Inlet/exhaust system design life is 20 years. For marine environments this requirement implies the use of SS, which is very expensive. 5.4.2.2Bolts, rivets, or fasteners are not permitted in inlet system. 5.5.3.9A gas turbine compressor cleaning system must be provided. This is quoted as an option by most manufacturers. 5.5.4.4Inlet filter system requires walkways and handrails. This is not necessary for very small gas turbine units. 5.5.5Inlet/exhaust silencer: Internals to be SS. External can be carbon steel. 5.7.3Unit must be supplied with gas detection system, fire detection system, and fire suppression system to meet NFPA. Fire suppression shall be automatic based on thermal detection. Typical systems are: Thermal, thermal gradient, lower explosive limit (LEL), and flash detector. CO2 and water mist are typically utilized for suppression. Water is safer for personnel but also requires more maintenance. 5.7.5When enclosure is supplied it must be weatherproof and include fire detection/suppression, ventilation/purging, lights, and doors and windows. 5.8.1.2Gas fuel system: Must include strainer, instrumentation, manifolds, nozzles, control valve, shutoff valve, pressure regulator, and vent valve. A strainer is often not supplied. Some vendors provide a cartridge filter instead. If a strainer or filter is used, they should be supplied with a differential pressure sensor. 5.8.1.3Fuel gas piping must be stainless steel. This is critically important, especially in sour gas applications. 5.8.1.4Liquid fuel system: Must include pump, atomizing air, two shutoff valves, instrumentation, control valves, flow dividers, nozzles, and manifolds. Newer systems usually employ a variable speed pump instead of a positive displacement pump with a control valve. 5.81.5.2If dual fuel system is provided it must allow bumpless bidirectional transfer. All gas turbine vendors require a reduction in load during fuel transfer. 5.8.2.2.2Vendor must review customers fuel supply system. This does not imply a vendor certification of guarantee. 5.8.2.4Heating value of fuel cannot vary by more than 10 percent. This often protects the manufacturer against warranty claims. 5.8.4.2Purchaser must specify required site emissions levels for NOx, CO, UHC, et al., to manufacturer (i.e., it is the purchasers responsibility to assure that the gas turbines meet environmental codes. The vendor only confirms [or not] the required emissions values). Section 6.0Inspection, Testing, and Preparation for Shipment In this section basic inspection, testing and shipping requirements are defined. Some important issues in this section are:
full speed (maximum continuous speed) test to verify that the complete gas turbine package (including all auxiliaries except for inlet/exhaust system) operates within vibration and operational control limits. Contract coupling should be used. Rotordynamic signature and vibrations must be recorded. This is a basic mechanical integrity test, and compliance with these requirements is critical. 6.3.4Optional tests: These tests are not optional if purchaser marks them in the API 616 (1998) data sheets. 6.3.4.1Optional performance test: unit full load tested to ASME PTC 22 (1997) (discussed later). 6.3.4.2Optional complete unit test: Similar to mechanical run test but to include driven equipment. This test is often combined with a performance test and is called a full load string test. 6.3.4.3Gear test: Gear must be tested with unit during mechanical run test. 6.3.4Other optional test: sound level, auxiliary equipment, fire protection, control response, spare parts. 6.3.4.6If unit fails mechanical run test, complete disassembly and reassembly are required. 6.4Preparation for shipment: Short-term shipping is just a plastic wrapping or tarp. Long-term shipping requires crating, surface anticorrosion treatment, inert gas (nitrogen) fill of all vessels including GT casing. Generally, long-term shipping preparation is recommended as there are often delays in the commissioning of a new plant, which may lead to extended storage times of the equipment. Section 7.0Vendors Data In this section the minimum documentation requirements from the proposal to as-built drawings are defined. The detailed document requirements are listed in the API 616 (1998) data sheets. Most manufacturers take exception to specific document requirements by crossing out lines in the API data sheets. As a minimum, the purchaser should insist on performance maps, performance calculations, mechanical, hydraulic, layout, and electrical drawings, process and instruments diagram (P&IDs), test and inspection results, as-built parts lists, operation and maintenance manuals, installation manual, and technical data.
API 614HIGHLIGHTS
API 616 (1998) refers to API 614 (1999) for the lubricating oil system. API 614 (1999) is a generic specification for all lubricating oil (LO) systems, but a couple of important requirements and issues should be emphasized:
6.3.2.3 Hydrostatic tests: Vessels and piping must be per ASME Section VIII (2004). 6.3.3Mechanical running test: A required four-hour no load,
LO system must be design for 20 years service life. This implies all SS for marine applications. A single full capacity pump with a full size backup pump is required. Most GT manufacturers provide smaller sized backup pumps that only allow for a safe shutdown but not continued operation. Duplex oil filters with smooth transfer are required (i.e., no shutdown should be required when replacing filter cartridges). All wetted surfaces must be stainless steel. Most manufacturers will provide a coated carbon steel LO tank integrated into the skid. Lube oil tank must have minimum of eight minutes retention time. Manufacturers with integrated skid LO tanks generally cannot meet this requirement, especially if a combined compressor LO system with wet seals is utilized. All pumps, valves, switches, and sensors to have block and bleed valves. Sight indicators and levels sensors for are required for tanks. A skid integrated lube oil tank is acceptable; an overhead gravity feed tank is preferred. This provides continues flow in case of an electrical failure and sometimes eliminates the need for a DC backup pump. Tank must have vent with flame trap, and an exhaust fan is preferred to avoid the flammable vapor gas mixture accumulation.
API 670 HIGHLIGHTS

API 670 (2000) covers machinery protection and control system such as alarm and shutdown switches. It is beyond the scope of this paper to cover API 670 (2000) but a couple of important items are listed below:
Divisions Within Class 1
API 670 (2000) provides minimum design, installation, and accuracy standards for package alarms/shutdown switches and sensors. It covers all instruments specified in the API 616 (1998) data sheets. All instrumentation and wiring must meet NFPA 70 (2002) hazardous area classifications. Signal and power wiring must be separate. Alarms and shutdown switches must be separate and separate stainless steel housings. This is often not practical with modern sensors and control systems. Bearings must have metal temperature sensors. Most manufacturers only offer this as an optional feature. Axial thrust bearings must have two proximity probes. Only one proximity probe is usually provided.
FACTORY PERFORMANCE TESTS
There are a number of tests that are performed on a gas turbine driven compressor in the factory. The gas turbine performance test specification most commonly used is ASME Performance Test Code 22 (ASME PTC-22, 1997) or manufacturer specific derivatives of it. For the driven compressor, API requires only a mechanical test run, but most operators insist on a closed loop test also to characterize the performance. Closed loop tests are usually performed per ASME PTC-10 (1997). PTC-10 Compressor
Class 1, Division 1 (Zone 0 and Zone 1)Ignitable concentration of flammable mixtures exist most of the time. Class 1, Division 2 (Zone 2)Ignitable concentrations of flammable mixtures are not likely to exist under normal conditions.
Most oil and gas applications require Class 1, Division 2. Some offshore and refinery applications require Class 1, Division 1. NFPA 70 (2002) applies principally to the following turbocompressor package items:
On-skid electrical
Wiring and cables Electric motors (pumps, fans, starter) Instruments and alarms Junction boxes Lights and heaters (if applicable) On-skid fire systems Ultraviolet/infrared (UV/IR) flash sensors Flammable mixture detectors Temperature, temperature rise detectors Smoke detectors (if used) CO2 (or other) fire fighting system Air inlet and lube oil electrical elements NFPA 70 (2002) does not apply to nonelectrical components and exception to NFPA 70 (2002) can often be taken for:
Closed loop test to determine performance of compressor Can also identify aerodynamic stability issues if tested to full load Type I: actual gas and full pressure; Type II: simulated gas and reduced pressure Type I typically used for high pressure/high energy applications Reduces risk of startup delays due to vibrations or lack of
performance PTC-22 Gas Turbine
Off-skid control system Battery charger and batteries Starter motor variable frequency drive Synchronization panel
There are many requirements in NFPA 70 (2002) for Class 1, Divisions 1 and 2 (details are in NFPA 70, 2002, Sections 500 and 501), but some of the most critical requirements are:
and efficiency
Full speed, full load test for four hours Typically against a water break or generator/load cells Determines maximum output power, specific fuel consumption,
Grounded explosion-proof conduit for all wiring Conduit and conduit connections are copper free Gas sealed conduit connections Instruments, connectors, terminals in explosion-proof boxes Motors rated explosion-proof Fire detection system (UV/IR, thermal gradient, flash detectors) Gas (flammable mixture) detectors (LEL) Fire fighting system (inert gas or water mist)
API 617Axial and Centrifugal Compressors and ExpanderCompressors for Petroleum, Chemical, and Gas Industry Services API 617 (2002) is a code that is applied to the centrifugal compressor used in GT packages. This specification is prevalent for offshore and refinery applications. It is usually less prevalent in pipeline applications. API 617 (2002) is often referenced within user-company specific purchase specifications, but all manufacturers have various amounts of comments and exceptions. API 617 (2002) has similar technical definitions as API 616 (1998), and these are, therefore, not repeated here. However, in general API 617 (2002) covers the following equipment:
Some operators opt to perform field testing instead of extensive factory testing. The risk of this test is that if problems are identified in the field it may be difficult to easily correct them. However, if testing is intended to only verify a very tight performance guarantee, field testing is typically adequate.
NATIONAL FIRE PROTECTION ASSOCIATION (NFPA) 70

API specifications refer to the NFPA 70 (2002) code for electrical wiring and safety requirement. To identify the proper classification of a subject, NFPA 70 (2002) provides the following guidelines. Hazardous Location Summary Electrical Requirements
Class 1Flammable gases, vapors, or liquids Class 2Dust and combustible dust that can form explosive mixtures Class 3Fibers or flyings suspended in air that are easily
ignitable
Centrifugal and axial compressors Integrally geared compressors Expander compressors

SCOPE OF API 617
API 617 (2002) covers the minimum requirements for axial and centrifugal compressors, single-shaft and integrally geared process
centrifugal compressors and expander-compressors for use in the petroleum, chemical, and gas industry services that handle air or gas. This standard does not apply to fans (covered by API 673, 2002) or blowers that develop less than 34 kPa (5 psi) pressure rise above atmospheric pressure. This standard also does not apply to packaged, integrally-geared centrifugal plant, and instrument air compressors, which are covered by API 672 (1996). Furthermore, hot gas expanders over 300C (570F) are not covered in this standard. As with all other API codes, the equipment vendor may offer alternative designs if these designs improve the safety or performance of the equipment. Otherwise all designs should comply with this standard. If exceptions to the standard are taken, they must be clearly stated in the proposal.
API Standard 670, 2000, Vibration, Axial-Position, and BearingTemperature Monitoring Systems, Fourth Edition, American Petroleum Institute, Washington, D.C. API Standard 671, 1998, Special Purpose Couplings for Petroleum, Chemical, and Gas Industry Services, Third Edition, American Petroleum Institute, Washington, D.C. API Standard 672, 1996, Packaged, Integrally Geared Centrifugal Air Compressors for Petroleum, Chemical, and Gas Industry Services, Third Edition, American Petroleum Institute, Washington, D.C. API Standard 673, 2002, Centrifugal Fans for Petroleum, Chemical, and Gas Industry Services, Second Edition, American Petroleum Institute, Washington, D.C. API Standard 677, 1997, General-Purpose Gear Units for Petroleum, Chemical, and Gas Industry Services, Second Edition, American Petroleum Institute, Washington, D.C. API Standard 686, 1996, Machinery Installation and Installation Design, American Petroleum Institute, Washington, D.C. ASME B133, A Series of Standards for Gas Turbines, American Society of Mechanical Engineers, New York, New York. ASME PTC-10, 1997, Performance Test Code on Compressors and Exhausters, American Society of Mechanical Engineers, New York, New York. ASME PTC-22, 1997, Gas Turbine Power Plants, American Society of Mechanical Engineers, New York, New York. ASME Boiler & Pressure Vessel Code - Section VIII - Pressure Vessels, 2004, American Society of Mechanical Engineers, New York, New York. ISO 3977, 1995-2002, Gas TurbinesProcurement, A Series of Nine Amendments, International Organization for Standardization, Geneva, Switzerland. NFPA 70, 2002, National Electrical Code, National Fire Protection Agency, Quincy, Massachusetts.
SUMMARY
This tutorial provides an overview of the applicable codes related to typical industrial gas turbine packages, including API 616 and the supporting API Standards 614 (1999), 617 (2002), 670 (2000), 671 (1998), and 677 (1997), as well as NFPA 70 (2002). As the authors have presented, one size does not fit all in the selection and procurement of gas turbines and many exceptions to API are made. However, the API standards represent sound engineering practice based on many years of experience. Therefore, exceptions to these standards should be kept to a reasonable minimum.
REFERENCES
API Standard 613, 2003, Special-Purpose Gear Units for Petroleum, Chemical and Gas Industry Services, Fifth Edition, American Petroleum Institute, Washington, D.C. API Standard 614, 1999, Lubrication Shaft-Sealing and ControlOil Systems for Special-Purpose Applications, Fourth Edition, American Petroleum Institute, Washington, D.C. API Standard 616, 1998, Gas Turbines for Refinery Service, Fourth Edition, American Petroleum Institute, Washington, D.C. API Standard 617, 2002, Axial and Centrifugal Compressors and Expander-Compressors for Petroleum, Chemical and Gas Industry Services, Seventh Edition, American Petroleum Institute, Washington, D.C.
Table of Contents
SURVEYING TILTING PAD JOURNAL BEARING AND GAS LABYRINTH SEAL COEFFICIENTS AND THEIR EFFECT ON ROTOR STABILITY
by John A. Kocur, Jr.
Machinery Engineer Plant Engineering Division ExxonMobil Research & Engineering Fairfax, Virginia
John C. Nicholas
Owner and President Rotating Machinery Technology, Inc. Wellsville, New York
and Chester C. Lee

Manager, Rotordynamics Design and Test Gas Compressor Engineering Department Solar Turbines, Inc. San Diego, California from the University of Virginia (1977) in rotor and bearing dynamics. Dr. Nicholas holds several patents including one for a spray-bar blocker design for tilting pad journal bearings and another concerning bypass cooling technology for journal and thrust bearings. Chester C. Lee is the Group Manager of Rotordynamics Design and Test in the Gas Compressor Engineering Department of Solar Turbines Inc., in San Diego, California. He has been with this group since he joined Solar 17 years ago. His major responsibility is to support rotordynamic analysis in design, manufacturing, testing, and field operation. Dr. Lee received his Ph. D. degree (Mechanical Engineering) from the University of Virginia, specializing in rotating machinery. Before joining Solar, he worked for Mechanical Technology Inc., in Latham, New York, on various rotating equipment.
John A. Kocur, Jr., is a Machinery Engineer in the Plant Engineering Division at ExxonMobil Research & Engineering, in Fairfax, Virginia. He has worked in the turbomachinery field for 20 years. He provides support to the downstream business within ExxonMobil with expertise on vibrations, rotor/aerodynamics, and health monitoring of rotating equipment. Prior to joining EMRE, he held the position of Manager of Product Engineering and Testing at Siemens Demag Delaval Turbomachinery. There Dr. Kocur directed the development, research, engineering, and testing of compressor and steam turbine product lines. Dr. Kocur received his BSME (1978), MSME (1982), and Ph.D. (1991) from the University of Virginia and an MBA (1981) from Tulane University. He has authored papers on rotor instability and bearing dynamics, lectured on hydrostatic bearings, has been a committee chairman for NASA Lewis, and is a member of ASME. Currently, he holds postions of API 617 Vice-Chair and API 684 Co-Chair. John C. Nicholas is owner and President of Rotating Machinery Technology, Incorporated, in Wellsville, New York. His company repairs and services turbomachinery, and manufactures bearings and seals. Dr. Nicholas has worked in the turbomachinery industry for 30 years in the rotor and bearing dynamics areas, including five years at Ingersoll-Rand and five years as the Supervisor of the Rotordynamics Group at the Steam Turbine Division of Dresser-Rand. Dr. Nicholas, a member of ASME, STLE, and the Vibration Institute, has authored over 40 technical papers concerning rotordynamics and tilting pad journal bearing design and application. He received his B.S. degree from the University of Pittsburgh (Mechanical Engineering, 1968) and his Ph.D. degree
1
ABSTRACT
The evaluation of rotor system stability, which has become an essential part of rotordynamic analyses and rotating machinery design, relies heavily on the bearing and seal dynamic coefficient modeling to obtain an accurate prediction of the turbomachinery behavior. Lacking experimental validation, analytical predictions can be widely varied and even divergent as more complex procedures and models are created. To measure the variability of bearing and labyrinth seal coefficient predictions, a survey of 60 turbomachinery users, manufacturers, consultants, and academicians was conducted under the auspices of the American Petroleum Institute (API). Coefficients received from the respondents were incorporated into a common rotordynamic model to determine the impact on the predicted rotor stability. In addition, several of the most popular analytical codes for the prediction of tilt pad journal bearings are compared. Starting with an iso-viscous prediction and proceeding through more complicated thermal and structural deformation solutions, the authors compare the variability and divergent nature of these codes. The measured variability of the data collected clearly illustrates the need for the resolution
Table of Contents
2 PROCEEDINGS OF THE THIRTY-SIXTH TURBOMACHINERY SYMPOSIUM 2007
of fundamental bearing issues (i.e., synchronously versus nonsynchronously reduced bearing coefficients) and labyrinth seal predictions based on repeatable experimental data.
INTRODUCTION
Centrifugal compressor instability remains a major concern causing reduced unit availability and project commissioning delays, leading to lost revenue for both users and vendors of the equipment. The evaluation of rotor stability has become an essential part of rotordynamic analyses and rotating machinery design. In the latest edition of American Petroleum Institute (API) 617, Seventh Edition (2003), specifications for performing stability analyses of centrifugal compressors were added for the first time. As a continued effort to improve API specifications, a study of the current state of bearing and labyrinth seal dynamic coefficient predictions is undertaken. Tilting pad bearing analysis started over 40 years ago spawned by work by Lund (1964). API 684 (2005), Section 2.5.4, and Nicholas (2003) provide an excellent historical perspective of the development of tilt pad bearing analysis. While work has advanced to include flexibility of the pad and pivot, nonsynchronous behavior, thermal effects of the fluid, and pad deformation, surprisingly little experimental work has been published on the dynamic behavior of tilting pad bearings. A series of papers was published in the 1990s on the measurement of coefficients concluding with Wygant, et al. (1999). However, bearing operating speeds were not representative of the high speed compressor applications that typically suffer stability problems. In addition, the use of synchronous versus nonsynchronous coefficients in stability analysis was not addressed. Cloud (2006) has investigated this argument from a rotor system standpoint. While the published research provides valuable insight into the appropriate coefficients to use in a stability analysis, it still relies on the assumed accuracy of the bearing prediction methods employed. Gas labyrinth seal research seems to follow a different path. While significant efforts to understand the dynamic behavior of these seals has been undertaken (Iwatsubo, et al., 1982; Childs and Scharrer, 1986; Kirk, 1990), the validity of these approaches is often called into question by comparison to experimental work, summarized by Childs (1993). Poor matching of the data produced by analytical methods has raised questions regarding continued modeling efforts of labyrinth seals from the single volume approach of Childs and Scharrer (1986), to the three volume approach of Nordmann and Weiser (1990). Current experimental efforts seem to indicate the more complicated analytic models are not providing better results. For the most part, these discussions focus on the tangential coefficients of the labyrinth seal. Radial coefficient predictions suffer such a wide variation to experimental data as to be ignored in their use. Testing of short labyrinth seals has also proved to be a challenge. Short seals with smaller pressure drops produce small forces that have resulted in high uncertainty levels during testing. While the individual impeller eye seal may not provide the same force magnitude compared to the balance piston (BP), the multiple stages usually result in a total force approaching the balance piston. In addition, the importance of data taken near application conditions was emphasized by Childs and Ramsey (1991), Elrod, et al. (1995), and Wagner and Steff (1996). The later, while extending test conditions, held much of the data proprietary. The main objective of this paper is to determine the magnitude of variations in tilting pad journal bearing and gas labyrinth seal dynamic coefficient predictions. This is approached in two fashions: first, as a survey sent to industry wide participants using a common data set; and second, by comparing coefficients obtained from several widely used tilting pad bearing analysis codes. The survey is intended to measure the existing variations in the industry when supplied with bearing and seal dimensions from
an unstable compressor. The code comparison effort attempts to identify the source of the tilting pad bearing coefficient variations as found in the survey. At the conclusion of this work, answers to the following questions are sought:
Is an API Level I type analysis still justified given the current

state of analysis technology?
Has sufficient experimental work been performed to act as the

basis for comparison of the existing analysis packages?
Do the available analysis codes give reasonable agreement when supplied with similar input? Why do different tiling pad journal bearing codes provide different bearing coefficients? Do
the different journal bearing codes converge to identical results after stripping out all of the parameters that cause the coefficient variations?
API SURVEY
A survey was conducted under the auspices of the American Petroleum Institute with the intent to determine the conformity level of bearing and labyrinth seal dynamic behavior predictions. Direct comparison of the coefficients and impact on rotor stability are the two methods used to determine the variability of the predicted coefficients. The working hypothesis of the survey can be stated as follows: To date, there remain significant differences across the industry in the prediction of dynamic coefficients for fluid film tilt pad bearings and labyrinth seals. The survey focuses on the prediction of tilt pad bearings and impeller eye and balance piston labyrinth seals as related to centrifugal compressor applications. Consequently, the goals of the survey are threefold:
First, to improve the recently introduced specifications in API 617 (2003) and 684 (2005) regarding rotordynamic stability. This can be realized by examining the necessity and/or appropriateness of the steps within the stability methodology.
highlight and communicate the disparity of the predictions within the industry. Hopefully this can be used as supporting evidence for the continued need of research funding from industry and government agencies.
Second,
Finally, determine the need for experimental data to act as the gold standard.
While some groups have started measuring tilt pad bearing coefficients, there remains a lack of verifiable data that can be used universally to determine the accuracy of analytical methods for either bearings or labyrinth seals. The survey was proposed and supported by API. Information relevant to the prediction of the tilt pad bearing and labyrinth seals was sent to 60 respondents directly from API. The respondents included industrial users and vendors of rotating equipment (both at a component and turbomachinery level), consultants, and educators. While the list of respondents was known (and developed) by the authors of this paper, the responses were kept anonymous by API. From the 60 requests for information sent, 16 respondents supplied bearing coefficients, 12 of which also sent seal coefficients, and several responded that they did not have the ability to calculate the requested information. Approximately one-third of the group replied in some fashion.
SURVEY INFORMATION
As preferred by the API Subcommittee on Mechanical Equipment (SOME), an actual compressor application was sought as the basis for the survey. The compressor used in the survey represents a multistage nitrogen compressor intentionally designed by the vendor to be unstable for stability testing. Extensive testing
Table of Contents
SURVEYING TILTING PAD JOURNAL BEARING AND GAS LABYRINTH SEAL COEFFICIENTS AND THEIR EFFECT ON ROTOR STABILITY 3
by the vendor determined the instability onset speed for several operating conditions. One such condition was selected for the survey representing the highest P across the compressor obtained during the testing program. To minimize efforts requested from the respondents and to eliminate modeling differences between respondents, a common rotor model and damped eigenvalue solution algorithm (Vzquez and Barrett, 1998; Vzquez, et al., 2001), was proposed for the study. By eliminating these differences, the variations in the predicted damped eigenvalues can be attributed to the supplied bearing and seal coefficients. Respondents were asked to supply bearing coefficients, the first and last impeller eye seal coefficients, and balance piston coefficients for a single operating point representing the onset of instability. Each was instructed to treat the information supplied as they would a new design compressor or one with a known stability problem. The appropriate level of analysis to use (including thermal analysis, pivot stiffness, etc.) was left to the respondent to determine. As noted, a single rotor model was employed throughout the study. Figures 1 and 2 display the mass (top) and stiffness (bottom) and 3D rotordynamic model of the survey compressor. The compressor is a five-stage tie-bolt design compressor with nitrogen as the working fluid. General information regarding the compressor and operating point in question can be found in Table 1. The operating point selected for the survey consists of an operating speed of 21,662 rpm with the compressor producing 209 bara at a pressure ratio of 2.53 in N2.
Table 2. Bearing Information, API Survey Compressor.
Labyrinth seals were described as tooth-on-rotor for all impeller eyes and the balance piston. Geometry dimensions included the tooth spacing, tooth tip width, tooth height, operating clearance, and diameter as shown in Figure 3. The impeller eye seals consisted of four tooth seals at a diameter of 133.35 mm (5.25 inches). The balance piston incorporated 11 teeth at a diameter of 127 mm (5.0 inches). The gas preswirl was also given due to the complexity of information needed to accurately compute this parameter. Preswirl was set at 70 percent for all seals. Only coefficients for the first and fifth (last) impeller eye seals were requested (also to minimize efforts). A simple linear variation was used to derive the eye seal coefficients for the middle three impellers.
Figure 3. Dimensions Supplied in Survey for the Impeller Eye and Balance Piston Labyrinth Seals. Gas conditions for each impeller eye seal were supplied in terms of suction and discharge pressure and temperature for the stage and for the compressor overall. Recognizing that some respondents chose methods similar to the modified Alfords method employed in the API 617, Seventh Edition (2003) Level I stability analysis, impeller diameter and discharge tip width and diffuser minimum width were also supplied in the survey. To further eliminate unintended variations, the gas properties of nitrogen were supplied at key operating points within the compressor. Figure 4 displays a typical set of data for nitrogen.
Figure 1. Mass and Stiffness Model of the Survey Compressor.
Figure 2. 3D Model of the Survey Compressor. Table 1. General Information, API Survey Compressor.
The tilt pad bearings consisted of five pads with the gravity load located between the pad pivots. The steel backed babbitted pad arc length was 60 degrees. Bearing clearance ratio was on the order of 1.5 mils per inch. Other details of the bearing can be found in Table 2. Survey information also included oil type, oil inlet temperature and flow, oil viscosity at referenced temperatures, and pivot stiffness.
Figure 4. Typical Nitrogen Gas Properties Supplied in Survey.
Table of Contents
Finally, an approximate location of the first critical speed was supplied at 6700 cpm. This was intended to be used by those respondents supplying bearing and seal coefficients for the whirl frequency rather than synchronous frequency. Currently, API states that synchronous reduction should be used in the Level I analysis for the bearing coefficients. It should be noted however that nearly half of the participants in that effort to develop the API stability specifications did use nonsynchronous reduction. No such statement is made for the calculation of labyrinth seal coefficients.
SURVEY COEFFICIENT RESULTS

Sixteen sets of bearing data were received from the respondents. The data were normalized using the minimum stiffness supplied for the principle stiffness in the X-direction. The same respondents principle damping in the X-direction was used to normalize the damping coefficients. Coefficients are presented on a graph plotting Kxx versus Kyy and Cxx versus Cyy. This display was selected to show variations in the coefficients due to loading. However, the survey compressors bearing was very lightly loaded producing almost identical coefficients in the horizontal and vertical directions. Figure 5 displays the variations in stiffness from the survey. Considering the effects of reduction, almost an order of magnitude of difference exists in the supplied coefficients. Using the information supplied by respondent #8, frequency reduced coefficients are assumed to occupy the higher range of the normalized coefficients. Figure 6 plots the variations in damping. As with the stiffness, the variation in damping from the smallest to largest is also nearly an order of magnitude. While using nonsynchronous reduction increases the bearing stiffness, damping is predicted to decrease with the reduction (Figure 6).
It should be noted that the approximate location of the first critical speed was erroneously supplied at 6700 cpm rather than the intended 9700 cpm. Unfortunately, this error was identified after the information was received from the respondents. While this information has no effect on the respondents using synchronous values for the bearing and labyrinth seal coefficients, it will add some unintended variation to those supplying frequency dependent coefficients. The frequency impact and reduction schemes are well known (Parsell, et al., 1983, and Barrett, et al., 1988). Examining the effect of this parameter on tilt pad bearings, a set of coefficients was predicted using a solution scheme developed by Branagan (1988). Comparing the frequency reduced coefficients using 6700 cpm, 9700 cpm, and 21,662 rpm, the change in the principle stiffness and damping coefficients are shown in Table 3. As noted, reducing the full coefficient matrix to the lower frequency, increases the stiffness and reduces the damping an additional 10 percent from the synchronously reduced coefficients when compared to the reduction to the intended frequency of 9700 cpm. Table 3. Predicted Impact of Frequency on Principle Bearing Coefficients.
Figure 7 illustrates the expected change in respondent #8s bearing coefficients if a frequency of 9700 cpm is used in the reduction process instead of 6700 cpm. The difference between synchronous and nonsynchronous reduction is decreased by 25 percent assuming the same variation as shown in Table 3. The impact on rotor stability is shown later.
Figure 5. Normalized Principle Stiffness Coefficients.
Figure 7. Approximate Impact of Frequency on Bearing Coefficients from Respondent #8. The coefficient variations for the labyrinth seals are presented on graphs of the tangential force components, Kxy versus Cxx. All respondents supplied symmetric coefficients for the seals. As previously mentioned, coefficients for impeller #1 and #5 eye seals and the balance piston were requested. The variations for these seals are plotted in Figures 8, 9, and 10. For the most part, the relative variations in Kxy are matched by the variations in Cxx. One exception is noted for a respondent supplying Alford type forces for the labyrinth seals. The bulk of the respondents impeller eye seal coefficients falls in the normalized range from 5 to 20. One respondent was less (at
Figure 6. Normalized Principle Damping Coefficients.
Table of Contents
Figure 8. Normalized Seal Coefficients for Impeller #1 Eye Labyrinth.
Figure 11. Normalized Destabilizing Force for the Balance Piston Labyrinth. Using a solution algorithm developed by Kirk (1990), the impact of the whirl frequency ratio (WFR) on the BP coefficients was studied. For the conditions specified, balance piston coefficients were calculated for 0.31 (6700 cpm), 0.45 (9700 cpm), and 1.0 (synchronous) whirl frequency ratios. Both cross-coupled stiffness and direct damping were smaller for the lower whirl frequency ratio. If the destabilizing force is defined as:
Qa = k Cwf
Figure 9. Normalized Seal Coefficients for Impeller #5 Eye Labyrinth.
Then calculating the Qa using the tangential coefficients derived from the three whirl frequency ratios produces the results shown in Table 4. As noted, Qa is only 10 percent greater for the higher subsynchronous WFR. The impact on stability of using 6700 cpm rather than 9700 cpm would be a secondary effect. Also notice the size of the destabilizing force if a WFR of 1.0 is used to predict the seal coefficients. This represents a 2 increase over the coefficients derived from a WFR of 0.45. Table 4. Impact of WFR on the Destabilizing Force of the Balance Piston.
STABILITY RESULTS
Coefficients supplied for the bearings and seals were incorporated into a common model of the compressor rotor. In addition, a single complex eigenvalue solution algorithm was used for each model. The bearings were represented by the principle terms only. Cross-coupling terms were consistently three orders of magnitude smaller than the principle terms and could thus be safely ignored. Labyrinth seal behavior was modeled using only the tangential terms supplied by the respondents due to the variations in the radial terms. Figure 12 plots the variation in the radial terms of impeller #1 eye seal. The principle stiffness varies by two orders of absolute magnitude and the cross-coupled damping by nearly three orders of magnitude, both ranging from positive to negative values. In addition, the relative size of each was large enough to significantly change the predicted frequency of the first forward damped eigenvalue.
Figure 10. Normalized Seal Coefficients for Balance Piston Labyrinth (Expanded Range Shown in Inset). 1.0) and one higher (at 42.) For the balance piston, the majority fell in the normalized range from 1 to 12 (an order of magnitude) with one at 37 and another at 137 (greater than two orders of magnitude.) (Confirmation of the values supplied by several respondents was requested but not received at the time of this paper.) Since the variations in cross-coupled stiffness and principle damping were nearly equal for all coefficients, it comes as no surprise that the destabilizing force, defined by Equation (1), also shows a similar range in normalized values, Figure 11.
Table of Contents
Figure 12. Variations in Radial Coefficients of Impeller #1 Eye Seal. The predicted compressor stability considering a rotor/bearings only model with the respondents coefficients is displayed in Figure 13. On the plot, the coefficients supplied by respondent #8 are highlighted for the two reduction schemes. A third point (in red) is plotted for an anticipated reduction to 9700 cpm. The impact on the base stability level of the rotor is secondary even with the 25 percent reduction in coefficient difference noted earlier in the paper. As is expected given the impact on the bearing coefficients, synchronous reduction of the bearing coefficients produces a lower first natural frequency (due to the softer stiffness values) and a higher logarithmic decrement (log dec) (due to the higher predicted damping.) The general trend can be seen on the plot as higher log dec values are produced at lower natural frequencies. This illustrates both the effect of reduction schemes and the increase in relative shaft-to-bearing stiffness making the bearing damping more effective in stabilizing the first mode.
Figure 14. Predicted Compressor Stability with Rotor/Bearings/ Labyrinth Seals. Two calculated points are not presented in the plotted range in Figure 14, one at approximately a log dec of 1.8 and the other at a log dec exceeding 2.0. These are omitted to more clearly define the influence of respondent #8s coefficients and due to the uncertainty of the validity of these points. An engineer analyzing this compressor at the onset of the design with the supplied coefficients would conclude a stable compressor design with a WFR less than tested in the majority of the cases. The subsynchronous frequency of the first mode as witnessed on the test stand was 10,700 cpm. This is plotted in Figure 14 against the predicted results. While the change is slight from 9700 cpm to 10,700 cpm, the change for most results would be in the wrong direction, to the left and up in Figure 14. Finally, the impact of the labyrinth seals on the rotor stability (log dec) is shown in Figure 15. The distribution seems skewed to the higher side ranging between a 0.2 to 1.5 change in log dec.
Figure 13. Predicted Compressor Stability with Rotor/Bearings Only. Adding the labyrinth seals to the stability model impacts the predicted log dec as shown in Figure 14. As before, the two reduction schemes supplied by respondent #8 are highlighted with their seal coefficients. A third point is added that includes both the bearing and seals coefficients altered representing a 9700 cpm location of the first damped forward mode. The change in log dec remains secondary to the overall variations in predicted rotor stability. In addition, a second respondents seal coefficients were changed by the amount indicated in Table 4. The 10 percent increase in destabilizing force was added to respondent #13s coefficients and the change indicated by the arrow in Figure 14.
Figure 15. Labyrinth Seal Impact on Predicted Log Dec.
TILTING PAD JOURNAL BEARING STIFFNESS AND DAMPING VARIANCE

From the previous sections, it is clear that there is a relatively wide variation in the stiffness and damping (K and C) properties provided by various computer codes for this stability study. In an attempt to investigate the source of this K and C variation, five different tilting pad journal bearing codes are utilized. References for the five codes are Nicholas, et al. (1979), Branagan (1988), San Andres (1995), He (2003), and Chen and Gunter (2005). The analysts were given the instructions to run the various codes for the tilting pad bearing from Table 5 as if they were about to perform an API stability analysis. Additional instructions included
Table of Contents
utilizing whatever thermal solution that was available within the code and to use a pad pivot stiffness of Kpiv = 8.0 105 lbf/in. Finally, synchronous coefficients were requested. Table 5. Stability Results Using Variable Viscosity Derived Coefficients from Four Bearing Codes at 21,662 RPM.
The results are illustrated in Figures 16, 17, 18, and 19 for Kxx, Kyy, Cxx, and Cyy, respectively. Note that only codes #1 through #4 are shown as code #5 did not have a variable viscosity capability (or the operator did not know how to use the capability). Additionally, code #2 had to be run with infinite pivot stiffness. Clearly, some K and C variation is evident.
Figure 18. Cxx - Variable Viscosity, Kpiv = 8.0105 lbf/in.
Figure 19. Cyy - Variable Viscosity, Kpiv = 8.0105 lbf/in. Stability results at 21,662 rpm using the variable viscosity K and C coefficients from the four codes discussed above are summarized in Table 5. Code #2 is listed twice for synchronous and nonsynchronous coefficients. All other results are for the synchronous coefficients from Figures 16, 17, 18, and 19. For synchronous coefficients, the variation in log dec is between 0.552 and 0.796. Code #2s nonsynchronous coefficients produce a log dec value of 0.128. Eliminating this variation in temperature predictions and thus in viscosity variation, the codes were rerun assuming a constant viscosity. The iso-viscous results are shown in Figures 20, 21, 22, and 23. While the damping variation is negligible (Figures 22 and 23), the stiffness values still show some variation. Closer examination of Figures 20 and 21 reveal that codes #1 and #5 predict the same values while codes #3 and #4 predict the same values. Codes #1 and #5 use the pad assembly solution technique (Lund, 1964) while codes #3 and #4 utilize the full pad solution technique.
Figure 16. Kxx - Variable Viscosity, Kpiv = 8.0105 lbf/in.
Figure 17. Kyy - Variable Viscosity, Kpiv = 8.0105 lbf/in. At 21,662 rpm, the maximum pad metal temperature (Tmax) predictions from the four codes varied from 173F (#1) to 221F (#4). Ignoring #2 as it was run with infinite pivot stiffness, the lowest predicted temperature resulted in the highest predicted K and C values (#1). Likewise, the highest predicted temperature resulted in the lowest predicted K and C values (#4).
Figure 20. Kxx - Iso-viscous, Kpiv = 8.0105 lbf/in.
Table of Contents
Figure 24. Kxx - Iso-viscous, Kpiv = Infinite. Figure 21. Kyy - Iso-viscous, Kpiv = 8.0105 lbf/in.
Figure 25. Kyy - Iso-viscous, Kpiv = Infinite.
Figure 22. Cxx - Iso-viscous, Kpiv = 8.0105 lbf/in.
Figure 26. Cxx - Iso-viscous, Kpiv = Infinite.
Figure 23. Cyy - Iso-viscous, Kpiv = 8.0105 lbf/in. Eliminating the pad pivot stiffness as a possible source of variation, the codes were rerun assuming infinite pivot stiffness and constant viscosity. The iso-viscous, infinite pivot stiffness results are shown in Figures 24, 25, 26, and 27. Now, all five codes predict essentially the same results for speeds above 4,000 rpm. Below 4,000 rpm, four out of the five codes predict essentially identical results.
Figure 27. Cyy - Iso-viscous, Kpiv = Infinite.
Table of Contents
CONCLUSIONS
This work presents surveyed bearing and seal coefficients from various industry sources including academia, manufacturers, users, and consultants. The variation in those coefficients and their impact on an example compressor rotor stability prediction was highlighted. In an effort to explain the origin of the bearing coefficient differences, five widely used bearing codes were compared. Various input options affecting the thermal solution and pivot stiffness were investigated for their effect on the predicted stiffness and damping of the tilting pad bearing. The industry survey supplied sufficient (but maybe not exhaustive) data to confirm the working hypothesis that significant differences exist in the prediction of dynamic coefficients for tilting pad journal bearings and gas labyrinth seals. The differences arise from several sources, some due to the solution algorithm and some traced to the user. Whatever the source, sufficient evidence exists to validate the following conclusions: pad journal bearings and gas labyrinth seal dynamic coefficients. The data should be used to validate analytical prediction methods. Preferably, the experimental data should be obtained for identical components from several sources. Due to variations in testing procedures, test equipment, etc., multiple sources should be encouraged and funded to obtain the data. Emphasis should be placed on obtaining the component information. While system behavior may provide insight into overall component behavior, i.e., nonsynchronous versus synchronous reduction (Cloud, 2006), it does not provide enough information to validate component coefficient predictions.
decreases by 25 percent the difference between synchronously and subsynchronously reduced bearing coefficients and increases the destabilizing force by 10 percent for the balance piston seal. Both were shown to produce only secondary effects. From the computer code survey, code-to-code variations in the tilting pad journal bearing stiffness and damping coefficients were found to be due to several sources:
Synchronous versus nonsynchronous coefficients
This is the major source of the K and C variation.
The temperature solution technique and the resulting pad metal

temperature prediction and, thus, the viscosity variation The pad assembly codes handle the inclusion of pivot stiffness differently compared to the full solution codes. Stripping out the above three variations, all five codes produce essentially identical synchronous coefficients for the infinite pivot stiffness, iso-viscous case. With regard to the industry survey, variations in the bearing coefficients stem from user assumptions affecting pivot stiffness and the frequency reduction of the coefficients. These variations appear to be related more to the user controlled analysis options rather than the analytical methods employed. The authors hope the presented work can be used to stimulate funded research in the areas of tilting pad and labyrinth seal coefficient predictions/measurements and the impact of frequency dependency on both. As compressor development continues to expand in size and power with an accompanied increase in the cost of unexpected downtime or project delays, accurate prediction of the compressor dynamic behavior is essential.
The methodology of the inclusion of pivot stiffness
A gold standard of experimental data is needed for both tilting
The Level I analysis in the API specifications is still needed. The original intent of that analysis was to provide a screening tool to identify rotors requiring only a simplified analysis and to provide a common methodology for stability calculations. The wide variations in bearing and seal coefficients and continued debate on synchronous versus nonsynchronous reduction of tilting pad bearing coefficients, still necessitate a common methodology to permit valid comparisons across the industry. For the survey compressor, nonsynchronous reduction of bearing coefficients appears to represent the rotor support situation more accurately. Synchronous coefficients tended to underpredict the frequency and overpredict the stability level of the first forward mode. However, strong caution is advised in drawing widespread conclusions without addressing the significant variations in bearing and seal coefficients shown in the study. Further research, both on a component and system level, is needed in this area.
The predicted stability of the survey compressor was greatly affected by the variations in the bearing and seals coefficients. Frequency predictions for the first forward mode ranged from 6000 to 11,300 cpm. Log dec magnitudes from +1.0 to 1.0 even after ignoring the extremes. While the authors understand that not all compressors will show this sensitivity, the survey compressor shares traits with rotors typically showing instability problems, namely, lightweight rotors operating at high speeds and pressures.
NOMENCLATURE
Cs C, Cxx D Hc K, Kxx k, Kxy Qa Tw Th Ts WFR wwf = Seal diametral clearance, mm (mils) = Principle damping, N-s/mm (lbf-s/in) = Labyrinth seal diameter, mm (in) = Minimum width of the impeller or discharge volute, mm (in) = Principle damping, N-s/mm (lbf/in) = Cross-coupled stiffness, N/mm (lbf/in) = Destabilizing force, N/mm (lbf/in) = Labyrinth seal tooth tip width, mm (in) = Labyrinth seal tooth height, mm (in) = Labyrinth seal tooth spacing, mm (in) = Whirl frequency ratio = wwf/operating speed = Whirl frequency of first natural frequency, rad/sec
REFERENCES
American Petroleum Institute Standard 617, 2003, Axial and Centrifugal Compressors and Expander-Compressors for Petroleum, Chemical and Gas Industry Services, Seventh Edition, American Petroleum Institute, Washington, D.C. American Petroleum Institute RP 684, 2005, API Standard Paragraphs Rotordynamic Tutorial: Lateral Critical Speeds, Unbalance Response, Stability, Train Torsionals, and Rotor Balancing, Second Edition, American Petroleum Institute, Washington, D.C. Barrett, L. E., Allaire, P. E. and Wilson, B. W., 1988, The Eigenvalue Dependence of Reduced Tilting Pad Bearing Stiffness and Damping Coefficients, ASLE Transactions, 31, pp. 411 - 419. Branagan, L. A., 1988, Thermal Analysis of Fixed and Tilting Pad Journal Bearings Including Cross-Film Viscosity Variations and Deformation, Ph.D. Dissertation, University of Virginia, Charlottesville, Virginia.
Analytical predictions of labyrinth seals, both short and long, are still incomplete and, in some cases, insufficient. As noted in the study, the radial force coefficients of the seals were ignored due to the extreme variations (approaching three orders of magnitude) in the coefficients. The study also showed that for the balance piston seal, the destabilizing force, Qa, produced from synchronously derived coefficients to be 2 greater than that produced from coefficients derived from a WFR of 0.45.
either the bearing or seal coefficients was also studied using two specific solution algorithms. A change in WFR from 0.33 to 0.45
The impact of the subsynchronous frequency used in solving for
Table of Contents
Camatti, M., Vannini, G., Fulton, J., and Hopenwasser, F., 2003, Instability of a High Pressure Compressor Equipped with Honeycomb Seals, Proceedings of the Thirty-Second Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 39-48. Chen, W. J. and Gunter, E. J., 2005, Introduction to Dynamics of Rotor-Bearing Systems, Victoria, British Columbia, Canada: Trafford Publishing. Childs, D. W., 1993, Turbomachinery Rotordynamics: Phenomena, Modeling, and Analysis, New York, New York: John Wiley & Sons, Inc. Childs, D. W. and Scharrer, J. K., 1986, An Iwatsubo-Based Solution for Labyrinth Seals: Comparison to Experimental Results, ASME Journal of Engineering for Gas Turbines and Power, 108, (2), pp. 325-331. Childs, D. W. and Ramsey, C., 1991, Seal RotordynamicCoefficient Test Results for a Model SSME ATD-HPFTP Turbine Interstage Seal With and Without a Swirl Brake, ASME Journal of Tribology, 113, pp. 113-203. Cloud, C. H., 2006, Stability of a Rotor Supported on Tilting Pad Journal Bearings, Ph.D. Dissertation, University of Virginia, Charlottesville, Virginia. Elrod, D. A., Pelletti, J. M., and Childs, D. W., 1995, Theory Versus Experiment for the Rotordynamic Coefficients of an Interlocking Labyrinth Gas Seal, ASME Paper 95-GT-432, Presented at the International Gas Turbine and Aeroengine Congress and Exposition, Houston, Texas. He, M., 2003, Thermoelastohydrodynamic Analysis of Fluid Film Journal Bearings, Ph.D. Dissertation, University of Virginia, Charlottesville, Virginia. Iwatsubo, T., Matooka, N., and Kawai, R., 1982, Spring and Damping Coefficients of the Labyrinth Seal, NASA CP-2250, pp. 205-222. Kirk, R. G., 1990, A Method for Calculating Labyrinth Seal Inlet Swirl Velocity, ASME Journal of Vibration and Acoustics, 112, (3), pp. 380-383. Kocur, J. A., Jr. and Hayles, G. C., Jr., 2004, Low Frequency Instability in a Process Compressor, Proceedings of the Thirty-Third Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 25-32.
Lund, J. W., 1964, Spring and Damping Coefficients for the Tilting-Pad Journal Bearing, ASLE Transactions, 7, pp. 342-352. Nicholas, J. C., 2003, Lunds Tilting Pad Journal Bearing Pad Assembly Method, ASME Journal of Vibrations and Acoustics, 125, (4), pp. 448-454. Nicholas, J. C., Gunter, E. J., and Allaire, P. E., 1979, Stiffness and Damping Coefficients for the Five Pad Tilting Pad Bearing, ASLE Transactions, 22, (2), pp. 112-124. Nordmann, R. and Weiser, H., 1990, Evaluation of Rotordynamic Coefficients of Look-Through Labyrinths by Means of a Three Volume Bulk Model, Rotordynamic Instability Problems in High-Performance Turbomachinery, NASA CP-3122, pp. 141-157. Parsell, J. K., Allaire, P. E., and Barrett, L. E., 1983, Frequency Effects in Tilting-Pad Journal Bearing Dynamic Coefficients, ASLE Transactions, 26, pp. 222-227. San Andres, L. A., 1995, Bulk-Flow Analysis of Flexure and Tilting Pad Fluid Film Bearings, TRC-B&C-3-95, Turbomachinery Laboratory, Texas A&M University, College Station, Texas. Vzquez, J. A. and Barrett, L. E., 1998, Representing Flexible Supports by Polynomial Transfer Functions, ASME Paper 98-GT-27. Vzquez, J. A., Barrett, L. E., and Flack, R. D., 2001, A Flexible Rotor on Flexible Bearing Supports: Stability and Unbalance Response, Journal of Vibration and Acoustics, ASME Transactions, 123, (2), pp. 137-144. Wagner, N. G. and Steff, K., 1996, Dynamic Labyrinth Coefficients from a High-Pressure Full-Scale Test Rig Using Magnetic Bearings, Rotordynamic Instability Problems in High-Performance Turbomachinery, NASA CP-3344, pp. 95-111. Wygant, K. D., Barrett, L. E., and Flack, R. D., 1999, Influence of Pad Pivot Friction on Tilting-Pad Journal Bearing MeasurementsPart II: Dynamic Coefficients, STLE Tribology Transactions, 42, (1), pp. 250-256.
ACKNOWLEDGEMENT
The authors recognize API (Roland Goodman), ExxonMobil, Solar Turbines, and RMT, Inc., for their assistance and support.
Table of Contents
ROTOR BEARING LOADS WITH HONEYCOMB SEALS AND VOLUTE FORCES IN REINJECTION COMPRESSORS
by Leonardo Baldassarre
Engineering Manager & Principal Engineer for Centrifugal Compressors General Electric Oil & Gas Company Florence, Italy
and John W. Fulton

Senior Engineering Advisor ExxonMobil Research and Engineering Company Fairfax, Virginia
Leonardo Baldassarre is currently the Engineering Manager & Principal Engineer for Centrifugal Compressors with General Electric Oil & Gas Company, in Florence, Italy. He is responsible for all requisition, standardization, and CAD automation activities as well as for detailed design of new products for centrifugal compressors both in Florence (Nuovo Pignone) and Le Creusot (Thermodyn). Dr. Baldassarre began his career with General Electric Nuovo Pignone in 1997. He has worked as Design Engineer, R&D Team Leader for centrifugal compressors in Florence, Product Leader for centrifugal and axial compressors, and Requisition Manager for centrifugal compressors both for Florence and Le Creusot teams. Dr. Baldassarre received a B.S. degree (Mechanical Engineering, 1993) and Ph.D. degree (Mechanical Engineering/Turbomachinery Fluid Dynamics, 1998) from the University of Florence. He has authored or coauthored 20 technical papers, mostly in the area of fluid dynamic design of 3D transonic impellers, rotating stall, and rotordynamics. He presently holds three patents. John W. Fulton is a Senior Engineering Advisor with Exxon Mobil Research and Engineering Company, in Fairfax, Virginia. In his 35 years with Exxon, he has worked in all phases of machinery engineering and in research and development. Mr. Fulton enjoyed years of assignments in Libya, Venezuela, Alaska, London, and Kuala Lumpur. He is co-inventor of six U.S. Patents. Mr. Fulton has a B.S. degree (Mechanical Engineering) from New Jersey Institute of Technology.
the usual position in the lower part of its journal bearing, degrading the unbalance response. In this paper the authors show that the honeycomb seal can produce a large disturbance in the equilibrium position of the rotor for certain values of negative stiffness, resulting in high bearing loads in unusual directions. It is also shown that aerodynamic forces on the rotor from the volute need to be considered.
INTRODUCTION
In the oil and gas industry, the typical centrifugal compressor for reinjection duty has its impellers placed in the casing between two bearings, with the shaft horizontal with respect to gravity. To calculate the vibration response of the rotor to unbalance, or the damped critical speeds (to evaluate rotordynamic instability) one has to know the bearing characteristics, which depend on the bearing loading. Most rotordynamic suites include a simple calculation that automatically finds the rotor weight and center of gravity. This calculation of static equilibrium then finds the load at each bearing by solving two equations, one found by setting the sum of forces equal to zero, and the other found by setting the sum of moments around one bearing equal to zero. This calculation will be shown in detail below. This paper is motivated by experience with several compressors on full-load test. The specific compressors involved, and the operating conditions on test are given in Table 1. Table 1. Specific Compressors and Operating Conditions.
ABSTRACT
The calculation for unbalance response of a rotor starts by calculating the bearing load to provide the basis for the bearing stiffness and damping characteristics. Measurements in test rigs at a major Texas university laboratory have shown that honeycomb-stator/drum-rotor annular seals can produce negative stiffness, in particular at zero to low whirl frequencies, which tends to pull the rotor off-center. Data given in this paper, from high-pressure factory tests of compressors using a honeycomb seal at the balance piston, have shown the rotor can be displaced from
11
These compressors use a honeycomb seal running against a drum rotor for the balance piston. In contrast to labyrinth seals with teeth, or in contrast to rotors with labyrinth teeth running against honeycomb, the honeycomb/drum type seals have significant direct stiffness (defined more completely below). Hole pattern type seals running against drum rotors are similar. Because the test experience related in this paper was with honeycomb seals, not hole pattern type, this paper will discuss honeycomb seals only. Figure 1 shows a photograph of the bore of a honeycomb seal. There are about 10,000 cells in this half of the seal. Each cell has a hexagonal width of about 2 mm (.08 inch) and a depth of about 2.3 mm (.09 inch). This seal is formed from a monolithic billet of aluminum, for all examples. In these examples the honeycomb seal acts on the balance piston that is just behind the last impeller.
Table of Contents
Figure 1. Photo of a Honeycomb Seal. These compressors also have a volute at the last impeller. The volute can produce asymmetric pressure gradients around the rotor. If the gas pressure is sufficiently high, then the asymmetric gradient may produce a radial force on the rotor that exceeds the rotor weight, as shown later.
EXPERIENCE ON FULL LOAD TEST

During full-load tests of compressor Example A, erratic response to unbalance was noted. Running at constant speed, the synchronous vibration varied from 6 to 25 microns peak-to-peak, as pressure and flow were varied. Clearly the high pressure gas was affecting the unbalance response of the rotor, either directly, or by affecting the bearing load, and thus the bearing dynamic characteristics. In an attempt to identify the cause, the pressure and flow were varied and repeated, over an appreciable range, as shown in Figure 2 and 3, respectively.
Figure 2. Synchronous Response (Filtered at 1 RPM) of Compressor Example A at the Discharge End X-Probe as a Function of Discharge Temperature.
From these figures it is clear that the change in unbalance response was well demonstrated, repeatable, and significantly affected. The three points at lower flow correspond to the three points at higher temperature. The separation of points is not as distinct when plotted against pressure. From previous work by Camatti, et al. (2003), it is known that the forces produced by a honeycomb seal in high pressure gas are sensitive to the clearance of the leakage annulus. From finite element analysis of the honeycomb seal in this compressor, it is known that temperature changes the clearance and causes a taper in the clearance as well. Therefore it is not surprising that the rotor response could vary with discharge temperature, which sets the temperature of the seal, distorting the taper, and thus changing the honeycomb seal forces on the rotor. These changes apply to both the static force and the dynamic stiffness and damping coefficients acting synchronously. It will be shown below, by calculation for the test stand conditions of Example C, that the volute force can exceed the rotor weight. The force and direction of the radial load produced by the volute depends on the ratio of volume flow to the design flow. Thus changing the flow can produce substantial additional load on the bearings, and thus affect their dynamic characteristics. Thus it is not surprising that the unbalance response might vary with volute-induced bearing loads. However, no practical method was obvious to distinguish between the bearing loads caused by the honeycomb seal versus the volute. The rotordynamic stability of the Example A (and Example B, discussed below) compressors were excellent as tested. The response-to-unbalance was also within contract vibration limits. Therefore, an extensive investigation during full-load testing was not warranted, and an exact comparison of the rotor position on test and the position calculated below was not made. However, the effects on unbalance response in Example A were significant, and could have caused a problem, if the rotor response were not so well behaved in the base case. This gave motivation to find an analytic solution to understand the lifted rotor position that was observed and estimate the bearing loads that might occur. In the case of Example B (and in other cases [Camatti, et al., 2003]) it was found that the proximity probes at the bearing journals showed the rotor journal was not resting in the bottom of the bearing when running, as assumed in the bearing load calculation that was outlined above. Instead the journal was often found in the top of the bearing, and the journal could be seen to lift as the compressor was started and loaded. Figure 4 shows the bearing journal position, measured by proximity probes, when the compressor was started and loaded. The left plot is for the balance piston end, and the right is the thrust end. The large circles represent the bearing clearance circle (nominal cold dimensions.) The parameter written by each point is revolutions per minute (rpm). The suction volume flow 76 percent of rated at full speed. From the calculated behavior of a tilt-pad bearing, one could expect that the journal should rise toward the center of the bearing, but not as high as the bearing center. Clearly, the journals in Example B rise above center, and more so, on the balance piston end.
Figure 3. Synchronous Response (Filtered at 1 RPM) of Compressor Example A at the Discharge End X-Probe as a Function of Suction Volume Flow.
Figure 4. Bearing Centerline Position as Compressor Example B is Started and Loaded. The Plot on Left is the Balance Piston End and the Plot on the Right is the Thrust End.
Table of Contents
ROTOR BEARING LOADS WITH HONEYCOMB SEALS AND VOLUTE FORCES IN REINJECTION COMPRESSORS 13
The tilt-pad bearings used on these compressors have thermocouples installed to measure the Babbitt temperature of the two bottom pads. On Example B, as a check on the validity of the journal center positions shown in Figure 4, one pad was moved from the bottom to the top of one bearing, with the second pad remaining in normal position. As expected from Figure 4, the top pad showed higher temperature than the bottom, confirming the load on the bearing was directed upward, and confirming the behavior shown there. The maximum temperatures measured were as shown in Table 2. Table 2. Maximum Babbitt Temperatures.
The honeycomb seal forces are effectively proportional to the
displacement of the drum within the honeycomb seal running clearance, That is, they have the characteristic of a spring rate that can be measured in Newtons per meter (pounds per inch).
The volute forces change as the flow rate changes with respect to the design (best efficiency) flow. Of course their magnitude changes with pressure. However, the volute forces are independent of small changes of the rotor position within its running clearance. That is, the volute force does not have the characteristic of spring. The
honeycomb seal spring rates are extremely sensitive variation of the running clearance along the length of the drum. A well-considered finite element analysis of the drum and of the honeycomb seal is necessary to define the clearance along the length as a function of temperature, pressure, shaft speed, and mounting conditions. Changes in these variables during testing may change the static equilibrium of the rotor position.
Thus the top pad (72-TE-29031A) of the bearing near the balance piston is 17C (62.6F) hotter than the bottom pad, indicating a substantially larger load on the top pad. Just to give a complete picture, these temperature records are plotted versus time in Figure 5.
honeycomb seal spring rates in both the direct and cross-coupled directions are strong functions of the whirl frequency of the drum orbital motion. For calculation of static equilibrium the displacement occurs at zero whirl frequency. seal spring rate, K, in the direction of drum displacement may affect the first bending frequency of the rotor. This reduced whirl frequency can fall into the range where the honeycomb has negative damping. This can cause rotordynamic instability, resulting in catastrophic vibration. Such a problem is reported in a previous paper (Camati, et al., 2003) but is not of concern here as the subject compressors were very stable. The honeycomb seal spring rates were calculated using the seal code developed by Kleynhans and Childs (1997). The volute forces were calculated using a computational fluid dynamics code. Figure 6 shows the coordinate system used. In a free body diagram for static equilibrium all forces are considered to be acting on the rotor. Lhc is the distance from Bearing 1 to the center of the honeycomb seal.
The
Honeycomb
Figure 5. Trend Plot of the Bearing Pad Temperatures (Top Four Traces) on Example B, Bottom Trace is Speed in RPM.
CALCULATING THE EFFECT OF HONEYCOMB SEAL ON STATIC EQUILIBRIUM OF THE ROTOR

Because both the potential honeycomb seal force and the potential volute force could be responsible for the anomalous journal position and bearing loads discussed above, the two effects were investigated by calculations after the testing was complete. The bearings themselves were not suspected of causing the anomalous position, as multiple disassemblies to change dry gas seals during some of the testing did not implicate them. On one instance with Example B, an increase once-per-revolution vibration prompted a bearing inspection that showed faulty assembly. However correcting this did not eliminate the anomalous position. The remainder of this paper will discuss the results of the calculation. As a basis for understanding the calculations, it is necessary to compare and contrast the general behavior of the honeycomb seal direct stiffness and the volute force as follows:
Figure 6. Rotor Coordinate System. The simple calculation of the bearing load in the typical rotordynamic code uses two equations. The sum of the forces, and the sum of the moments, must be equal to zero, for static equilibrium. For the simple calculation without the honeycomb seal, it is not necessary to know the spring rates of the bearings as one can find the two bearing loads with the two equations. The static equilibrium can be written by summing the forces to zero (Equation 1) plus summing the moments to zero (Equation 2). Only the vertical plane need be considered, as there are no horizontal forces.
The balance drum/honeycomb seal was next to the volute (as usual) in the above cases, so that it was not obvious by comparing the behavior of the bearing on one end of the casing to the other, whether the honeycomb seal or the volute was responsible for the anomalous journal position in the bearing, as might be expected from the static equilibrium calculations.
fb1 + fb2 Wr = 0
Lcg Wr Lcl fb2 = 0
Table of Contents
The standard model for forces acting on the rotor is given in Equation (3) (Kleynhans and Childs, 1997).
K Fx = 1 Fy k
k X C c Xdot + K Y c C Ydot
gives the displacement due to rotor bending under force from the honeycomb seal. The displacements in the horizontal plane, shown in Equation 11, are similar but of course do not include bending due to gravity.
x1 +
For static equilibrium, the velocities Xdot and Ydot are zero, which eliminates the damping coefficients, C and c, from consideration. This equation can be applied to the static force caused by the honeycomb seal acting on a rotor, as expressed by Equations (4) and (5):
Lhc Lcl
(x 2 x1) +
Fhx ( x 3, y 3) Krotor
x3
HCradclear
=0
Fhy (x , y ) : = kh x Kh y
Fhx (x , y ) : = Kh x kh y
In Equations (4) and (5) the direct stiffness is K and the cross-coupled stiffness is k. Note that the cross-coupled stiffness requires that the horizontal displacement be included in the equations for vertical forces and vice versa. The subscript y indicates the force from the honeycomb acts in the y (vertical) direction and the subscript x indicates action in the horizontal direction. When the honeycomb seal is added, there are now three unknown force vectors, the two force vectors of the bearings on the rotor and now the force vector of the honeycomb seal on the rotor. However, there are only two force equations (one in the horizontal direction and one vertical) and two moment equations. Thus the calculation becomes a statically indeterminate problem (Popov, 1968), because the force caused by the honeycomb seal depends on its drum displacement. Six equations are required to solve for the six displacements. Two force balances are provided by Equations (6) and (7) and two moment balances are provided by Equations (8) and (9).
Figure 7. Definition of the Rotor Displacements in the Vertical Plane. The spring coefficient representing the rotor bending stiffness is easily found by using a rotor response code to calculate an asynchronous response at very low frequency (1 Hertz) to find the displacement at the honeycomb seal caused by an asynchronous force at the honeycomb location As mentioned, Equation (10) also includes the displacement of the rotor at the honeycomb seal location, yhcstatic due to rotor weight. It is handled separately as it remains constant while the other displacements in Figure 6 vary. Handling it separately avoids complication due to the weight action not being at the honeycomb seal location. For the examples given, yhcstatic is a significant term. The authors estimated it from the first bending frequency of the rotor, using the concept of the resting displacement of a single spring-mass oscillator. The yhcstatic is adjusted for the actual bending curve of the rotor, using the concept of the Raleigh natural frequency method, using the first bending mode shape as calculated by a rotordynamic code. This system can be solved for the six particular values of displacement, using a standard numerical solution method. Because the force equations are written as functions of displacement, no further algebra is required when using the solver in a popular technical calculation software. Figure 8 shows the result graphically in the same format as used in Figure 4. Please note that the direction of rotation as viewed in Figure 4 is clockwise, while it is counterclockwise as viewed in Figure 8. (This reverses the horizontal displacements between the two figures, as the view is from opposite ends of the rotor.) The displacements are normalized by their respective clearance circles to indicate the limit of validity of the solutions, due to contact if a displacement exceeds the clearances. The honeycomb seal clearance is larger than the bearing clearance. The displacement of the seal drum is marked by the solid triangle symbol, bearing 1 (near the honeycomb seal) by the circle, and bearing 2 by the square. When the honeycomb seal bore is not concentric with the centerline between the two bearing bores, a variable called offset can be introduced to represent the honeycomb bore position. To include this in the above calculation, the offset must be included in the right-hand side of Equations (4) and (5) by adding it to the variable displacements. The honeycomb seal stiffness and damping, calculated by the seal code developed by Kleynhans and Childs (1997) are 0.150 E9 N/m (0.857 E6 lbf/in) and 0.200 E9 N/m (1.142 E6 lbf/in), respectively. These values are for a honeycomb seal with diverging clearance. They are for the compressor of Example B, but with slightly different
Fby (x1, y1) + Fby (x 2, y 2) + Fhy (x 3, y 3) Wr Wr
=0
Fbx (x1, y1) + Fbx (x 2, y 2) + Fhx (x 3, y 3) =0 Wr

Lcg Wr Lhc Fhy (x 3, y 3) Lcl Fby (x 2, y 2) Wr Lcl
=0
Lhc Fhx (x 3, y 3) + Lcl Fbx (x 2, y 2) =0 Wr Lcl

To account for the cross coupling, the forces are written as functions of both horizontal (x) and vertical (y) displacements. The force balance equation is normalized by rotor weight, Wr, to suit the tolerance limit of the numerical method used. The moment equations are normalized by moment Wr Lcl. Figure 7 shows the displacements in the vertical direction as defined by Equation (10).
y1 +
Lhc Lcl
( y 2 y1) +
Fhy ( x 3, y 3) Krotor
+ yhcstatic y 3
HCradclear
=0
That equation gives the displacement of the rotor drum at the honeycomb seal, y3. The first term, y1, is the displacement of the bearing at the coordinate origin. The second term uses the displacement of the second bearing, y2, to find the centerline displacement at the honeycomb seal location. The third term represents the bending of the rotor due to gravity. The fourth term
Table of Contents
internal parts, and slightly lower pressure and speed than tested for Figure 4. The cross-coupled stiffness causes the rotor center to be displaced to one side, instead of remaining under the center of the tilt-pad bearing (which has no cross-coupled stiffness).
values and the displacement suddenly changes direction downward to negative values. This system of equations indicates very large forces on the bearings near a particular value of honeycomb seal stiffness. Mathematically, the force is unbounded and approaches an asymptote (defined in Thomas, 1968) located at a particular magnitude of direct stiffness of the honeycomb.
A GRAPHICAL ILLUSTRATION OF THE ASYMPTOTIC BEHAVIOR

The following explanation is intended to aid visualization of how the asymptotic behavior occurs. It is based on the concept of solving two simultaneous equations graphically. To show the concept of the graphical solution, consider two simultaneous Equations (12) and (13) (where a, b, c, and d are numerical constants):
y = a*x +b
y = c*x +d
Figure 8. Journal and Drum Positions in their Clearance Circles Example B. Alternate Bundle, Diverging Clearance in Honeycomb, 9965 RPM. Figure 8 represents only one value of honeycomb seal direct and cross-coupled stiffness. The behavior of the static equilibrium calculation varies dramatically with direct stiffness of the honeycomb seal, when negative direct stiffness is considered. Figure 9 shows this vertical displacement of the honeycomb seal drum as a function of honeycomb direct stiffness, over a range of negative seal stiffness, with all other input values held constant. In this figure, the drum displacement is normalized by its radial clearance while the honeycomb seal direct stiffness is normalized by the bending stiffness of the rotor calculated at the seal location. The expected value of the direct stiffness is marked by the line labeled expected. At the expected stiffness, the rotor is displaced upward to the level marked by the line labeled Root2, thus explaining how the bearings may run against their top pads as shown in Figure 4. Each equation can be plotted as a straight line on x-y coordinates. If the two equations are independent and consistent equations that apply simultaneously, then their solution (a particular value of x and of y that satisfies the equations) is found where the two lines cross (Ayres, 1958). To form the first line of the graphical solution, conduct a thought experiment on the rotor. Take the rotor running on its bearings. Apply an arbitrary force in the positive vertical direction on the rotor at the location were the honeycomb seal acts. Measure the displacement of the rotor at that location. Plot the arbitrary force on the Y axis and the resulting displacement on the X axis. This is done in Figure 10, forming the red line, Y1. The slope of the red line represents the stiffness of the rotor-bearing system. This slope is not constant here because the static bearing stiffness acts only in the direct axis and is not cross-coupled for the tilt-pad bearings used. The nonlinearity is included in the analysis to accurately portray the bearing characteristic at high eccentricity. Figure 11 shows the values used.
Figure 9. Vertical Force on Bearing Journals as a Function of Arbitrary Direct Stiffness of the Honeycomb Seal. As the honeycomb seal stiffness becomes more negative, the upward displacement increases at a larger rate, reaching 100 percent of its clearance. If it were not limited by rubbing, the calculated displacement would increase without bound, to reach the line labeled asymptote. Going further left to more negative
Figure 10. Graphical Solution with Positive Honeycomb Seal Stiffness.
Table of Contents
Figure 11. Bearing Force and Bearing Static Stiffness, as a Function of Eccentricity. If the force at the honeycomb seal position is zero, the equilibrium position occurs where the red line crosses the X axis (force is zero there). In this instance, one can see in Figure 10 that equilibrium occurs at about 25 percent of the radial clearance below the center of the honeycomb seal. (The X axis of the plot represents the honeycomb seal drum vertical displacement normalized by radial clearance of the seal. The Y axis is force on drum normalized by rotor weight, Wr.) Now do another experiment. Center the rotor drum in the honeycomb seal and lift the rotor up. For a honeycomb seal with positive stiffness the honeycomb seal will resist with a downward (negative) force on the rotor. Do this for a series of points and plot as the blue line (Y2) as done in Figure 10. The stiffness of the honeycomb seal in this plot is minus one times the slope of the blue line. The two lines represent two equations that are solved where the lines cross. The static equilibrium occurs where the red and blue line cross, which is about 15 percent of the radial clearance above the center of the honeycomb seal, at the vertical line marked Root1. This is expected, as the positive stiffness of this honeycomb seal is helping to support the rotor, lifting it up from the 25 percent position (below center) found without considering the honeycomb seal support. Knowing the force and displacement at the seal, the moment balance equation can be used to solve for the force at bearing two, and then the force balance equation can be used to solve for the force at bearing one. To demonstrate how the bearing load reaches the asymptote when the negative stiffness of the honeycomb increases, the above plot will be repeated with a honeycomb seal stiffness that has a slope near the slope of the rotor characteristic (red). This represents a honeycomb seal having negative stiffness. That is, as the drum is displaced away from the center, the honeycomb seal tends to pull the drum further off center. Figure 12 shows this graphical solution for rotor equilibrium in the vertical plane with negative honeycomb direct stiffness.
The graphical solution is also useful to show honeycomb seal offset from the bearing centerline. In Figures 10 and 12 the blue line is offset from the center representing an upward displacement of the honeycomb bore. The green line is offset by an equal and opposite amount showing a low honeycomb bore. The possible range of solutions for this offset falls between the blue and green lines, representing an acute sensitivity to concentricity. In Figure 12 the red and blue lines cross near 100 percent of the radial clearance below the honeycomb sea, at the displacement marked Root 1. This intersection is the equilibrium position. Note that with negative stiffness small changes in the slope of the blue line can move the crossing point with the red line to very large positive or negative values, thus representing large forces on the bearings. Of course, checking this graphical solution against the algebraic solution, shown earlier, gives identical results. However, the graphical solution gives more insight. From this graph, the learning is to avoid negative stiffness whose absolute value is near the stiffness of the rotor bearing system, that is, near the asymptote. Note that the position of the asymptote depends on the stiffness of the rotor bearing system as well as the stiffness of the honeycomb seal. Even larger negative stiffness, to the left of the asymptote may create a larger problem, as it is likely to depress the whirl frequency (typically the first bending mode) possibly causing the honeycomb to produce negative damping, and thus causing rotordynamic instability. Actually plotting the graph is not necessary to find the solution, as the intersection of the rotor and honeycomb seal characteristic lines is easily found by setting the equations for the two lines equal and finding the displacement of the rotor in the honeycomb seal that satisfies the two equations. However plotting the graph is useful to visualize if the solution is near the asymptote. For a rigid rotor solution it is possible to solve for the asymptote in closed form based on the lengths of the components along the rotor. Given two identical bearings of stiffness Kyy, the critical honeycomb negative stiffness where the asymptote occurs is shown in Equation (14).
Kcrit : = ( Kyy )
Lb2 Lb2 + 2 Lhc2 2 Lhc Lb
This equation only applies for zero offset.
BEHAVIOR DISPLAYED BY THE GRAPHICAL SOLUTION

For practical application, rotor bending, rotor sag due to gravity, honeycomb seal offset from the bearing centerline, and bearing nonlinearity are usually highly significant. The honeycomb seal offset needs to be controlled by tolerances, or by adjustment on assembly. For the compressor of Example B, all these effects were included in the calculation of static equilibrium. Figure 12 gives the graphical solution for this case. The red line shows the force-displacement characteristic of the rotor-bearing system. It shows a hardening spring rate due to the tilt-pad bearing reaching high eccentricity, as found by a Reynolds code for laminar flow in bearings. In Figure 12, the dashed green line shows a honeycomb seal with negative stiffness, whose center is displaced 60 microns (2.3 mils) below the bearing centerline, shown on the graph as a displacement to the left on the horizontal axis, at about 40 percent of the radial clearance of the honeycomb seal. The intersection of the red and green lines shows the equilibrium position in honeycomb seal just above center (Root2). The dashed blue line shows a honeycomb seal with negative stiffness, whose center is displaced 60 microns (2.3 mils) above the bearing centerline, shown as a displacement to the right on the horizontal axis. The intersection of the red and blue lines shows the calculated equilibrium position of the drum in the honeycomb seal would heavily depress the rotor below its centerline. The displacement normalized by the honeycomb seals
Figure 12. Graphical Solution for Rotor Static Equilibrium for Negative Stiffness of the Honeycomb Seal.
Table of Contents
radial clearance is near minus one, indicating the drum is close to rubbing the honeycomb surface at the calculated equilibrium. This behavior may be self-limiting, because if the rotor drum rubs the honeycomb seal surface, then, given a rotor whirl amplitude approaching the honeycomb seal clearance, the surface will be worn to give an axial leakage path of constant clearance. Based on calculations using the seal code developed by Kleynhans and Childs (1997), such a clearance does not usually have significant negative direct stiffness. Note that the red line is nearly parallel with the blue and green ones, which are offset by the concentricity tolerance on the position of the honeycomb seal. The practical result is that the static equilibrium cannot be accurately calculated in this case, because the intersections within the tolerance band cover a range from where the drum is near the bottom of the honeycomb to being lifted above the centerline of the two bearings. The behavior of the rotor when it is near the asymptote can be understood by the concept of indifferent equilibrium. This concept applies to the case were all spring rates in the system are linear. Indifferent equilibrium is most easily seen by an example where the honeycomb seal (with negative direct stiffness) is at the center span of a symmetric rigid-rotor bearing assembly. In that case the asymptote will occur when the honeycomb stiffness is equal to minus the sum of the bearing stiffness (twice the stiffness of the two identical bearings.). Thus the rotor will have zero stiffness to parallel translation, because the negative honeycomb seal stiffness will cancel the positive bearing stiffness, as the three spring rates are additive because the springs act in parallel. (The resistance to angular displacement will remain.) At this asymptote, if the rotor moves a honeycomb seal with negative stiffness will pull it up, and the bearings will push down, giving a zero net force resisting the displacement. Therefore this rotor can be moved to any lateral position without applying an external force, assuming perfectly linear stiffness. (Nonlinear stiffness of the bearing or honeycomb seal may not cancel at all displacements.) When the solution for static equilibrium is at the asymptote, a rotor is indifferent to its lateral position, and can be moved to different positions with little or no force, provided the rotor drum does not hit the honeycomb bore. However, the forces on the bearings can be very large.
geometry of the impellers, diaphragm-impeller cavities and seals it is possible to calculate the resulting radial force by means of a static equilibrium analysis. Figure 13 shows a cross section of last wheel stage of a reinjection compressor, Example C.
Figure 13. Cross Section of Last Wheel of a Tested Compressor. The control volume used to make the static equilibrium of the impeller is highlighted in red color. For the analysis the authors assumed that the pressure gradient extends down to the diameter of inlet eye seal on the impeller cover, and down to the hub at the foot of the impeller disk. Pressure gradients have been considered constant along the two cavities. This assumption leads to make an overestimation of the forces. Figure 14 shows the result of the above analysis applied to a very high pressure compressor, Example, C for the full density full pressure conditions in the authors testing facility. In particular circumferential pressure gradients are evident.
INFLUENCE OF VOLUTE FORCES ON ROTOR STATIC EQUILIBRIUM

In contrast to the honeycomb seal, which is characterized as a spring rate, the volute forces can be represented as a static force on the rotor, independent of small displacements of the rotor. They can then be incorporated into the above static equilibrium model in the same manner as the weight of the rotor. As a simplification, the volute force can be applied at nearly the same location as the honeycomb seal force. However, obtaining a good estimate of the volute force is laborious. In order to calculate the force generated by the volute a complete computational fluid dynamics (CFD) model of the last impeller plus diffuser plus discharge volute has been developed. The analysis has been carried out using the well-known mixing plane approach. According to this technique the complete model has been divided in two parts: first the impeller plus the first part of the diffuser, second the diffuser plus the discharge volute. First analysis is done on the impeller by imposing the boundary conditions at the inlet plus the mass flow. Once this calculation is completed total pressure and total temperature distributions as well as yaw angle are extracted at the interface plane and are used as boundary conditions for the CFD analysis of the volute. This analysis also requires imposing the mass flow (the same value applied for the CFD analysis of the impeller). The analysis is able to capture the static pressure gradients in the tangential directions close to the outlet of the impeller. By knowing the pressure distribution at the outlet of the impeller as well as the
Figure 14. Static Pressure Distribution Around the Last Impeller. Figure 15 is a graph showing the static pressure distribution immediately at the discharge of the impeller and used in order to compute the radial force created on the rotor. For this case the pressure equilibrium of the rotor control volume leads to a horizontal force of 9561 N (2149 lbf) and vertical force of 9663 N (2172 lbf). The sum of these forces acts at an angle of 45 degrees from vertical, as shown in Figure 16. The weight of the rotor is 350 kg (771.62 lb), i.e., the vertical force from the volute is around three times the weight. The negative value means that the force is in the same direction of the weight.
Table of Contents
After the honeycomb seal was modified to have positive stiffness the compressor was retested. The measured bearing journal positions are shown in Figure 17. The journal in the nondrive-end (NDE) bearing did not lift high but is pushed to the side. The journal in the drive-end (DE) bearing (next to the honeycomb seal) still lifted to the top pads. The volute forces were not calculated, as the effort was not justified by the good behavior of the compressor.
Figure 15. Pressure Profile around Circumference of Last Impeller.
Figure 17. Calculated Static Force by Bearings for Example B, before Modification of the Honeycomb Bore. Figure 18 shows the calculated static force on the rotor by bearings for Example B before modification of the honeycomb bore. The operating conditions for Figure 18 are estimated to correspond to Figure 4 at maximum speed. The offset used corresponds to the green line in Figure 12. The vertical line labeled expected in Figure 18 represents the Keff shown in the above table for the rated conditions. At the expected Keff, the forces at both bearings are negative, pushing down on the rotor, in the same direction as shown in Figures 4 and 5. Note that the bearing force is large, being nearly twice the rotor weight on bearing number one. Because the volute forces are not considered, the calculation is not empirically confirmed. However, the calculated and measured journal conditions are not contradicted either.
Figure 16. Direction of Net Force Due to Pressure. Although this particular example is not correlated to the test data, is serves to demonstrate that the volute forces can dominate rotor weight for compressors in this size and pressure range.
COMPARISON OF CALCULATIONS WITH TEST DATA

The honeycomb seal stiffness values used in this paper were all calculated with the seal code developed by Kleynhans and Childs (1997). For Examples A and B, using a diverging clearance (leakage area increasing going downstream) in this seal code gives negative direct stiffness. A converging clearance gives positive direct stiffness of the honeycomb seal. The honeycomb seal for Example B was modified after the measurements in Figures 4 and 5 were made. The modification was made when an opportunity arose to limit the possibility of high bearing loads. Swirl brakes were used on the honeycomb seal to avoid subsynchronous instability. Using the operating (hot and spinning) clearance of the honeycomb seal for Example B, the calculated direct stiffness at zero frequency before and after changing the clearance is shown in Table 3 for the respective clearances. Table 3. Calculated Direct Stiffness at Zero Frequency Before and After Changing the Clearance.
Figure 18. Journal Positions in the Bearings of Example B after Revising the Honeycomb Seal. Figure 19 shows the calculated force on the rotor by the bearings for Example B as modified. The offset used corresponds to the blue line in Figure 20.The vertical line labeled expected represents another value of the Keff = +0.259 E9 N/m (+1.43 E6 lbf/in) for the as-modified conditions. At the expected Keff, the force from Bearing 1 (on the DE of the rotor) is pushing down, as shown on
Table of Contents
the right-hand side of Figure 19. At the expected Keff, the force from Bearing 2 (on the NDE of the rotor) is pushing down slightly, as shown in the right-hand side of Figure 19. Note that the Bearing 2 force is small, with its absolute value being less than half the rotor weight. Without the honeycomb seal force or volute forces, the bearing force would be about half the rotor weight, pushing up. The graphical solution predicting the rotor equilibrium as modified is shown in Figure 20.
Figure 19. Calculated Static Force by Bearings for Example B, after Modification of the Honeycomb Bore.
Figure 20. Graphical Solution Corresponding to Figure 19. Figures 19 and 20 are consistent between the calculation and test, assuming a known offset (tolerance) of the honeycomb bore. However, the actual offset is not known. Note that the offsets are opposite between Figures 19 and 18. The actual offsets were unknown. Alternate compressor internal parts (rotor and bundle) were installed for the test shown in Figure 17 than were used in the test shown in Figure 4. Therefore, the offsets may have changed between Figure 18 and 19.
preload and tight clearances on tilt-pad bearings to make the bearing less sensitive to load. The disadvantage of high preload and tight clearances is loss of opportunity to maximize the bearing damping and thus optimize unbalance response and rotordynamic instability. The measured rotor position in the bearings of the examples was often against the top pads, in opposition to the standard assumption of gravity load. A more unexpected conclusion, from the equations of static equilibrium, is that a honeycomb seal having a negative static stiffness in the same range as the rotor-bearing system stiffness (that is, at the asymptote) can produce bearing loads of many times the rotor weight. These calculations suggest that such a load would only be bounded by contact of the rotor against the honeycomb (based on a linear stiffness characterization of the honeycomb). However, the contact force will be small due to the indifferent equilibrium effect discussed above. (The large force on the rotor in the honeycomb is carried by the gas pressure.) In the cases presented, the calculations predict the asymptote was not reached. Only moderate contact was noted on disassembly. The solution of the six equations of equilibrium (Equations 6 through 11) is straightforward and well based on the theory of statically indeterminate problems (Popov, 1968.) However, conceptually graphic solution gives more insight into the behavior of the system. The concept of static equilibrium could also be applied to honeycomb seal test rigs used for determining the dynamic coefficients. In the case of test rigs where the seal carrier moves with respect to a stiffly supported rotor, the direct stiffness of the tie-rod assembly could be used in the calculation in place of the rotorbearing system stiffness. When testing seals with negative direct stiffness contact with the rotor may occur. The graphical solution presented in this paper shows such contact will occur when the absolute value of negative stiffness of the honeycomb approaches the positive stiffness of the mechanical parts of the system (when approaching the asymptote). This is conceptually different from reaching contact only at very large values of negative stiffness. The test results presented prove that the bearing loads are not due to gravity alone, as it is customary to assume. The calculations presented here confirm that negative stiffness from the honeycomb seal, and volute loads should have a dramatic effect on the static equilibrium of the rotor on its bearings. Due to strong influence of the unknown magnitude and direction of offset between the honeycomb bore and the bearing centerline, the calculations suggest the bearing loads are not calculable, given typical offset tolerances. Furthermore, the aerodynamic force on the rotor can vary so that the estimated bearing loads may vary over a wide range. Indifferent equilibrium further confounds the problem. Nevertheless, the analysis presented requires that the compressor design must take the variability of the possible bearing loads into consideration. For oil film bearings this is easily considered in the design stage by calculating the loads at the extremes of offset. Magnetic bearings and bearings supported on squeeze film dampers typically have direct stiffness that is an order of magnitude softer than the tilt-pad journal bearings used in Examples A and B. If magnetic bearings or damper bearings are used with a honeycomb seal, especially one having negative stiffness, the calculated asymptote would occur at a much smaller negative stiffness of the honeycomb seal than with stiffer bearings. Such behavior could disrupt the functioning of a magnetic bearing or damper bearing even if the honeycomb negative stiffness were relatively small in absolute value.
CONCLUSIONS
Calculation of the bearing load based on only the rotor weight (for horizontal rotors) is not valid for high pressure centrifugal compressors having honeycomb/drum-rotor seals, or single-volute or collector type discharge hardware. The forces caused by these components are shown for the above examples to exceed the rotor weight.
DISCUSSION
The basic assertion of this paper is that honeycomb seal and volute forces in reinjection compressors can invalidate the bearing load calculations typically used in rotordynamics analysis to find the bearing characteristic. A practical solution is to use high
Table of Contents
Introducing the honeycomb/drum type seal into the bearing load calculation as direct and cross-coupled spring coefficients makes the problem statically indeterminate and thus requires consideration of displacement of the journal in its bearing, the drum in the honeycomb bore, and the rotor as a beam in bending. In this case the bearing loads can be solved by the standard methods of the mechanics of solids. When the spring coefficient of the honeycomb seal is negative, the indeterminate beam calculation can show asymptotic behavior. In this case the drum displacement will only be limited by contact with the honeycomb seal bore. At the asymptote, the rotor can be either lifted or depressed. This paper introduces a conceptually graphical solution to the indeterminate beam calculation for the purpose of visualizing the asymptotic behavior (here shown in a vertical plane.) This solution also makes clear the importance of small deviations of the honeycomb bore from the centerline between the two bearings. The asymptotic behavior does not depend merely on the magnitude of the negative stiffness of the honeycomb seal, but instead depends on the ratio of that stiffness to the stiffness of the rotor bearing system. In the graphical solution this occurs when the slope of the honeycomb seal displacement versus force line is nearly parallel with the slope of the rotor-bearing system displacement versus force line. This solution predicts the rotor can be lifted in its bearings by the negative stiffness of the honeycomb seal, at sufficiently high gas pressures. This solution also predicts that the rotor can be lifted in its bearings by positive stiffness of the honeycomb seal, when the honeycomb bore is offset above the centerline between the two bearings. For examples given in this paper, these predictions are not inconsistent with the observed behavior. However, the unknown values of honeycomb bore concentricity defeat an exact calculation, because the bearing position is acutely sensitive to the honeycomb bore offset. The volute forces compound this problem.
K N/m k C c x1, x2 y1, y2 Xdot Ydot Kyy = K Lb Kcrit Keff Subscripts 1 2 3
= Radial spring coefficient of linearized restoring direct force N/m (lbf/in) = Tangential spring coefficient of linearized cross-coupled force N*s/m (lbf*s/in) = Damping coefficientdirect N*s/m (lbf*s/in) = Damping coefficient cross-coupled microns (mil) = Horizontal displacement of rotor in bearing 1 and 2, respectively microns (mil) = Vertical displacement of rotor in bearing 1 and 2, respectively mm/sec (in/s) = Horizontal velocity of whirl orbit mm/sec (in/s) = Vertical velocity of whirl orbit N/m (lbf/in) = Bearing direct stiffness in the vertical direction mm (inch) = Lcl N/m (lbf/in) = Honeycomb seal stiffness at asymptote for the rigid rotor, linear bearing case N/m (lbf/in) = Direct stiffness of the honeycomb seal at zero frequency
(lbf/in)
= Bearing 1 = Bearing 2 = Drum of honeycomb seal
REFERENCES
Ayres, F. Jr., 1958, First Year College Mathematics, New York, New York: McGraw-Hill Book Company, p. 28. Camatti, M., Vannini, G, Fulton, J., and Hopenwasser, F, 2003, Instability of a High Pressure Compressor Equipped with Honeycomb Seals, Proceedings of the Thirty-Second Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 39-48. Kleynhans, G. W. and Childs, D. W., 1997, The Acoustic Influence of Cell Depth on Rotordynamic Characteristics of SmoothRotor/Honeycomb-Stator Annular Gas Seals, Journal of Engineering for Gas Turbines and Power, 119, pp. 949-957. Popov, E. P., 1968, Introduction to Mechanics of Solids, Englewood Cliffs, New Jersey: Prentice-Hall, Inc., Chapter 12, Statically Indeterminate Problems. Thomas, G. B., Jr., 1968, Calculus and Analytic Geometry, Fourth Edition, Reading, Massachusetts: Addison-Wesley Publishing Company, p. 317.
NOMENCLATURE
fb1 fb2 Wr Lcg Lhc Lcl Fbx(x1,y1) N (lbf) N (lbf) N (lbf) mm (inch) mm (inch) mm (inch) N (lbf) = = = = = = = Vertical load on bearing number 1 Vertical load on bearing number 2 Vertical force due to rotor weight Horizontal distance from bearing 1 to the rotor center of gravity Horizontal distance from bearing 1 to the rotor center of the seal Distance between horizontal centers of the two bearings Force by the bearing on the rotor in the horizontal direction due to displacements x1 (direct) and y1 (cross-coupled) Force by the bearing on the rotor in the vertical direction due to displacements y1 (direct) and x1 (cross-coupled) Force by the honeycomb seal on the rotor in the horizontal direction due to displacements x3 (direct) and y3 (cross-coupled) Force by the honeycomb seal on the rotor in the vertical direction due to displacements y3 (direct) and x3 (cross-coupled)
BIBLIOGRAPHY
Mathcad, 2007, Version 14, Parametric Technology Corporation, Needham, Massachusetts. Wagner, N. G., 1999, Reliable Rotor Dynamic Design of High Pressure Compressor Based on Test Rig Data, Transactions of the ASME, Journal of Engineering for Gas Turbines and Power, October 2001, 123, pp. 849-856.
Fby(x1,y1)
N (lbf)
Fhx(x3,y3)
N (lbf)
ACKNOWLEDGEMENTS
The authors thank General Electric Oil & Gas, and ExxonMobil Research & Engineering for permission to publish this paper. Many thanks also to Luca Poli for his support and contributions, and also to Vittorio Michelassi in particular for the contribution given in the aero CFD analysis.
Fhy(x3,y3)
N (lbf)
Table of Contents
RECIPROCATING COMPRESSOR CONDITION MONITORING

by Steven M. Schultheis
Senior Engineer
Charles A. Lickteig
Senior Engineer Shell Global Solutions (US) Inc. Houston, Texas
and Robert Parchewsky

Principal Engineer Shell Global Solutions International Calgary, Alberta, Canada
Steven M. Schultheis is a Senior Engineer in Shell Global Solutions (US) Inc., in Houston, Texas. His primary responsibilities are to support Shell worldwide in regard to condition monitoring and vibration analysis of rotating machinery. In addition, his specialties are in the areas of machinery and piping dynamics, rotordynamics, structural vibration, and pulsation. Mr. Schultheis received a B.S. degree (Mechanical Engineering, 1984) from New Mexico State University, is a registered Professional Engineer in the State of Texas, and is certified Category 4 vibration specialist through the Vibration Institute. He has published and presented at meetings of a number of technical societies and symposium including the Texas A&M Turbomachinery Symposium, the Vibration Institute, the NPRA, the ASME, and others. Charles A. (Chuck) Lickteig is a Senior Engineer in Shell Global Solutions (US) Inc., in Houston, Texas. He is responsible for providing technical support for rotating and reciprocating machinery at Shell affiliated companys worldwide as well as commercial customers. Mr. Lickteig is currently developing advanced performance and condition monitoring applications. Since joining Shell in 1990, he has had assignments in projects involving specification, evaluation, systems integration, installation, commissioning, and startup of rotating equipment for existing and new process plants, as well as extensive field troubleshooting in performance and vibration analysis. Mr. Lickteig has a B.S. degree (Mechanical Engineering, 1990) from the University of Kansas, and is a registered Professional Engineer in the State of Louisiana. Robert Parchewsky is a Principal Engineer in Shell Global Solutions International, in Calgary, Alberta, Canada. He is
107
responsible for managing the implementation of global standards and systems on condition monitoring of rotating equipment for remote diagnostics. He has worked in a variety of business units at Shell including Chemicals, Oilsands, Refining and Exploration, and Production. Mr. Parchewsky is currently developing advanced performance and condition monitoring applications. He has provided support on new project design, troubleshooting, failure analysis, condition monitoring, performance assessments, standards development, and field support. Mr. Parchewsky received a B.Sc. degree (Mechanical Engineering, 1988) from the University of Saskatchewan, and is a registered Professional Engineer in the Province of Alberta in Canada.
ABSTRACT
Technology for reciprocating compressor condition monitoring has been around since the 1950s. However until the last 15 years or so it seemed that only the pipeline companies spent much effort on this activity. Technology has advanced, and there are very effective approaches to monitoring and protecting reciprocating compressors on the market today. While pipeline operations are pulling out their reciprocating compressors, this machine is still the workhorse of refineries, chemical plants, and oil production facilities. As a result a new generation of interest has developed in effective condition monitoring of reciprocating compressors. This paper will discuss risk-based decision making in regard to measurements and protective functions, online versus periodic monitoring, proven and effective measurement techniques, along with a review of both mechanical- and performance-based measurements for assessing machine condition. Case histories will also be presented to demonstrate some of the concepts.
INTRODUCTION
Each year at the Turbomachinery Symposium in the reciprocating compressor discussion group the focus of the discussion is primarily on condition monitoring. With all the other technical issues related to reciprocating compressors that could be discussed, this is usually the topic that generates the most interest. Past topics have included vibration monitoring, rod drop monitoring, pressure-volume analysis, and temperature measurements. In these discussions there are several thematic questions that have come out:
Table of Contents
How is condition monitoring of reciprocating compressors justified? What measurement techniques are really effective? Should these parameters be used as protective functions or not? Should the measurements be online or periodic? What can I do easily to improve the monitoring on my existing Is it really worth it?
All legitimate questions, but unfortunately there is not one answer for all equipment. Based on risk, a compressor in a refinery hydrotreater service may require a very different condition monitoring approach than a gas lift machine in an oil field application, or a nitrogen compressor in a chemical plant. Sorting out the risk and applying the right monitoring approach is the primary objective of any conditionmonitoring project.
machine, especially without doing a major retrofit project?
AN OVERVIEW OF PROVEN MEASUREMENT TECHNIQUES

For centrifugal compressor trains there is good agreement across industry on how to effectively monitor machine condition. These approaches are summarized in American Petroleum Institute (API) Standard 670 (2000), and there is little question as to what suite of instrumentation will be used to monitor centrifugal machines. When it comes to reciprocating compressors in process plants, there is much less agreement on which monitoring techniques should be standard but at least API 618 (1995) contains some basic requirements. For an ISO 13631 high speed reciprocating compressor, particularly in oil field service, there is even less agreement in the industry on applicable monitoring systems. API Standard 618 (1995) monitoring and protection requirements include high discharge temperature, low frame lube-oil pressures and level, cylinder lubricator system failure, high oil filter differential pressure, high frame vibration, high level in the separator, and jacket water system failure. API Standard 670 (2000) describes the requirements for installing proximity and casing transducers on reciprocating compressors, but the details of what measurements to make, and how to apply those measurements are left to the original equipment manufacturer (OEM) and the purchaser to decide. The following are some techniques that have been proven effective across a wide range of applications. Vibration There are two primary vibration measurements that have been proven effective; measurement on the crankcase, and measurement on the crosshead/distance piece. This is due to the way forces are applied in reciprocating machines. The most common machine design is a balanced opposed configuration and in this configuration the reaction forces in the cylinders are balanced across the machine by the opposing cylinder. Since the cylinders are offset, moments are also set up in the crankcase, and these moments are balanced as much as possible by cylinder placement and cylinder timing. The balance of forces and moments is fine-tuned by designing the weight of reciprocating parts and by applying counterweights on the crankshaft. But it is rare for the balance to be perfect and if a malfunction occurs that upsets this balance of forces and moments, the result is high vibration at 1 or 2 running speed. The flip side is that if the machine support stiffness is reduced, for instance due to grout deterioration or the loosening of foundation bolts, then even in the presence of normal forces and moments the vibration will increase. Catastrophic events such as breaking a piston rod or losing a counterweight result in a sudden increase in unbalanced forces and moments, and may result in very high crankcase vibration.
Figure 1. Crankcase Vibration Transducers. Crankcase vibration has been used as a basic protection parameter for decades, usually with a mechanical earthquake switch. API 618 (1995) has specifically eliminated the mechanical switch as an option because of instrument reliability issues. This measurement is now mostly made with solid-state electronic devices. Most compressor OEMs specify a crankcase velocity alarm/shutdown level and the most common configuration is to mount the transducers on each end of the crankcase about halfway up from the baseplate in line with a main bearing (Figure 1). This configuration enables the transducers to see the effects of both unbalanced forces and moments while minimizing the number of measurements. It is then also possible to monitor changes in relative vibration amplitude and phase in the event that grout breaks down or forces and moments become unbalanced in the machine. An alternate approach is to install a transducer at each main bearing, but this is much less common. The advantage of this approach is that information is available online regarding the operating deflection shape of the crankcase. Since the purpose of these transducers is to measure running speed related vibration, a low frequency transducer is required, and velocity is the normal measurement parameter. The crankcase vibration acceptance criterion used by most OEMs is also in velocity. An option is to use a low frequency accelerometer and run the signal to a dual path monitor, allowing both the measurement of low frequency running speed related vibration as well as the high g excursions associated with a major impact event in the crankcase. The other primary vibration measurement that has proven to be effective is to measure acceleration on the crosshead or distance piece of each cylinder (Figure 2). In this case the idea is to measure the mechanical response of the assembly to impact events. Malfunctions such as liquid carryover, loose piston nuts, loose crosshead attachment valve problems, clearance problems, and many others can be identified with this measurement. For more than 100 years mechanics have been listening to the knocks and rattles of reciprocating compressors to assess the health of their machines. An accelerometer on the distance piece of a cylinder is in essence a microphone that allows those sounds to be recorded, and allows for consistent alarms and shutdowns. Of all the vibration measurements that could be made, this is probably the most effective vibration protection measurement available. If a machine is undergoing catastrophic distress, it will typically be picked up on the crosshead accelerometers. As a result, even for very small, spared, or noncritical compressors in hydrocarbon service a simple (and cheap) accelerometer measurement on the crosshead or distance piece of the machine is easily justified.
Table of Contents
RECIPROCATING COMPRESSOR CONDITION MONITORING 109
discharge temperature. Valve temperatures have proven to be valuable in identifying individual valve problems, but are most effective if the measurement is made in a thermowell in the valve cover (Figure 4) so that the reading is taken close to the valve plates or poppets, rather than measuring contact temperature of the valve covers. Packing case temperature or packing leak off temperature can give an indication of packing leakage, while main bearing temperature measurement has been proven effective in preventing major damage due to main bearing failures. Many machines have been saved by shutdowns associated with eutectic or turkey popper temperature devices in wrist pin and big end bearings, but there are technology developments currently underway to make this a wireless measurement allowing for a temperature trend. Fossen and Gemdjian (2006) describe one such technology in their paper on radar-based sensors. This and other similar technologies are a major improvement in the protection and condition assessment of connecting rod bearings.
Figure 2. Crosshead Accelerometer Installation. (Courtesy of Bently Nevada) An extension of vibration technology is ultrasound analysis. Ultrasound has proven to be the preferred approach to analysis of valve condition. Ultrasonic energy is most often associated with gas leaks, so a valve that leaks is a strong generator of ultrasonic energy. Ultrasound measurements are usually taken in conjunction with compressor pressure-volume analysis, which will be discussed later (Figure 3).
Figure 4. Valve Temperature Measurement, Thermo Wells in the Valve Covers. Rod Drop/Rod Runout In their paper, Rod Drop Monitoring, Does it Really Work, Schultheis and Howard (2000) discuss the significant problems associated with rod drop monitoring. While advances have been made to this technology since then, it is still a difficult measurement with many sources of error. At the same time, rider band wear and failure is still a significant maintenance issue with many machines and as a result rod drop monitoring still has a place in the reciprocating compressor condition monitoring toolbox (Figure 5).
Figure 3. Pressure and Ultrasound Traces for Valve Analysis. Critical to successful implementation is properly set trip levels that are just high enough over the normal operating level to react to mechanical failures, but not so high as to miss the failure prior to catastrophic release. Unfortunately, even when systems are properly configured, catastrophic failures often still occur due to operations restarting the machine to confirm that the trip was real (Case History 1). In these cases when restarted the failure is escalated, leading to a large consequence failure. Proper operating procedures need to be in place to ensure that reciprocating compressors are not restarted without proper inspection and engineering assessment. Temperature Machine temperatures are a valuable indication of machine condition and are a primary tool for reciprocating compressor condition monitoring. The primary temperature measurements include cylinder discharge temperature, valve temperature, packing temperature, crosshead pin/big end bearing temperature, and main bearing temperature. Cylinder discharge temperature is one of the protection parameters recommended by API 618 (1995) since leaks in rings and valves result in recompression of gas that will raise the
Figure 5. Rider Band Failure.
Table of Contents
An underutilized parameter however is rod runout monitoring. Rod drop measurement is accomplished by using the direct current (DC) component of the proximity probe signal, which is proportional to position. This locates the average distance between the probe tip and the target. In a reciprocating compressor proximity probes are typically located under the piston rods, and are used to measure the rod position, which can be converted to rider band wear (Figure 6). There is also an alternating current (AC) component of the probe signal, which is representative of operating rod runout. Cold runouts are usually held to about 2 mils peak-to-peak, and operating runouts under normal conditions are typically somewhat greater than that, on the order of 2 to 6 mils peak-to-peak. In the event of a malfunction such as a cracked piston rod attachment, a broken crosshead shoe, or even a liquid carryover to a cylinder, the operating rod runout will increase significantly. These are data that have historically been ignored, since the rod drop monitors on the market were only measuring the DC component of the signal. Paralleling the signal from rod drop probes into a vibration channel, or configuring the rod drop monitors to display rod runout in addition to rod drop, adds tremendous value to the condition data set on a reciprocating machine.
Figure 7. PV Measurement Valves on the Side of a Cylinder.
Figure 8. Permanent Pressure Transducer Installation Through a Valve Cover. (Courtesy of Bently Nevada)
EFFECTIVELY USING THE DATA ALREADY AVAILABLE

Figure 6. Rod Drop Probe Installation. (Courtesy of Bently Nevada) PV Analysis Pressure velocity (PV) analysis is a technique that has proven to be very effective in assessing the condition of reciprocating machinery and has been used for more than 50 years. Personal computer technology has significantly reduced the cost of this kind of measurement, and improvements in transducer technology have overcome the technical obstacles such that PV analysis is now available online. Dynamic pressure transducers are used to measure the pressure inside the cylinder over the course of the stroke (Figure 3). This allows the analyst to evaluate the condition of the rings, valves, and packing, while at the same time calculating the dynamic rod load, which is the source of the forces and moments described in the discussion on vibration. This requires that a pressure transducer be installed in the cylinder, either on a temporary basis using a valved port (Figure 7) or for an online measurement using a permanent transducer installation (Figure 8). As described in the next section, it is often possible to predict the PV diagram from existing instrumentation and modeling. In many reciprocating compressor applications, a tremendous amount of data is gathered as a matter of course, and while they are displayed to the operator through a distributed control system (DCS) or other control system it is otherwise wasted. In many if not most applications, those data are stored, most often in a data historian, but this is functionally like the black box on an aircraft; data gathered and stored so that they are available in the event of a problem. Those data are in essence underutilized and since they are available they should be used as a part of the condition monitoring for a compressor, rather than leaving them to collect dust until a problem arises. One approach to utilizing these data that is proving successful is to use a modeling program that compares the theoretical performance of the compressor to the actual performance in order to determine if the compressor performance is deviating from predictions. Such a model can easily reside on the same computer with the data historian, gathering its input from that database, and writing results back to the historian. The modeling program is first populated with design data and physical properties of the compressor, e.g., piston diameter, stroke, rod diameter, etc. Then the model is validated against predicted performance information. Next the model is integrated into the network where the data reside and data on the actual process conditions are fed
Table of Contents
into the model to determine the actual performance of the machine. For a given set of process conditions the theoretical and actual performance of several compressor performance variables can be calculated. Finally performance and deviation from theoretical are written back to the data historian for trending. Software alarms on the calculated values can also be generated to alert operations or maintenance of developing performance issues. The theoretical model uses first principal type equations to calculate: volumetric flow rate, volumetric efficiencies, power, rod loads, interstage pressure, outlet temperatures, and rod reversal. Each of these values for each cylinder or stage is then evaluated against the design or actual values to determine if the performance of the compressor is deviating from prediction. Deviations generally indicate degradation in cylinder wear parts or off-design operating conditions. There may be accuracy issues associated with the field instrumentation. However trending performance deviations from a baseline can be as valuable as absolutely accurate performance data. While it is true that a direct measurement of the compressor pressure volume trace utilizing a dynamic pressure transducer gives the most accurate picture of cylinder condition, a performance calculation based on the pressures, temperatures, and flows already available in the process control system provides a valuable overview of the condition of the machine. Establishing a baseline of performance when it is known that there are no problems with the machine makes it possible to identify deviations from normal, alerting operations and maintenance to a developing problem, and perhaps giving operations a chance to take action to mitigate the problem before it causes a machine failure.
Figure 9. Risk Assessment Matrix.
CASE HISTORIES
Case History 1 A 1500 kW compressor was installed in 2002 that was deemed critical enough to justify an advanced monitoring and data acquisition system with remote monitoring and diagnostic capability. After 3000 running hours a failure occurred at the crosshead for cylinder 1. The machine had experienced a four-hour shutdown after a power failure and when this was resolved a restart was initiated. Shortly after startup a trip occurred due to vibration/impact at the accelerometer sensor for cylinder 1. The weather was cold and stormy (0F), and after examination of the crosshead the failure was obviously a matter of poor lubrication. This compressor had one common oil supply line to the top and bottom crosshead guides and based on the design there was certainly a preference in the oil supply to the top. That issue, in combination with the low temperature and associated viscosity changes in the oil along with startup conditions, resulted in a lack of oil supply to the crosshead and the resultant failure (Figure 10). The compressor tripped on impact level set in the online monitoring system, which prevented the machine from more extensive damage and longer outage time. However, the system showed a change earlier in both vibration and temperature. Unfortunately no response was taken to this change in vibration level and temperature increase until the failure occurred (Figures 11 and 12).
A RISK-BASED APPROACH TO CONDITION MONITORING

The most common objections to condition monitoring on reciprocating compressors include: We never needed it before, why now? We dont have problems with these machines, why monitor them? Monitoring is too expensive. All of this may be true, it may not be necessary, the machine may not have problems, and a complete system may be expensive. However if the compressor is moving hydrocarbon gas, it becomes easy to justify at least a simple accelerometer-based vibration system costing on the order of $1000 USD, to shut down the machine in the event of a catastrophic event and to install a performance calculation tool using existing instrumentation and databases. Unfortunately purchasing a condition monitoring system is much like buying an insurance policy. If there is never a failure then there is really no need to have it. One thought to consider is that at least 10 percent to 20 percent of all reciprocating compressors (based on the authors experience) suffer a catastrophic failure or a failure that could have been catastrophic if protection systems did not stop the event from progressing. In buying insurance, or in buying condition monitoring, the first step is to assess the risk, and then purchase what is appropriate to mitigate the risk specific to the machine and the service. The outcome of the risk assessment should be a list of parameters that will be used to protect the machine, as well as parameters for determining machinery condition. The approach to risk assessment is usually a risk matrix that includes aspects of safety, business impact, environmental impact, and reputation impact. Figure 9 is an example of such a risk matrix. The outcome from this kind of analysis can be either a level of criticality, a safety integrity level, or a standard monitoring approach. One of the ways this type of analysis can be used is to determine if monitoring should be online or periodic. Machines with high criticality ranking will typically require online monitoring and protection. Machines that are spared, and thus represent significantly less risk, may only require minimal protection with periodic monitoring. Machines with high criticality ratings may not only justify a complete set of monitoring and protection instruments, it might also make sense to establish the information technology (IT) infrastructure for remote monitoring and diagnostics.
Figure 10. Damage to the Lower Crosshead Guide.
Figure 11. Vibration Trend Prior to Failure (About 1g).
Table of Contents
Figure 12. Vibration Trip on 9 January, Spiked up to 10g or More. Case History 2 At a Canadian gas plant, a series of reciprocating compressor failures occurred due to yielding of the aluminum pistons from high internal pressure loads resulting in rod failures (Figure 13). The compressors were in acid gas service with high molecular weight gas, which resulted in high valve losses. By instituting online performance monitoring of the compressor through the use of a compressor performance model interfaced to the data historian, and by highlighting to operations the appropriate operating envelope based on the results of the performance calculations, repeat failures have been eliminated. The same viewer was used for the data historian that the operators were already using for other tasks, so there was little training required to make this tool an effective operator interface (Figure 14). Online monitoring utilizing existing instrumentation and requiring no additional capital expenditures or machine modifications resulted in multimillion-dollar savings as the failures were eliminated.
Figure 15. Routine Crosshead Vibration Trace Shows Development of the Vibration as the Crosshead Crack Progressed.
Figure 16. Broken Crosshead Slipper. Case History 4 Analysis identified a leaking crank-end discharge valve as the reason for decreased performance. Inspection of the discharge valve during replacement showed what looked like pieces of piston ring or rider band material in the valve. The piston was removed from the cylinder and it was confirmed that the piston rings were broken into segments and the edges of the rings were broken (Figure 17). The pressure time (PT) trace (Figure 18) shows the crank-end discharge valve, 1CD3. Ultrasonic data indicated this valve leaking during the suction portion of the stroke. Although the piston rings were modified by the failure to have more than one end gap, they stayed in the piston ring groove, between the piston and the cylinder, and performed their intended sealing purpose. There was probably more bypass than intended, but it was not noticeable in the flow rate.
Figure 13. Piston Rod Failure.
Figure 14. Data Display-Rod Load Curves. Case History 3 A broken crosshead slipper was identified in the periodic data collection on the machine. During regular PV data gathering, the analyst also routinely gathered vibration data on the crosshead. Figure 15 shows the vibration trace of the cracked crosshead slipper on the top, along with two historical traces. The vibration signal on this crosshead has changed significantly, and the predominant feature is the impact ring down seen in the waveform between 240 and 300 degrees. The crack in the slipper is shown in Figure 16.
Figure 17. Broken Piston Rings Laid out on the Cylinder Support Pedestal During Disassembly.
Table of Contents
such as valves, rings, and packing, some form of performance analysis is key to the early identification of cylinder problems, allowing for maintenance cost reduction through proactive maintenance planning. When properly configured and utilized an online condition monitoring system that is tailored to the criticality of the equipment can significantly reduce the likelihood of catastrophic failures, and has the capability to create proactive action from operations, while being cost effective and repeatable. Online monitoring can be quite effective even if only utilizing existing instrumentation and vibration transducers.
REFERENCES
Figure 18. Pressure Time and Ultrasonic Trace Showing Crank End Discharge Valve Leak. Case History 5 Again crosshead acceleration measurement saves the day as it tracks the development of a loose piston nut. The acceleration trace (Figure 19) changed significantly between the May and June measurements, and there is a distinct impact/ring down pattern present after top dead center (TDC) and bottom dead center (BDC) indicating an impact as the load shifts from tension to compression and back. API Standard 618, 1995, Reciprocating Compressors for Petroleum, Chemical, and Gas Industry Services, Fourth Edition, American Petroleum Institute, Washington, D.C. API Standard 670, 2000, Vibration, Axial-Position, and BearingTemperature Monitoring Systems, Fourth Edition, American Petroleum Institute, Washington, D.C. Fossen, S. and Gemdjian, E., 2006, Radar Based SensorsA New Technology for Real-Time, Direct Temperature Monitoring of Crank and Crosshead Bearings of Diesels and Hazardous Media Reciprocating Compressors, Proceedings of the Thirty-Fifth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 97-102.
BIBLIOGRAPHY
Leonard, L., June 1996, The Value of Piston Rod Vibration Measurement in Reciprocating Compressors, Orbit Magazine, pp. 17-19. Schultheis, S., June 1996, Vibration Analysis of Reciprocating Compressors, Orbit Magazine, pp. 7-9. Schultheis, S. M. and Howard, B. F., 2000, Rod Drop Monitoring, Does it Really Work, Proceedings of the Twenty-Ninth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 11-20. Figure 19. Crosshead Acceleration Traces Showing the Development of a Loose Rod Nut. Crosshead Knocks after TDC and BDC (Ring Down). Smith, T. and Schultheis, S., 1997, Protecting and Managing Your Reciprocating Compressors, Proceedings of the 1997 NPRA Refinery and Petrochemical Plant Maintenance Conference, pp. 85-98.
CONCLUSIONS
Reciprocating compressor condition monitoring has developed well beyond theory to the point that there are a number of proven and practical approaches to monitoring and protection of these machines. The first step to effective condition monitoring is to take advantage of the information already available on the machine. From there a well thought out risk analysis should be used to guide the application of protective functions and condition analysis parameters, as well as determining whether monitoring should be online, periodic, or a combination of both. For normal wear parts
ACKNOWLEDGEMENTS
The authors would like to acknowledge the following people for contributing to the pictures and case histories: Dwayne Roberts, Shell Norco Chemical, Norco, Louisiana; Bill Robertson, Staff Machinery Engineer, Oilsands Projects, Shell Canada Limited; Bert van het Maalpad, Shell, Nederlandse Aardolie Maatschappij B.V., Netherlands; Brian Howard, GE Optimization and Control, Seattle, Washington; Steve Stoddard, Shell Chemical, Mobile, Alabama.
Table of Contents
Table of Contents
A PRACTICAL GUIDELINE FOR A SUCCESSFUL ROOT CAUSE FAILURE ANALYSIS

by David L. Ransom
Senior Research Engineer Mechanical and Materials Engineering Division Southwest Research Institute San Antonio, Texas
David L. Ransom is a Senior Research Engineer at Southwest Research Institute, in San Antonio, Texas. His professional experience over the last 10 years includes engineering and management responsibilities at Boeing, Turbocare, and Rocketdyne. His research interests include rotordynamics, structural dynamics, seals and bearings, finite element analysis, and root cause failure analysis. He has authored eight technical papers in the field of rotordynamics and thermodynamics. Mr. Ransom received his B.S. degree (Engineering Technology, 1995) and M.S. degree (Mechanical Engineering, 1997) from Texas A&M University. He is also a licensed Professional Engineer in the State of Texas.
they also represent organizational inability to successfully manage the competing interests of time, quality, and money. Therefore, in the interest of continuous improvement, it is in ones best interest to learn all one can from these failures, allowing us to avoid making the same mistake twice. The objective of this tutorial is to provide the reader with a practical guide for performing root cause failure analysis and determining the appropriate corrective/preventive action necessary to avoid the same failure in the future. The root cause failure analysis (RCFA) process begins with the collection phase, followed by the analysis phase, and concludes with the solution phase. Each of these phases is shown in Figure 1.
ABSTRACT
Root cause failure analysis is a process for identifying the true root cause of a particular failure and using that information to set a course for corrective/preventive action. From a technical standpoint, it is usually a multidisciplinary problem, typically focused on the traditional engineering fields such as chemistry, physics, materials, statics, dynamics, fluids, etc. However, it seems that too often the analysis stops with the technical aspects that are easily understood in an engineering environment, where the real root cause may exist in the human organization. In this tutorial, a practical guide to root cause failure analysis will be provided, followed by case studies to demonstrate both the technical and organizational nature of a typical root cause failure analysis.
INTRODUCTION
Despite the best efforts to avoid them, failures are still a common occurrence in every industry. Of course, there are the more obvious and well-publicized failures in the automotive, petrochemical, aerospace, and mining industries, just to name a few, but there are also many less catastrophic failures occurring at any point in time. Failures not only represent imperfection in the technical attempt to design and operate complicated systems, but
149
Figure 1. RCFA Flow Chart.
Table of Contents
Within collection, there are several key steps including team forming, problem definition, and of course data collection. The analysis phase is simply represented by determining the immediate, contributing, and root causes of the defined problem. The solution phase consists of determining corrective/preventive action, and then testing and implementation, which of course is the final step in RCFA. In each phase of the process, there are critical steps and simple guidelines to consider that will keep the investigation focused and practical. There are also some practical methods for organizing the investigation, depending on the size of the system under review. Finally, two turbomachinery related case studies are presented and discussed throughout the length of the tutorial to assist in demonstrating the overall process and guidelines.
questions or assist in the development of viable solutions. These expert team members do not need to be permanent members, and can be released once their contribution is complete. Their role is to support the investigation so that it is not halted for technical reasons.
CASE STUDIES
As stated in the introduction, two turbomachinery related case studies are threaded throughout the length of the tutorial, and are used to demonstrate steps and guidelines along the way. The first case study (Case 1) is from Alpha Company, who manufactures after-market parts for industrial turbomachinery. Just like every other day, an order comes in for the manufacture of a product for which drawings already exist in the engineering files. The engineering department pulls the drawings, selects the materials, and issues the work order to the shop. The shop, in turn, manufactures the hardware per the print and sends all the pieces to final inspection. During the course of final inspection, it is found that the pieces of the assembly do not fit. This is particularly troublesome for two reasons. First, this is a priority job and must ship immediately. Second, this is a product that has been manufactured in the past, so any defects in the design or manufacturing process should have been worked out in the past through the engineering change request (ECR) process. The second case study (Case 2) is from the Bravo Power Company, owner of several power generation gas turbines operating in combined cycle. After completing the spring outage (just in time for the summer heat wave), one of the gas turbines begins to exhibit a higher than normal temperature spread in the turbine exhaust. The unit is shutdown and inspected, and is found to have cracked crossfire tubes on one of the combustors. The damaged hardware is replaced, and the unit is returned to service. However, the temperature spread is still unacceptably high. A second unscheduled outage reveals that yet another combustor has cracked crossfire tubes. This time, the full set of combustors is replaced, and the unit is returned to operation without further anomalies.
Figure 2. Team Expansion for Either Technical or Organizational Limitations. The second reason to add team members is to increase the circle of influence of the team. As the investigation matures, it may become apparent that the real root cause lies outside of the current teams influence. For example, an engineering investigation may point to an issue in manufacturing. In such a case, it is important to add a team member that has the desired influence (i.e., manufacturing lead/manager), so that the investigation is not prematurely halted due to organizational boundaries. In Case 1 above, the natural team formed includes the engineering manager, two designers, one sales staff, and the quality manager. In this case, the natural team seems to have the ownership, technical knowledge, and the influence necessary for the resolution of this problem. In Case 2, the natural team is smaller, consisting only of the plant manager and the operations manager. Since neither of them has the technical knowledge necessary for the investigation, a third party consultant is also hired for the investigation. It is also possible to have an investigation team lead by an outside, independent investigator. This is most likely to occur when there is suspicion of overall organizational failure, and is typically imposed by a superior authority, such as senior management, industry regulator, etc. In such a case, the investigation may be lead by another division manager or a recognized industry expert. In this scenario, the people who would have otherwise been likely candidates for the natural team become technical team members, and will likely only be involved in the actual data collection phase. The next step in the collection phase is to define the problem. Defining the problem is a team activity, usually requiring some amount of brainstorming to come up with just the right definition. The quality of the investigation depends heavily on the quality of the problem definition. A good problem definition is short, simple, and easy to understand. In fact, if a problem statement is complicated, it merely reflects a poor understanding of the real problem. It is important that everyone on the team understands and agrees with the problem statement. The problem statement must also not be biased toward a specific solution. The consequence is the potential to either completely
STEPS TO ROOT CAUSE FAILURE ANALYSIS

As described in the introduction, RCFA is generally divided into three major phases: collection, analysis, and solution. Each of these steps is described in detail below. It is best to proceed through the phases as they are presented (i.e., one should not consider solutions until the analysis is complete), but it is not a one way path. There are plenty of reasons to possibly back up and repeat or revisit any step in the process before proceeding further. Collection Collection is used to describe all of the work necessary to prepare for the analysis phase. Naturally, the first step is to form a team that will participate in the RCFA. Team members should have ownership of the problem, and will therefore probably include engineers, technicians, operators, sales, management, etc. These team members are considered the natural team, as they have a first hand interest in the results of the RCFA. There are two main reasons why other team members might be added over the course of the investigation (Figure 2). First, it may be necessary to bring expertise into the team to help resolve key
Table of Contents
A PRACTICAL GUIDELINE FOR A SUCCESSFUL ROOT CAUSE FAILURE ANALYSIS 151
miss the real root cause, or at a minimum, miss some important contributing causes. In Case 1, the problem statement is determined to be, Why did the product fail to pass final inspection? On the contrary, if the team jumped immediately to what they intuitively determined as the root cause, the problem statement might be, Why didnt engineering update the drawings after the first manufacture? As will be shown later, this second problem statement prevents the team from identifying an important contributing cause. The final portion of the collection phase is the actual data collection. Table 1 is a listing of the most common sets of data that should be collected for an industrial turbomachinery failure analysis. Generally speaking, there are three common types of data: physical evidence, recorded evidence, and personal testimony. Table 1. Common Data Types.
In addition to the failed components, it may be important to the investigation to provide good, undamaged components for study as well. For example, in blade failure analysis, the most effective method for evaluating blade modes is to rap test a good blade. Undamaged parts can also be important for extracting geometric information to be used in a computer simulation (finite element analysis [FEA], computational fluid dynamics [CFD], etc.). Depending on the type of failure, it may also be important to capture physical evidence such as lube oil samples, water samples, air filter samples, deposit samples, etc. Generally, it is experience that will determine what other physical evidence needs to be retained. If there is doubt, it is certainly better to retain the samples. They can always be discarded once the investigation is over. Recorded evidence is the next significant type of data to be collected. Pictures are clearly necessary for the investigation. The tendency is to take too few pictures, because at the time, it seems impossible to forget what is being witnessed. However, experience will show that there cannot be too many pictures. There are two good concepts to keep in mind when taking pictures. First, for each detail picture, include a series of pictures that start from a very large view, and then gradually (perhaps three steps) zooms into the desired level of detail (Figure 4). This technique is vital to maintaining perspective and orientation.
The most critical aspect of collecting the physical evidence is to resist the urge to clean. Although it may seem desirable to provide clean, easy to handle samples to the various technical experts for review, the odds are that valuable data will be lost in the cleaning process. Figure 3 is an example of just such an occasion. The air compressor impeller and the stationary passage are contaminated with chlorides and sulphur, leading to the stress corrosion cracking (SCC) failure of the impeller blades. Cleaning these parts before the completion of the investigation will add uncertainty to the metallurgical analysis, as well as eliminate the evidence of the corrosive source (contaminated air). There are times when it is necessary to further damage evidence just to remove it from the scene. In this case, care should be taken to not impact the actual damaged portions of the evidence.
Figure 4. Photo Sequence Captures Orientation. Another important concept is to take pictures in orthogonal views, as if they are intended to be used as manufacturing drawings. Although isometric views are handy for seeing the overall layout, they are very difficult to scale. Orthographic views can easily be used as pseudo drawings, especially if there are at least three views recorded (front, top, and side). In Figure 5, the top photograph shows an isometric view of this small, auxiliary power unit (APU) gas turbine. Although this view is helpful to see where the fuel lines are located, it is very difficult to extract line dimensions from this view. On the other hand, the lower two photos provide the proper view, allowing for dimensions to be scaled, if necessary.
Figure 3. SCC Failure Due to Chloride and Sulfur in Air Stream.
Table of Contents
collect personal testimony will definitely have adverse effects on the quality of the testimony. Second, it is important to stay focused only on data collection, building a consistent timeline, etc. Any premature discussion of the cause of failure will likely adversely impact the interview process. It is up to the investigating team to resolve all conflicts in the data, whether it is in the personal testimony, in the operator logs, etc. Unfortunately, due to the human influence, none of the data sources will be pristine. But, by comparing all of the data, filling in the gaps, and resolving the conflicts, a clear and consistent picture of the failure can be obtained. Analysis The analysis phase is solely focused on using the collected data to build the cause chain and determine the immediate, contributing, and root causes of the failure. The immediate cause is typically the first one in the cause chain, thus directly leading to the failure. The root cause is the last one in the cause chain, while the contributing causes are the ones in between the immediate and the root. Although the process is referred to as root cause failure analysis, it is important to identify all of the causes. There are several common structures used in the analysis phase. The why chart is a simple series of questions that guides the team to the root cause. This is generally applied to small systems, or problems that do not span over to more than a couple of systems. This method is generally useful for most rotating machinery failures. The chart begins with the first problem statement, followed by the first answer to the problem statement. The questions are answered in small steps, which help to prevent missing any contributing causes. For Case 2, the why chart starts with the event question, Why did the gas turbine register an increased T5 spread? The rest of the chart is provided in Figure 7.
Figure 5. Orthographic Views Are Easier to Scale. The other forms of recorded data (operator logs, supervisory control and data acquisition (SCADA) logs, etc.) can be critical to the complete understanding of operating conditions at the time of failure. Figure 6 is an example plot from a pump SCADA system. These data are used to assist in a failure analysis of an overheating bearing. Since log data typically are dated, they are ideal for generating a timeline of events. However, due to the costs associated with data storage and management, high resolution data are often retained only for a short period, replaced by lower resolution data for long term storage. Therefore, it is critical to capture these electronic data as soon as possible.
Figure 6. Example SCADA Plot. The other important category of data to collect is the personal testimony. Theoretically, since everyone involved is discussing the same event, all of the various stories should converge. If the information from the various personnel does not agree, it may be a sign of multiple failures. Obviously, there is significant potential for finger pointing, or at least perceived finger pointing during this phase of data collection. To minimize this perception, it is important that the interviews be conducted by a rational, coolheaded person. Sending in an irritated and irrational person to
Figure 7. Case 2 Why Chart. At the conclusion of this RCFA, the immediate cause is determined to be uneven combustion, and the root cause is
Table of Contents
determined to be an assumption of service provider quality practices, not simply a technical flaw. This is important to note, as it drives the types of solutions that are considered in the next phase. Notice also that the contributing cause (poorly welded parts delivered by the service provider) lies outside of the current teams influence. At this point, it is recommended that the service provider be included in the team. The why chart is simple to apply, and will work for most of the turbomachinery related failures found in industry. For much larger systems, it may be more practical to use a fault tree. Fault trees are commonly used in the aerospace and nuclear power industries, since these systems are typically very complicated and much more difficult to investigate. Basically, the fault tree method requires that the team start with the fault, and then work backward to identify all possible causes of this fault. For large, difficult to understand systems, this provides a map for dividing up the investigation. As the investigation proceeds, teams gradually rule out each potential cause, and mark it off on the tree. By the end, there should remain a short list of potential root causes. Figure 8 is an example of a fault tree for Case 2. In this case, each of the events is preceded by either an AND or OR logic statement. For example, Uneven Combustion can be caused by either Uneven Fuel Delivery OR Uneven Air Delivery. On the other hand, the Poor Weld Quality is the result of both Improper Welding Technique AND Inadequate Quality System. Although this fault tree is incomplete, it demonstrates the level of detail required when using this approach. Each branch must be expanded and evaluated by the team. By closing out each branch of the tree, the actual failure cause chain becomes apparent.
Since this method is more complex and relies on a system of symbols (similar to a process flow chart), there are many commercial software packages available to assist in the process. Due to the size and complexity of the systems for which fault trees are used, the investigation is usually managed by an experienced investigator. Another popular structure for the analysis phase is the cause and effect diagram (also known as a fishbone diagram). Where fault trees are useful for complex systems, cause and effect diagrams are useful for incorporating cross-functional influences. As seen in Figure 9, the head of the fish is the problem to be investigated, and each of the main branches (bones) represents a specific functional area. To complete the fishbone diagram, the investigation team continues to list all of the possible connections each functional area might have with the failure. This format allows the team to see the overall picture and begin to focus the investigation as each of the functional branches is evaluated. In this case, it is clear from the beginning that the failure is contained within the engineering function, eliminating the need to further investigate the other branches.
Figure 9. Case 1 Cause and Effect (Fishbone) Diagram. There are some other important key points to remember during the analysis phase. It is helpful to keep these handy as the investigation proceeds, so that each team member is reminded of these guidelines.
Follow the dataThe most difficult aspect of the analysis phase
is avoiding preconceived notions regarding the root cause. It is up to the team members to protect each other from this trap. The investigation team must stick to the data and exclude gut feel from the investigation. both technical and organizational causesFinding the technical answer is often difficult, but the investigation should not stop there. Organizational influences can be just as significant and must also be included in the investigation.
Consider
Concentrate on analysisSave the problem solving for the next phase. The key at this point is to identify the immediate, contributing, and root causes.
operator or maintenance error?It is rarely actually an operator or maintenance craftsmen error. We all work in organizations with norms, procedures, and external pressures. What appears to be operator error is most likely a broken process, missing check, or unclear expectations. The analysis phase is complete once the immediate, contributing, and root causes are identified. Keep in mind that the root cause is dependent on the reach of the team. If the last contributing cause exists at a boundary that cannot be crossed (by either adding technical or organizational influence), then it is effectively the root
Really
Figure 8. Case 2 Fault Tree.
Table of Contents
cause. In other words, this is where the solution phase should focus. It is only of academic value to identify a root cause over which the team has no influence. For example, consider again the cause chain for Case 2 (Figure 7). The contributing cause is identified as GT service provider delivered poorly welded combustors. Clearly, this cause lies outside of the current teams influence. If the gas turbine (GT) owner hopes to eliminate this cause, they must convince the service provider to join the investigation. Solution Certainly, failure prevention starts with good design, manufacture, and operation practices, providing protection against the already considered failure modes. For complicated systems, cause and effect diagrams and fault trees are used to study potential failure modes, allowing them to be incorporated into the design phase. However, it is just as important to learn from experience, preventing the recurrence of the same mistake, and this is where the solution phase of root cause failure analysis fits in. The fundamental objective of the solution phase is to break the cause chain. Of course, this means that the quality of the solution depends heavily on the quality of the cause chain developed in the analysis phase. To emphasize this point, consider again the problem statement for Case 1. As shown in Figure 10, the correct initial problem statement is Why did the product fail to pass final inspection? Suppose the team instead started with Why werent the drawings corrected after the first manufacture? This statement is located in the lower half of the why chart, below a critical split. Such a poorly posed problem statement would prevent the investigating team from identifying the real root cause, which is the assumption that all previously manufactured products have updated drawings.
Therefore, using the well-developed cause chain as a starting point, the solution phase begins by identifying all the possible ways to break the chain. These solutions are referred to generically as corrective/preventive actions. Each corrective/preventive action must be evaluated for effectiveness (i.e., does it reduce the likelihood of the event recurring to an acceptable level) and realism (i.e., is it reasonable to implement with respect to cost, time, organizational influence, technical requirements, etc.). Table 2 is a list of corrective/preventive actions for both of the case studies, along with an assessment of effectiveness and realism for each potential action. Table 2. Corrective/Preventive Actions.
Figure 10. Case 1 Why Chart. Another important feature of the cause chain is that since most failures are the result of both root and contributing causes, there are usually multiple areas that can be addressed. This is important to recognize in the solution phase, as it helps to open up the number of possible solutions to the original problem. It is also possible that preventing some of the contributing causes can also lead to improved reliability in other areas not presently considered. For example, in Case 1, one of the possible contributing causes is Shop fixed the problem without engineering involvement. Clearly, if this occurs, there is no feedback through the ECR process, and the design cannot possibly be corrected. Although this did not occur in this particular case, it is identified as a potential failure mode, and preventive action is taken to minimize the possibility of this failure.
For Case 1, perhaps the most obvious, or at least the initial response, is to thoroughly check each drawing before it is issued to the shop. But, when this possible solution is evaluated for effectiveness and realism, it becomes clear that it is a poor option. It is later determined that the best corrective/preventive action plan is to eliminate the ECR back log (and discontinue practice of adding to the back log), and begin to review previous job files for each repeat order, thus significantly improving the probability of eliminating repeat mistakes. For Case 2, although the root cause is an assumption of service provider quality, it is also clear from the cause chain that there are two reasonable approaches to the corrective/preventive action plan. First, it might make sense to begin the practice of inspecting combustors for weld quality before they are installed. This provides the GT owner with the direct control of the combustor weld quality. However, it is also in the GT owners interest to get involved in the service providers quality system, making it possible to eliminate the extra level of quality inspection. Obviously, this approach requires teaming with the service provider, and therefore may not be relied upon for the immediate correction that is necessary before the next outage.
SUMMARY
Failures in human-made systems reflect both technical and organizational flaws. Although it is unreasonable to expect perfect performance with perfect reliability from these systems, it is just as unreasonable to allow the same failure to occur multiple times. Therefore, the objective of this tutorial is to provide the reader with a practical guide for performing root cause failure analysis and determining the appropriate corrective/preventive action necessary to avoid the same failure in the future.
Table of Contents
RCFA starts with the collection phase, consisting of team forming, problem definition, and of course data collection. Next is the analysis phase, determining the immediate, contributing, and root causes of the defined problem. Finally, the solution phase consists of determining the appropriate corrective/preventive action plan that will effectively break the cause chain. In each phase of the process, there are critical steps and simple guidelines to consider that will keep the investigation focused and practical. These, of course, are the key characteristics of a successful root cause failure analysis.
BIBLIOGRAPHY
AlliedSignal Aerospace FM&T, 1997, Root Cause Analysis and Corrective/Preventive Action Workshop. Vesely, W. E., Goldberg, F. F., Roberts, N. H., and Haasl, D. F., 1981, Fault Tree Handbook, NUREG-0492, U.S. Nuclear Regulatory Commission, Washington, D.C.
Table of Contents
TURBINE FAILURE AND RECONSTRUCTION

by Charles. R. Rutan
Senior Engineering Advisor, Specialty Engineering Lyondell Chemical Company Alvin, Texas
Table 1. Design Data. Charles R. (Charlie) Rutan is Senior Engineering Advisor, Specialty Engineering, with Lyondell Chemical Company, in Alvin, Texas. His expertise is in the field of rotating equipment, hot tapping/plugging, and special problem resolution. He has three patents and has consulted on turbomachinery, hot tapping, and plugging problems all over the world in chemical, petrochemical, power generation, and polymer facilities. Mr. Rutan received his B.S. degree (Mechanical Engineering, 1973) from Texas Tech University. He is a member of the Advisory Committee of the Turbomachinery Symposium, and has published and/or presented many articles.
Turbine Number of stages Rating, bhp (kW) Speed, rpm Maximum continuous speed, rpm Tripping speed, rpm Normal inlet pressure, psig Normal inlet steam temperature, F Normal exhaust pressure, psig Wheel diameters First rotor response speed (critical), rpm Rotor material Rim welds for the first and second wheels Rim welds for the third wheel Bucket material Compressor Number of stages Speed, Rpm Gas handled Molecular weight K Z Inlet pressure, psia Inlet temperature, F Discharge pressure, psia Discharge temperature, F Rating, bhp Gear Type Rating, bhp Input speed, rpm Output speed, rpm Gear ratio Service factor Motor Type Rating, bhp Speed, rpm Phase Volts Efficiency, % Hertz, cycles per second Service factor Induction, premium efficiency 15,030 1793 3 12,470 97.3 60 1.25 Double helical, speed increaser 17,285 1793 8676 1:4.8388 1.4 1 8676 Recycle mixture 25.79 1.305 0.964 235 85 317 135.5 13,390 3 5050 (3768) Original 7680, New 8676 Original 7757, New 8676 9000 760 626 250 22.616 4700 ASTM A470 Class 8 F8 F7 Original #1 AISI 422 SS, New TG-410AB001
ABSTRACT
This paper is an extended case history of a turbine catastrophic failure due to stress corrosion and the reconstruction of the turbine, compressor, gearbox, and the drive motor after a fire. Wheel stresses and the design upgrades to minimize the potential for failure while reducing the subsequent damage if a failure occurred of the complete system will be discussed.
INTRODUCTION
A Lyondell manufacturing facility has a recycle gas compressor that continuously recycles ethylene and ethylene oxide to the reactors. On December 28, 1996, the drive turbine of the compressor machine train catastrophically failed. General Description The equipment consists of a compressor driven at one end by a steam turbine and at the other end by a motor through a speedincreasing gear. The driving motor, gear, compressor, and turbine are mounted on a common bedplate. The compressor train is equipped with nonlubricated diaphragm couplings, water seal system, control and lube oil systems, and the accessories necessary for the safe and efficient operation of the unit. Design data can be found in Table 1. The operation of the compressor train is somewhat unique for a chemical plant. The turbine is started and once it reaches 6000 rpm, the motor is started and the compressor train comes up to 8676 rpm. Figures 1 and 2 show the layout of the compressor platform. During the summer, at peak energy consumption, the turbine is used to augment power when electrical power shedding occurs for the local electrical grid. This process lets steam down to a level required by the process and uses the steam developed in the process reactors. A major overhaul and an uprate of the turbomachinery train had been performed in the spring of 1996. The new motor, couplings, and gearbox were installed at this time. However the turbine, compressor, and foundation did not require modifications as the uprate
1
PROCEEDINGS OF THE THIRTY-FOURTH TURBOMACHINERY SYMPOSIUM 2005
Turbine
Compressor
G e a r
Motor
Low lube oil level Overspeed Low sealing water flow Low sealing water pressure Low sealing water flow
With the exception of the motor electrical overloads located in the motor control center, all the instrumentation was locally monitored and controlled from a foul weather building located on the compressor platform. When the turbine failed (Figures 3, 4, 5, and 6), the inboard bearing cover split into two pieces (Figure 7) and the outboard bearing cover rolled to one side of the turbine (Figure 8). Oil began spewing out all over the platform and the atomized oil ignited. The ignition source was the molten metal caused by the heat generated from the metal-to-metal contact rub in the bearing areas. The platform was engulfed in flames burning all the instrumentation, electrical wiring, and the foul weather building. The conduit melted and the electrical insulation melted away exposing the wire that came in contact with other wiring and the metal structure (Figure 9). At this point, the production operator arrived on the seen and was about to turn on the firewater monitor, located at grade level on the platform. However, when he saw all the electrical arcing he decided that it would be unsafe and elected to let the fire burn itself out, which occurred when all the lube oil had been consumed. This was not a good idea. The compressor mechanical water end seals did not fail and the trip throttle valves located at the inlet and inner stage admission did close, but the motor continued to turn at-speed until the motor current overloads tripped.
Figure 1. Compressor Platform Layout Plan View.
Figure 2. Compressor Platform Layout. was a speed increased from 7757 rpm to 8676 rpm and an increase in horsepower occurred to meet the new load requirements of the compressor. The compressor manufacturer completed a thorough dynamic and torsional analysis and found no issues. The turbine and compressor rotors were removed, inspected, and refurbished locally. Inspection included:
A thorough cleaning. Visual inspection. Dimensional verification. Magnetic particle

Shafts. Integral wheels. Buckets. Shroud bands. Tenons. Impellers.
Complete electrical runouts. Complete mechanical runouts. Incoming low speed balance.
Minor wear of the turbine and compressor was reported, which required some handwork. The rotors were low-speed balanced, then at-speed balanced, and returned to the facility where they were installed in their respective cases. The nozzles, diaphragms, packing boxes, etc., were similarly inspected with the same result. New packing and bearings were installed. The plant had most of the normal spare parts including a new spare compressor rotor. Plant management did not see the value in a spare turbine rotor, spare gears, or a spare motor. The vibration monitoring system and turbine controls (completely manual) were not upgraded. The time limitation for the installation was the time it took to dump the reactor catalyst and reload new catalyst. Due to the process safety concerns the facility had, as per the standard operating procedure (SOP), all of the automatic shutdowns of the entire compressor train were bypassed: Figure 3. Inboard Compressor Turbine.
Vibration Thrust Low lube oil pressure
Figure 4. Valve Rack Support.
Figure 8. Turbine Outboard Bearing Cap.
Figure 5. Turbine Outboard Wobble Foot.
Figure 9. Typical Electrical Instrumentation Junction Box. The turbine was supplied with steam from two sources. The inlet or head end steam was purchased from another chemical company that has a plant located near this plant. Low-pressure steam was generated in the unit reactors, and then introduced between the second and third stages. At this location inside the turbine the stationary nozzles are located in the upper half and a diaphragm in the lower half of the case. As the incident investigation progressed, a notification of a caustic excursion in the boiler feedwater makeup had occurred on December 24, 1996. The source was the company that supplied the high-pressure inlet steam to the turbine. No additional information was supplied at that time. Operations personnel took a condensate sample and found that it had a 13 pH. Supervision did not realize that a pH of 13 was a serious issue and could cause stress corrosion cracking of steels. The steam supplier had sent, by mail, a formal notification of the caustic excursion to the plant manager, but he did not receive it until December 30, 1996. This notification was too late by six days. On the morning of January 3, 1997, the compressor platform was released to maintenance to begin the job of damage assessment and repair. Structural consultants, piping experts, instrumentation, electrical, piping, and rotating equipment engineers were brought in to see what could be salvaged and what items had to be replaced. The motor, gearbox, and compressor case (with rotor) were returned to the original manufacturers for inspection. The compressor rotor, turbine end, was severely damaged (Figure 10). Bearing journal, labyrinth oil seals, and coupling areas would require submerge arc welding and machining to bring these areas back to the original specified tolerances. Additionally, the rotor had
Figure 6. Exhaust End Wobble Foot.
Figure 7. Turbine Inboard Bearing Cap Split.
Figure 12. Turbine Third Stage.
Figure 10. Compressor End. a 5 mil kink, turbine side, starting at the compressor end seal area extending to the shaft end. A decision was made that no effort would be made to repair the compressor shaft at this time because a new spare compressor rotor was available; so all the attention was focused on the turbine repairs. The motor sustained minimal damage to the bearing journal and oil seal areas. After the repair of the journal areas and the installation of a new bearing, the motor was tested and returned to the plant with a fresh coat of paint. Minor damage was found upon inspection of the gear. The bearing journal areas were hand dressed. The gear teeth required regrinding and new bearings were installed. The gearbox was then tested and returned to the plant with a fresh coat of paint. While the above was transpiring, the decision was made that repairing the turbine rotor was more expedient than waiting for a new forging. A detailed list of all the components that could be repaired, replaced with stock inventory, and fabricated was developed. An outside consultant was contacted to determine what caused the turbine failure. The first two wheels suffered severe damage. It was learned that one complete segment of the third wheel was missing (Figures 11, 12 and 13).
Figure 13. Third Wheel. When the remainder of the turbine rotor was magnetic particle inspected the Christmas trees (bucket tenons) of the third wheel looked like they had grown hair (Figure 14). The cracks were sectioned and inspected. Caustic was present and the wheel had failed due to stress corrosion cracking. A section of the wheel, identical to the section that had failed (an eight-bucket packet), was removed and weighed. A quick calculation with this weight, radius from the center of the rotor and speed, was performed. It was determined that when the section came off the rotor, the turbine experienced an unbalanced force of 160,000 lb. If any of the shutdowns had been activated, and the compressor train had automatically been shut down, then the damage would have been bad, but not catastrophic. Fortunately, the compressor mechanical water end seals did not leak and remained intact. If these seals had failed and the ethylene oxide gas had been released to the atmosphere, there was a very high probability that an additional event would have occurred.
STEAM CONTAMINATION
Stress corrosion cracking is one failure mechanism that has been well documented. Please refer to the attached list of references at the end of the paper for more information. Turbine manufacturers are all in agreement that the steam purity must be maintained at the lowest practical level of contaminants. It should not exceed 3.0 ppb Na, cation conductivity of 0.2 mho/cm. Total suspended solid shall not exceed 0.1 ppm (100 ppb) and pH 8.0 to 10.0 (inclusive) during normal operation. The manufacturers recommendations are intended to limit sodium compounds, such as caustic (NaOH) and sodium chloride (NaCl). One manufacturer has published the following limits during periods of abnormal operation: for short
Figure 11. Turbine First and Second Stages.
Table 3. ASME GuidelinesBoiler Feedwater, Boiler Water, and Steam Specifications Applicable for Steam Drums Operating Between 0 to 300 PSIG.
Boiler Feedwater Iron Copper Total hardness pH Boiler Water Silica Total alkalinity1 Specific conductance 150 ppm as SiO2 350 ppm2 as CaCO3 3500 mmho Report 0.100ppm or 100 ppb as Fe 0.050 ppm or 50 ppb as Cu 0.300 ppm or 300 ppb as CaCO3 Report
Figure 14. Third Stage Sectioned. periods, not to exceed 100 hours per incident and accumulating 500 hours or less in a 12 month operating time, 6.0 ppb Na and 0.5 mho/cm should not be exceeded. During emergency conditions, for periods of 24 hours or less with accumulations not exceeding 100 hours in a 12 month operating time, 10.0 ppb and cation conductivity of 1.0 mho/cm should not be exceeded. The plant did not, at the time of the incident, have the capability of measuring anything other than the pH of the steam condensate. Just as a reminder, the plant had measured a 13 pH of the steam condensate. This situation was rectified during the period of time it took to rebuild the compressor train. Typical specifications for steam quality are found in Tables 2, 3, 4, and 5. Table 2. ABMA GuidelinesBoiler Feedwater, Boiler Water, and Steam Specifications Applicable for Steam Drums Operating Between 0 to 300 PSIG.
Total dissolved solids1 Total alkalinity2 Suspended solids Total dissolved solids2,3 steam 700-3500 ppm
pH
Notes: 1 Minimum level of hydroxide alkalinity in boiler below 1000 psi must be individually specified with regard to silica solubility and other components of internal treatment. 2 Maximum total alkalinity consistent with acceptable steam purity. If necessary, the limitation on total alkalinity should override conductance as the control parameter. The above parameters and limits should be reviewed with the site water treatment provider.
Table 4. ABMA GuidelinesBoiler Feedwater, Boiler Water, and Steam Specifications Applicable for Steam Drums Operating Between 301 to 450 PSIG.
Total dissolved solids1 Total alkalinity2 Suspended solids 600-3000 ppm 120-600 ppm as CaCO3 10 ppm 0.2-1.0 ppm
140-700 ppm as CaCO3 15 ppm (max) 0.2-1.0 ppm (max expected value)
Total dissolved solids (steam)2,3
Notes: Actual values within range reflect the total dissolved solids in feedwater. Actual values within the range are directly proportional to the actual value of total dissolved solids of boiler water. 3 These values are exclusive of silica.
1 2
Notes: 1 Actual values within range reflect the total dissolved solids in feedwater. 2 Actual values within the range are directly proportional to the actual value of total dissolved solids of boiler water. 3 These values are exclusive of silica.
Water/steam related parameters are found in Tables 6, 7, and 8. The overall quality of the combined fresh and recoverable condensate (vacuum and/or suspect) makeup ultimately impacts boiler efficiency and operating continuity and steam purity. Poor steam purity can lead to carryover related problems such as steam turbine fouling. Lastly, continuous and intermittent blow-down control can help ensure optimum boiler water cycles of concentration, and reduce unwanted impurities. The following information covers online sodium analysis and its relationship to total solids/total-dissolved solids. A sodium analyzer is used to determine steam purity by measuring the sodium ion present in steam with the use of specific ion electrodes. Specific sodium ion electrodes (Note: Samples streams must be cooled with a sample cooler to prevent damage to various components
of the sodium analyzer) have been developed which are extremely accurate (2 percent) and reliable. As a result of this and because sodium is the most common ion found in boiler water, it is an excellent indicator of boiler water carryover problems. When using the specific sodium ion technique, the approximate total solids present in steam is calculated by multiplying the sodium ion content by a factor of three. Thus, a sodium ion reading of 0.33 ppm would represent approximately 1.0 ppm of total solids. By running steam purity evaluations, determination can be made of the carryover TDS (total dissolved solids) in steam as well as evaluate methods for reducing TDS in the steam by making mechanical or chemical changes in boiler operation. Other applications for sodium analyzers include monitoring demineralizer effluent and condensate. (Drew, 1994)
Table 5. ASME GuidelinesBoiler Feedwater, Boiler Water, and Steam Specifications Applicable for Steam Drums Operating Between 301 to 450 PSIG.
Boiler Feedwater Iron Copper Total hardness Boiler Water Silica Total alkalinity
1
Table 7. Water/Steam Related Parameters Specifically for 901 to 1000 PSIG Steam Generators (ASME Guidelines Unless Otherwise Stated).
Boiler Feedwater Limits
0.050 ppm or 50 ppb as Fe 0.025 ppm or 25 ppb as Cu 0.300 ppm or 300 ppb as CaCO3
Organic Iron Copper Total hardness Dissolved oxygen*
Nondetectable (TC/TOC min. detection ~ 0.5 ppm) 0.020 ppm or 20 ppb as Fe 0.015 ppm or 15 ppb as Cu 0.050 ppm or 50 ppb as CaCO3 10 ppb as O2 (w/o oxygen scavenger)
90 ppm as SiO2 300 ppm2 as CaCO3 3000 mmho
Boiler Water Silica** Total alkalinity Specific conductance Total solids Suspended solids Steam Purity Limit Total dissolved solids Silica 0.1-0.5 ppm (max. expected value) 0.02-0.03 ppm, or 20-30 ppb as SiO2 (ABMA limit) 8 ppm as SiO2 100 ppm as CaCO3 1000 mmho/cm 1250 ppm (ABMA max.) 40 ppm (ABMA max.)
Specific conductance
Notes: 1 Minimum level of hydroxide alkalinity in boiler below 1000 psi must be individually specified with regard to silica solubility and other components of internal treatment. 2 Maximum total alkalinity consistent with acceptable steam purity. If necessary, the limitation on total alkalinity should override conductance as the control parameter. The above parameters and limits should be reviewed with the site water treatment provider.
Boiler Feedwater Limits Organic Iron Copper Total hardness Dissolved oxygen* Boiler Water Silica** Total alkalinity Specific conductance Total solids Suspended solids Steam Purity Limit Total dissolved solids Silica 0.2-1.0 ppm (max. expected value) 0.02-0.03 ppm, or 20-30 ppb as SiO2 (ABMA limit) 35 ppm as SiO2 (ABMA max.) 250 ppm as CaCO3 2500 mmho/cm 2500 ppm (ABMA max.) 100 ppm (ABMA max.) Nondetectable (TC/TOC min. detection ~ 0.5 ppm) 0.030 ppm or 30 ppb as Fe 0.020 ppm or 20 ppb as Cu 0.200 ppm or 200 ppb as CaCO3 10 ppb as O2 (w/o oxygen scavenger)
Notes: * Well-designed and operated deaerators can reduce oxygen to as low as 7 ppb. ** Silica limit based on limiting silica in steam. ASME guidelines unless otherwise stated. The above parameters and limits should be reviewed with the site water treatment provider.
Boiler Feedwater Limits Organic Iron Copper Total hardness Dissolved oxygen* Boiler Water Silica** Total alkalinity Specific conductance Total solids Suspended solids Steam Purity Limit Total dissolved solids Silica 0.1 ppm or 100 ppb (max. expected value) 0.02-0.03 ppm or 20-30 ppb as SiO2 (ABMA limit) 2 ppm as SiO2 Not specified, dictated by boiler water treatment program 150 mmho/cm 1000 ppm (ABMA max.) 20 ppm (ABMA max.) Nondetectable (TC/TOC min. detection ~ 0.5 ppm) 0.010 ppm or 10 ppb as Fe 0.010 ppm or 10 ppb as Cu None detectable 10 ppb as O2 (w/o oxygen scavenger)
Comment: It is assumed that the author considers total solids and total dissolved solids as being synonymous. Therefore, if the TDS in steam concentration (ppm) is determined analytically, or taken from ASME/ABMA steam purity guidelines, the theoretical sodium concentration can be determined by multiplying the TDS concentration by 0.33.
Cleaning Cleaning rotors after being exposed to steam contaminants involves the removal of water-soluble and water-insoluble deposits. The original equipment manufacturer (OEM) should be contacted for recommendations concerning water washing and grit blast specifications and procedures. The author prefers to grit blast prior to water washing. For the general cleaning of steam turbine rotors, the recommended material is aluminum oxide with a particle size of 220 mesh maximum. All the sensitive areas must be protected by masking, such as the journal bearing areas, thrust disk, threads, rotor shaft ends, vibration probe, etc. It is vital to have an experienced individual performing this task. If the blast gun is held in one place too long, metal will be removed in addition to the contaminants. The only reason for water washing is to remove water-soluble deposits in the dovetail regions of the bucket and wheel assemblies and the tenon areas of the bucket and shroud band areas. In general water washing consists of three steps: 1. A general rinse and high-pressure water spray to remove the majority of the water-soluble deposits on the surface. 2. Completely submerging the turbine rotor assembly in a water bath. The bath must consist of demineralized water with a conductivity of less than 1 micro siemens (S) per centimeter and have a pH range of 5.0 to 7. 3. A high-pressure water wash to remove seepage products following the soaking is the final step. The seepage must be tested and the pH must be below 7 to 10. If the seepage is above a pH of 10, then the rotor must be resubmerged in the cleaning bath. Step 3 must be repeated as many times as required to lower or raise the seepage pH into the recommended range. Turbine rotors can be water washed at slow roll speeds. At-speed water washing is not recommended by any of the turbine manufacturers. This will remove water-soluble deposits such as salts (NaCl) and turbine blading if not executed very carefully. The deposits in the turbine that are not water-soluble, such as silica, can only be removed by abrasive blasting of the rotor with the critical areas, e.g., bearing journals and vibration probe, masked.
Figure 15. Initial Weld Pass of Turbine Rotor.
Figure 16. Submerged Arc Welding.
BACK TO THE STORY

The turbine case was sent to welding and after a thorough cleaning the rotor was stress relieved then put into a lathe. The rotor was turned (machined) true. The wheel areas were actually undercut in the rotor shaft. The next step in the process was to begin submerged arc welding. Weld metal was deposited about 3 inches on the radius, then the rotor was turned to a rough dimension and the rotor was magnetic particle inspected. If the rotor passed the magnetic particle inspection the weld material was ultrasonically tested to find any inclusions in the welds. When an inclusion was found, greater than 1/32 inch, the inclusion was ground out and repaired via welding (Figures 15 through 19). This procedure was repeated many times until the rotor was sent to final machining and inspection (Figures 20 through 22). During this period discussions were conducted to determine if there was any way to reduce the stresses on the third wheel. An agreement was reached that the steel bucket and shroud bands would be replaced with titanium Z lock buckets that would reduce the forces on the third wheel. A full engineering study was completed and no issues were found with the change of the bucket material. As the rotor work was proceeding, the turbine case was being weld repaired. As can be seen this was not an easy matter (Figures 23 through 26). First, the split lines of the upper and lower halves were weld repaired and rough machined. Then, the register fit areas were weld repaired and rough machined. The turbine case halves were then sent to stress relieving. Once this operation was completed, the case was returned to the machine shop and final machining was completed and the case was dimensionally checked.
Figure 17. Machining after Initial Weld Pass. The original cast bearing bracket was fabricated while other work was performed. The valve lift beam had damaged supports and had to be fabricated along with other miscellaneous components. The rotor was at-speed balanced to 0.5 mm/sec/bearing (the authors specification), then installed in the repaired case and returned to the plant (Figure 27). All the above work detailed took six weeks. Since all the controls and vibration/temperature monitoring equipment were destroyed in the fire, this opportunity was taken to upgrade these systems. New state-of-the-art turbine controls and monitoring systems were installed that allowed the startup of the train to go smoothly. The startup procedure was:
Figure 18. Turbine Rotor Stress Relieving.
Figure 21. Cutting Bucket Dove Tails in Turbine Wheels.
Figure 19. Turbine Rotor Welding Completed.
Figure 22. Loading Buckets into Turbine Rotor Wheels.
Figure 23. Upper Half of Case. Figure 20. Turbine Rotor Final Machining. 1. Line up and start the water to the compressor water seals. 2. Line up and start the lube/control oil system. 3. Test the shutdown turbine overspeed shutdown devices. 4. Perform three shutdowns: a. Low oil trip b. Thrust trip c. Overspeed trip 5. Couple the turbine to the compressor train. 6. Slow roll the turbine at 500 rpm for 1 hour to allow the turbine case, turbine rotor, and oil systems to warm up close to their operating temperature.
Figure 24. Case Split Line.
Figure 25. Bottom Half of Case.
Figure 26. Upper Half of Case after Welding Stress Relieving and Machining.
Figure 27. Finished Turbine Rotor in the At-Speed Balance Bunker. 7. Step the turbine speed up at 500 rpm per step until the speed set point, which is 500 rpm below the critical speed. 8. Press fast ramp on the electronic governor, the turbine ramps up at 250 rpm/second until 500 rpm above the first critical speed. 9. Set at this speed for 30 minutes, allowing all the turbine component temperatures to become stable. 10. Once the minimum operating speed set point is reached, continue to jog up at 500 rpm steps until the high-speed stop (maximum continuous operating speed) is reached. 11. Bypass the high-speed stop and run the turbine up to the overspeed trip set point. Once the turbine was soloed it was shut down and coupled to the compressor/gearbox/motor and restarted. The startup of the
compressor train steps is identical until step 10 (listed above), when 6000 rpm is reached, the motor is started, and the equipment is brought up to the normal operating speed of 8676 rpm. This step takes about four seconds, then the machine is turned over to operations. The vibration levels were less than 1 mil in any plane. Bearing temperatures, thrust positions, compressor process temperatures, etc., were at their normal operating conditions and life was good! Approximately an hour after startup, the compressor inboard and outboard bearing temperatures begin to rise, and the vibration in both planes started to rise about 0.2 of a mil per hour. All the other operating parameters remained unchanged. The maximum radial shaft vibration level, in any plane, was considered to be 4.5 mils. The orbit of the compressor shaft in the bearings was perfectly round and at synchronous speed. A spectrum analysis of the vibration also indicated that the vibration was predominately at synchronous speed. It appeared that an unbalance existed and was slowly moving away from the compressor shaft centerline. The compressor was shut down and allowed to come to ambient temperature and then restarted. At speeds below the first critical speed, the vibration levels of the compressor rotor were slightly above those seen initially on the first startup. Once the critical speed of the compressor was gone through, the vibration levels rapidly increased to 4.5 mils and continued to climb. Analysis of the vibration indicated that at this point a rub was being picked up and thus the vibration climbed at a faster rate. Two additional attempts to start up were made before the conclusion to split the compressor case and determine the cause of the vibration. The compressor case was split and some minor rubbing was apparent, but it was not significant enough to cause the vibration levels to be seen until a dial indicator was put on the compressor lower half and the runout was checked at the center of the rotor. A 45 mil bow was discovered. The rotor was removed from the case and returned to the manufacturers machine shop. Dimensional checks were repeated. Runouts were repeated. Except for the instance when the runout at the center of the rotor was performed, 5 mils of runout were measured, which was acceptable for this rotor. But this was not close to the 45 mils seen in the field. Since all readings were back to within tolerances, the rotor was returned to the plant and reinstalled in the compressor. The thought was that something must have shifted on the compressor rotor during the transport by truck to the machine shop. After the reinstallation of the rotor, the compressor was closed up and recoupled and the machine train started up. At this point, the repair was started on the compressor rotor that was originally installed in the compressor during the initial wreck. The compressor train was restarted, as per procedure, and the exact same sequence of events occurred as detailed above. The compressor case was shut down and split, once again, and the condition of the rotor was as it had been left. Management was not happy and life was not very good. The rotor was again checked. There was a 45 mil bow in the center of the rotor, so the rotor was returned for a thorough inspection. The same results were seen in the shop as before, rotor runout at the midspan was 4 mils. The rotor was unstacked and reassembled and everything was to specifications. A decision was made to return the compressor rotor to the facility. Once installed in the lower half of the compressor case, a runout check at midspan would be performed, and it was. The runout at midspan was exactly as measured in the shop, 4 mils. Armed with this information a decision was made to complete the installation and startup. As expected by some, the same sequence of events occurred. Life at this time was terrible! When the rotor was removed from the case this time, it was transported with the thrust disk and the end seals were left on the rotor. The rotor was set into v-blocks, seals supported, and runouts were retaken. The 45 mil runout at midspan was found this time. As the rotor was disassembled, a dial indicator was placed at the midspan so any changes could be seen.
10
In the shop, with at least a dozen spectators, the thrust disk nut was loosened and the midspan went from 45 mils to four mills. The OEM representative and the author were watching the dial indicator when it changed. We would have called each other a liar if we had not both seen the change. Fortunately, the compressor rotor from the initial wreck was repaired and the rotor was ready for installation. It was installed into the compressor and the compressor train was restarted, as per the operating procedure. The plan was to get the unit up then go back to the new rotor and perform a very detailed inspection and determine what was causing the 41 mil change in the rotor assembly. The machine train was restarted and vibration levels were less than 1 mil peak-to-peak at the operating speed of 8676 rpm. Life was good again. The problem compressor rotor was stripped (impellers removed) and every type of test to the 17-4ph shaft was performed. Nothing could be found that would explain why the rotor would bow. The final test was to hang the shaft vertically and stress relieve it. Once the stress relieving was completed it was allowed to cool, then it was placed in a low speed balance stand and shaft runouts were taken. The problem was immediately determined, a 45 mil bow was found in the shaft at its midspan. The shaft had not been stress relieved properly. The OEM bought us a new shaft and reassembled the rotor. Several years later during a major compressor overhaul the new rotor was installed and it ran perfectly. One would think that everyone lived happily ever after but they would be wrong. The compressor train ran about three months, then the author received a call that the vibration levels at the bearing journal had pegged the vibration monitor at 5 mils and the temperature of the turbine bearings was starting to climb. Operations was advised to shut down immediately and life had turned again. The unit was shut down. The turbine bearings were inspected and nothing was found. Digital data recordings of the vibration were reviewed and it appeared that a mass unbalance suddenly occurred. The turbine case was split and four titanium buckets had broken off at the blade root and were lying in the bottom of the case. Now a way had to be found to start up the compressor train without the turbine. A modal and tensional analysis was completed. The only way the compressor train could be brought up was to leave the turbine side coupling that was mounted on the compressor rotor on the end of the rotor; that overhung moment was needed to have a stable machine train. Startup procedures were rewritten to prevent the motor from tripping due to the extended rampup time that was expected for motor overloads. The time out was set from 8 seconds to 15 seconds and it actually took 13 seconds to reach 8676 rpm with the compressor completely unloaded. About three weeks later the motor tripped due to overloads and this was a surprise to operation personnel because they were not aware this shutdown set point was not increased to give them any warning. This did prove that they could trip without warning and the plant would not exceed a safety critical variable. All the safety systems functioned as designed; ethylene oxide gas was directed to the flare. Now the reason the titanium buckets failed could be investigated. The buckets were removed from the third wheel and the rotor was inspected thoroughly and the wheel was in as-new condition. It was finally determined that the steam at the inlet of the third stage was causing very high alternate loading of the buckets because the purchased steam was still super heated and the reactor steam was just at the saturation temperature at this point in the turbine. Options were defined and it was decided to add thickness, about inch to the third stage wheel and install thicker buckets and these buckets would have shroud bands. The modifications were made to the turbine rotor and the rotor was installed in the case on the compressor platform awaiting an opportunity to couple up to the compressor. We did not have to wait longtwo months before an electrical outage brought the unit down. During the outage the turbine was coupled to the compressor and started up, as per the operating procedures. The turbine rotor forging that
had been ordered was then received and machined to the new specification. It was then put into a climate controlled storage hanging vertically as a spare.
ANOTHER TURBINE FAILURE

In mid-1999, after several years of smooth operation the turbine rotor vibration began to rise. At 2.5 mils the first vibration alarm sounded in the control room alerting the operation personnel that the turbine had a problem. The other equipment vibration levels remained normal. Fortunately a field operator was in the immediate area and reported a strange noise emitting from the turbine. The resident mechanical engineer was called and he told operations that if the vibration reached 4.5 mils to shut down the compressor train. Within a minute, operations had to shut down the compressor train. Upon reviewing the vibration data, all indications were that there was a mass unbalance and it was getting worse. The decision was to uncouple the turbine and run the compressor with the motor alone. Maintenance forces were mobilized and the turbine was dismantled after a quick inspection of the bearings showed no signs of damage. The case was lifted; the problem was obvious. The shroud band of one of the bucket packets was pealing off like an 18-wheeler retread. Metal pieces were recovered from the recycle gas drive compressor turbine after its failure in May. They were believed to be part of the shroud assembly that covered the turbine buckets. The metal pieces, all smaller than 1 inch in length and inch in width, were heavily banged and deformed. However, one round nippleshaped piece, which is believed to be one of the tenons, appeared to have an undamaged fracture surface. The turbine components had been in service for about two years. Analysis SEM Fractography The metal pieces were first ultrasonic cleaned in a concentrated chemical cleaner to remove grease and loose surface films, and then the nipple and one of the larger pieces were examined under a scanning electron microscope (SEM).
Tenon (or nipple) pieceIt was quite fortunate that the fracture
surface was mostly undamaged except in areas near the circumference of the tenon (Figure 28). However, the fracture surface is severely oxidized, which masked out the fine details on the surface. Still, it could be determined to be a fatigue failure (Figure 29) with the typical feature of fatigue striations (Figure 30).
Figure 28. Bucket Tenon.
11
Figure 29. Fatigue Failure.
Figure 31. Ductile Overload Failure.
Figure 30. Fatigue Striations. On one end of the fracture, there was clear sign of final ductile overload fracture (Figure 31). Also known as microvoid coalescence (MVC). Another final fracture area is found about 90 degrees to the first one. Here the microvoids appeared to be distorted or elongated (Figure 32), which indicates the fracture is at a different orientation. Efforts to locate the crack initiation site by tracing back from the final fracture areas to the opposite end of the fracture surface were not successful, as the area was too damaged to be recognizable (Figure 33). A feathery area (Figure 34) across from the second final fracture area was determined to be a fatigue area (Figure 35). When viewed at higher magnification, all in all, the fatigue area was estimated to cover over 70 percent of the fracture surface.
Figure 32. Elongation of Microvoids.
Shroud
pieceThere was not much to look at, because the fracture surface had been damaged (rubbed) and also heavily oxidized (Figure 36). In a cracked area, near one end of the surface, it was also damaged, but it had some scale attached to it (Figure 37). Even the surface inside the crack was found to be severely oxidized (Figure 38). Heavy scaling was also found on the side surface of the shroud piece (Figure 39). Chemical Analysis (by X-Ray Fluorescence)
Figure 33. Badly Damaged Area.
Based
metalThe chemical compositions of the tenon and shroud pieces were found to be close to that of 422 stainless steels. They are listed below in Table 9.
Oxidized surfacesThe oxygen content ranged from 8 to 14 wt.
percent or 24 to 35 atomic percent. In some areas the oxide films were so thick that only 2 to 3 percent of chromium was detected. Other than the basic elements like iron, nickel, and manganese,
12
Figure 34. Feathery Area.
Figure 37. Scale at One End.
Figure 35. Final Fracture Area.
Figure 38. Severe Oxidation.
Figure 36. Fracture Surface. there were also silicon (0.4 to 1.5 percent), vanadium (0.8 to 1.9 percent), aluminum (0.3 to 0.5 percent), zinc (up to 1 percent), copper, (up to 2.4 percent), and sulfur (0.3 percent).
Figure 39. Shroud Piece.
ScaleThe surfaces of the shroud and tenon pieces, including
the fracture surfaces, were covered with scale. The scale was found to be rich in calcium (3 to 4 percent), silicon (0.7 to 3.3 percent),
13
Table 9. Chemical Compositions.

422 SS Chromium Molybdenum Nickel Manganese Silicon Vanadium Phosphorus Iron 11.5 13.5 0.75 1.25 0.5 1.0 1.0 max. 0.75 max 0.15 0.3 0.04 max. Balance Tenon 12.26 1.03 0.79 0.78 0.62 < 0.1 < 0.1 84.5 Shroud 12.46 0.97 0.86 0.75 0.46 < 0.1 < 0.1 84.5
With the above information engineers revised the shroud design by reducing the overall width of the band by 1/2 inch, thus reducing the stress on the tenons.
LESSONS LEARNED
Stress corrosion crack can occur very rapidly when all the right
conditions are present.
Recommended steam quality Using todays technology almost anything can be repaired. If the equipment is critical to the operation of the unit or to the
facility, spare rotors are essential.
phosphorus (1 to 3.2 percent), vanadium, (0.7 to 1.7 percent), copper (0.5 to 1 percent), aluminum (0.15 to 1.42 percent), and zinc (0.4 to 1.1 percent). The failure of the tenon has been confirmed by SEM fractographs as fatigue, or more precisely corrosion assisted fatigue. It is also likely a high cycle (or low stress) fatigue failure because:
New rotors can have unseen problems. 15-5ph material is far better than 17-4ph, relative to stability. High thrust shutdowns should not be bypassed. The perception of danger is not always accurate or understood.
REFERENCE
Drew, 1994, Principles of Industrial Water Treatment, Eleventh Edition, Published by Drew Chemical Corporation, Boonton, New Jersey, p. 310.
The ratio of fatigue area to final fracture area is over 2 to 1. This
means the fatigue crack propagated through a large portion of the cross-sectional area of the tenon before final fast fracture occurred.
The fatigue area was heavily oxidized. In order to have such a
BIBLIOGRAPHY
The ASME Handbook on Water Technology for Thermal Power Systems, Paul Cohen, Editor-in Chief, Sponsored by the ASME Research and Technology Committee on Water and Steam in Thermal Power Systems, EPRI Research Project No. RP 1958-1. ASTM A-380, 1999, Standard Practice for Cleaning, Descaling, and Passivation of Stainless Steel Parts, Equipment, and Systems, American Society for Testing and Materials, West Conshohocken, Pennsylvania.Elliott Specifications for Steam Quality. General Electric Instructions GEK-63430, Turbine Steam Purity. General Electric Instructions GEK-72281, Steam PurityStress Corrosion Cracking. General Electric Power Generation Global Product & Technology Support, TIL 1231-3, Maintenance Recommendations for Cleaning Deposits and Chemical Contamination from Steam Turbine Rotors, August 15, 1997. General Electric Power Systems GEK 98965, Steam Turbines, Steam Purity for Industrial Turbine. Lindinger, R. J. and Curran, R. M., 1981, Experience with Stress Corrosion Cracking in Large Steam Turbines, The National Association of Corrosion Engineers, The International Corrosion Forum, Corrosion/81, Paper Number 7, Toronto, Ontario, Canada. McIntyre, D. R. and Dillon, C. P., 1985, Guidelines for Preventing Stress Corrosion Cracking in the Chemical Process Industries, MTI Publication No. 15.
thick oxide film on the surface, the oxidation is believed to have occurred over a long period of time. The final fracture areas, on the other hand, were not heavily oxidized because the fracture tenon was removed from the turbine shortly after the failure. The shroud and tenon were confirmed to be 422 stainless steel. With the small amount of alloying element molybdenum and nickel, 422 SS offers better corrosion resistance and higher hardenability than the lower alloy 400s martensitic SS like 410 or 416. The excess deformation and rubbing damages on the pieces, the lack of secondary cracking, and the MVC appearance of the final fracture all indicate the material has good ductility and toughness. The severe oxidation and scaling on the failed pieces are not believed to be beneficial to the service life of the turbine components. It was believed that the steam is of poor quality and is corrosive to 422 SS. The steam also contains a high amount of impurities. The elements found in the scales like calcium, phosphorous, silicon, and vanadium are believed to be water treatment chemicals, while the zinc, copper, and aluminum are likely to be corrosion products that were carried over by the steam. Since steam is water vapor, it is not supposed to carry many impurities. The steam used in the turbine is thus believed to be very wet and contained liquid entrainment. Besides oxidation, there was no other sign of environmental degradation like localized corrosion, stress corrosion cracking, or caustic embrittlement on the failed pieces. The failure of the tenon is believed to be high cycle fatigue. The actual fatigue duration is not known, but is estimated to be months. The quality of steam is questionable; it is believed to have contributed to the scaling and heavy oxidation on the failed components. However, there is not enough evidence to indicate the low quality steam, or the scaling and oxidation on the parts, are the root causes of the failure. The materials of construction of the tenon and shroud are 422 SS, which has good corrosion resistance and mechanical properties. So the failure is not believed to be material related.
14
REPAIR TECHNOLOGIES OF MECHANICAL DRIVE STEAM TURBINES FOR CATASTROPHIC DAMAGE

by Manabu Saga
Manager, Turbine Design Section
Shugo Iwasaki
Acting Manager, Materials Laboratory
Yuzo Tsurusaki
Mechanical Engineer, Turbine Design Section
and Satoshi Hata

Manager, Turbine Design Section Mitsubishi Heavy Industries, Ltd. Hiroshima, Japan
Manabu Saga is a Manager of the Turbine Design Section in the Turbomachinery Engineering Department, Mitsubishi Heavy Industries, Ltd., in Hiroshima, Japan. He has had experience with basic and detail engineering of steam turbines and gas turbines for power generation and development. Mr. Saga has a B.S. degree (Mechanical Engineering) from Keio University. Shugo Iwasaki is Acting Manager of the Materials Laboratory in the Hiroshima Research and Development Center, Mitsubishi Heavy Industries, Ltd., in Hiroshima, Japan. He has 12 years of experience in the evaluation of materials, the development of new manufacturing processes, and troubleshooting for turbines, compressors, heat exchangers, and vessels. Mr. Iwasaki has B.S. and M.S. degrees (Materials Engineering) from Kyushu University. Yuzo Tsurusaki is the Mechanical Engineer of the Turbine Design Section in the Turbomachinery Engineering Department, Mitsubishi Heavy Industries, Ltd., in Hiroshima, Japan. He is a specialist in rotor design and has five years of experience with troubleshooting of steam turbines. Mr. Tsurusaki graduated from Kurume College of Technology (Mechanical Engineering).
well as other monitoring and protection systems. However, in some cases, a turbine will suffer severe mechanical damage due to improper operation or failure to activate the protection system as a result of human error. For urgent plant recovery and to minimize the duration of risky operation with no spare rotor, a damaged turbine has to be repaired in as short a time as possible. This paper introduces actual experiences in repairing and reviving catastrophically damaged turbine rotors through special welding procedures, based on element tests to find the optimized welding conditions, detailed strength calculations to confirm the integrity, and heat transfer analysis for proper heat treatment process conditions. These basic procedures are discussed to show useful data. The revived rotor of the extraction condensing turbine was placed back into the casing and operated. The turbine was uniquely modified in order to balance the required amount of power and minimize the repair time, to restart the plant as quickly as possible. The case study for this optimization is discussed by showing thermodynamic calculations, performance, and repair schedules. Root cause analysis for the process of the catastrophic failure is explained, for the integrated control system governing the control and operation positioner systems.
INTRODUCTION
Mechanical drive steam turbines play important roles in petrochemical plants. For safety in operation, these turbines are protected by an overspeed trip device, as well as other monitoring and protection systems. However a turbine still can suffer severe mechanical damage from improper operation or failed activation of protection systems, caused by human error and improper control of steam purity. When rotor damage does occur, a rotor is normally replaced with a spare rotor and the plant continues to operate with the running risk of having no replacement rotor. For urgent plant recovery and to minimize the operating time with no spare rotor, a damaged turbine has to be repaired in as short a time as possible. However, in cases where the damage is catastrophic and existing technology cannot repair the damaged rotor, it has to be scrapped and a new rotor must be fabricated. Consequently, the end user has to wait a minimum of six months to one year for a new rotor. In order to repair and revive a catastrophically damaged turbine rotor in emergencies in response to end user requirements, a new welding technique has been developed based on risk analysis of repair and laboratory element tests. In addition, to optimize welding conditions, the authors studied detailed strength calcula15
ABSTRACT
Mechanical drive steam turbines play an important role as core equipment in petrochemical plants, and these turbines are protected for safe operation by an antioverspeed trip device, as
16
tions to confirm integrity, and performed heat transfer analysis for finding proper heat treatment process conditions. The authors introduce typical experiences in rotor repair by discussing useful results of the above analysis and the test results.
FAILURE MODE OF STEAM TURBINES

The failure damage modes of mechanical steam turbines are classified into main components for steam flow paths, rotating components such as rotors, grooves, disks, and blades, as well as stationary parts such as control valve bearings (Table 1). The basic root causes for each damage mode are listed. Severe damage modes include destruction of blades due to excessive centrifugal force, blade corrosion fatigue failure, and rotor bowing due to interference/rubbing in transient operations such as initial start up. Table 1. Failure Damage Mode.
Components
Control Valve Nozzles (HP)
Damage Mode
Valve Stem Bending Fatigue Failure Solid Particle Erosion Fouling Solid Particle Erosion Fouling Fouling Drain Attack Erosion Fouling High Cycle Fatigue Failure Centrifugal Force Failure High Cycle Corrosion Fatigue Centrifugal Force Failure Rubbing & Shaft Bow & High Vibration Disk SCC Rubbing & Scratch/Wear Rubbing & Melting Creep Deformation/ Creep Rapture, Erosion/Corrosion Diaphragm Bending Deformation Erosion & Corrosion Rubbing & Erosion
Root Causes
High Steam Velocity Fluid Excitation Hard Foreign Materials Inside of Pipes Hard Foreign Materials Inside Steam Impurity Steam Impurity High Moisture & High Velocity Steam Impurity Excessive Excitation Force Excessive Over Speed Improper Water Treatment Excessive Over Speed Improper Start Up Drain Intake, Steam Impurity Excessive Thrust Force Improper Oil Supply (UPS) Excessive Thrust Force Improper Oil Supply (UPS) Operation Over Allowance Excessive Drain / Galvanic Operation Over Allowance Excessive Drain / Galvanic Improper Start Up Excessive Drain
Figure 1. Failure Damage Map. continued to increase, exceeding EOST speed. After the operating governor function key and the turbine control were shifted to the backup system, the governing valve suddenly opened from 12 percent to 60 percent in a no load condition and the turbine rotor was accelerated to 170 percent of MCR. Eventually, the turbine rotor disks and blades became damaged, resulting in oil leakage.
Flow Path Parts
Blades (HP) Nozzles (LP) Blades (LP) Blades (HP) Blades (LP)
Rotating Parts
Rotor Thrust Collar Journal Shaft Bearings Casing
Rotor Disk & Blades Broken at 170% of MCR
Stationary Parts
Diaphragm Seals
EOST Module Recorded Max. 17557rpm
Cross Section of Extraction Condensing Turbine

TTV Control GOV Control GOV Trip 11407rpm
Oil Leakage due to Excessive Vibration Control shifted to Back Up System by GOV F Key GV suddenly opened to 60% due to Bias Signal
In Figure 1, a typical damage condition is shown in the steam turbine cross section. The first stage nozzle profiles are eroded at the trailing edges, due to contact with hard solid particles. The flow path of the diaphragm and blade profiles at the leading edges in the low-pressure (LP) high moisture section are eroded by water droplets. In some cases, the welded zone of the nozzles are affected by a combination of erosion and corrosion. Corrosion fatigue failure will occur in the blades of the LP section at the wet and dry enrichment zone under corrosive conditions, if proper water treatment is not performed during steady and transient operation, or if a corrosive chemical leakage occurs at a heat exchanger. Excessive contact friction at the rotor thrust collar will induce melting damage of thrust bearing pads, due to excessive thrust force from drain invasion. In other cases, lubricant oil additives may generate oil sludge contaminants on pad lubricating surfaces, and this will reduce bearing durability. If lubricant oil is not selected properly, in the worst case the rotor will rub against the bearings.
EOST Trip 11307rpm MCR 10279rpm Customer Design EOST Lower than GOV Trip
SPEED (rpm)
Normal 9500rpm MGS 8075rpm 7000rpm 5000rpm 4840rpm OST Displayed on EOST Module and DCS No Signal from Interlock Circuit of DCS at OST
Valve Opening Deviation was observed.
1st Critical 4400rpm
Preparations for Over Speed Test Valve Opening Deviation was observed. GV Open 12%, ECV Open 100% However, Back Up System indicated GV and ECV Open 60%
Critical Zone
3960rpm 3000rpm
2000rpm
500rpm
Solo Running Test

17:30 19:00 TIME
16:30
17:00
Figure 2. Typical Failure Experience (Overspeed Burst). The control system for this turbine is shown in Figure 3. The control system has backup modules operating by position control, and this position control is used to supply the actuator output signal instead of the electric governor in the case of a governor failure. The backup system monitors the actuator output signals from the governor and controls two signals to switch the actuator signal source from governor to position control. Two backup modules are utilized for governing and extraction control valve actuators. As a result of detailed investigation of the system and signals, it was found that circulating current from the distributed control system (DCS) into the signal system of the backup modules caused the supply signal to the actuator to increase, thereby opening the valve. This caused a large amount of steam flow leading to the abnormal increase in rpm, and the interlock that should have performed an emergency valve close did not operate properly. Circulating current causes a 12 percent to 60 percent deviation of signals between the governor and governor backup module. The actual damage conditions are shown in Figure 4 and Figure 5. The second (high-pressure section) and third stage blades (lowpressure section) have broken off from the rotor grooves, and the disks and grooves are deformed at the first stage and other stages
TYPICAL FAILURE EXPERIENCE

Rotational equipment manufacturers design, manufacture, and deliver steam turbines to customers based on well-proven technologies and supply experiences, and therefore they do not experience catastrophic accidents. However, this paper introduces typical catastrophic damage in using steam turbines. Figure 2 shows trends in rotation speed and key events from start to excess speed rotor damage, during the site-based solo running test. This extraction condensing turbine for driving a charge gas compressor is equipped with a position control backup system. The turbine was coasted up smoothly to a maximum continuous speed of maximum continuous rate (MCR) 10,279 rpm. Rotating speed was increased for a moment to electric overspeed trip speed (EOST); however, the trip interlock did not activate and speed

Extraction Flow
DC 4-20mA
17
Circulating Current from DCS DCS
Speed Pickup-NO.1 Speed Pickup-No.2
DC 4-20mA
Suction Pressure ESD SEQ Trip SW TTV ESD
GV E/H Actuator ECV E/H Actuator
60% 12% Open 60% 100% Open
B/UP Module
Electric Governor
TRIP GOV TRIP
X X
SOV Manually SW OFF
B/UP Module
in as short a time as possible. In order to revive this catastrophically damaged turbine rotor, a special welding procedure was necessary. In order to analyze the applicability of overlay welding for this turbine, technical risks are evaluated in a common procedure for rotor repair according to experiment data. In addition, experience with welding rotors and risk factors are categorized as wellproven, possible, practical, or not well-proven. Table 2 shows details of this test. Table 2. Technical Risk Evaluation.
ALARM
Local Panel Raise/Lower
Over Speed Trip

Speed Pickup No.1, 2 & 3
OVER SPEED TRIP Signal
Risk Factor
Mechanical Properties Deterioration Creep Strength SCC Strength
Evaluation
2.25Cr-1Mo Welded Metal Lab. Test Stress of Damaged Parts Change of Impact Value and others 10% Decrease is Expected from Mech. Property and Temp. Test Confirmed by SCC Test. Residual Stress is Measured by Test and Actual Rotor.. Applied for Test and Actual Rotors including Heat Treatment & NDE. Well-Proven Heat Treatment Procedure Heat Transfer Analysis for Disks
Circumferential Equal Temperature Profile
Results/Countermeasures
8% Decrease of Tensile Strength Proper Safety Factor of Yield Stress Proper Safety Factor of Creep Life Expected 300 to 400 NO SCC Within Allowable WPS/PQR Issued Issued Basic Procedure is OK Heat Transfer Analysis is Required. WPS/PQR Issued
Basically Expected Within Criteria Stress Analysis & Actual Size Test are Required.
2/3 Voting Module
Figure 3. Overspeed Burst and Back Up System Error. of the low-pressure section. The rotor is bowed from a large amount of friction and contact due to a seriously unbalanced condition. The journal and thrust bearings are damaged and their Babbitt metal pad has melted due to a huge friction loss heating resulting from the excessive overspeed and high loads. Rotor shaft at bearing location had no damage and was protected by Babbitt metal melting promptly. As explained in Figure 5, damage of the rotor was catastrophic.
Element for Welding
Residual Stress Welding Procedure Cold Cracking Stress Relief Cracking Blow Holes Defects
Residual Deformation Spring Back Eccentricity & Unbalance Rubbing Spiraling vs Bonding Stress Long Term Capability
Applied for Test and Actual Rotors including Heat Treatment & NDE.
Applied for Actual Rotors and Within Criteria 0.005mm for Disks. No Experience for HP Balance Piston By Fine Setting and Machining After Welding & Balancing, They Can Be Within Criteria.
Confirmed Same Bonding Strength By Lab. Test Deformation due to Stress Relief by Contact Heating
Actual Size Test are Required for Damaged Parts.

Stress Relief and No Spring Back during operation to be confirmed by Actual Size Test Stress Relief and No Spring Back during operation to be confirmed by Actual Size Test
Element for Operation
No Impacts for Creep, SCC by Test and For Corrosion Fatigue due to HP Sect.
Element for Experience
Welded Position Boundary Condition Unexpected Defects
No Experience for HP Balance Piston Experience of HP and LP stage Disks Applied for Test and Actual Rotors including Heat Treatment & NDE.
Actual Size Test are Required. Expected 300 to 400 Heat Transfer Analysis is Required. No Blowhole Inspected
&+5-5 )4118'5 &'(14/'& 41614 4'5+&7#. $19 70$#.#0%'
6*4756 ,1740#. $'#4+0)5 &#/#)'&
,1740#. $'#4+0)5 &#/#)'&
0& 4& 56#)' $.#&'5 (#+.74' #0& $41-'0 1((
Figure 4. Damaged Condition of Turbine.
Elements of welding risk factors include mechanical properties, creep and SCC strength, residual stress, cold cracking, and residual deformation of the rotor. Risks in operation after repair consist of the bonding strength in frictional areas and spiraling around the labyrinth seal at the high temperature and pressure side. For the risk analysis and evaluation, element tests were performed to find the optimum welding conditions, detailed strength calculations were performed to confirm the integrity, and heat transfer analysis was performed to achieve proper heat treatment process conditions. These basic procedures are studied and discussed in the next section. As explained below, mechanical properties and strength can be maintained at the same levels as that of a new rotor. However, residual rotor deformation is affected by residual stress in the surface of the welded shaft section. This deformation must be less than 0.005 mm (.0002 inch) when considering required limit criteria for balance. Though it is possible to calculate residual stress and shaft deformation, it is very difficult to predict such a small amount of deformation precisely. Because of this, residual warpage of the rotor still remains as a risk factor, and an actual full scale test is necessary to confirm that residual warpage is within allowances, and that there are no changes when reheated.
TECHNIQUES FOR WELD REPAIR

Technical Issues and Solutions for Weld Repair Table 3 summarizes the technical issues on weld repair of damaged rotor disks and solutions for those issues. The rotor shaft at the bearing has no damage and was protected by the Babbitt metal melting promptly. Damaged pads can be easily replaced with spare pads. However, there are risks in replacing forged rotor material with weld deposited material. Forging gives the rotor material proper mechanical properties and reduces defects caused by casting, but weld repairs have no such effects. Therefore, it is necessary to apply welding material that will provide proper mechanical properties even though there is no forging in the repair process. The proper welding material was selected as explained in the following section.
Figure 5. Detail Damaged Condition of Turbine.
REPAIR RISK ANALYSIS

For urgent plant recovery and to minimize the duration of operation with no spare rotor, this damaged rotor had to be repaired
18
Table 3. Technical Issues and Solutions on Weld Repair of Damaged Rotor Disk.
Technical issues Process Solution Proper material selection Same allowable defect size for weld metal with base metal Uniform heat input Local heating Uniform heat input Vertical rotor layout Optimum heater layout based on FE analysis Proper heating temperature and time Check points on test Lab. scale Properties of weld metal Welding defects Welding GH Real scale GH
Welding and Post Weld Heat Treatment on a Laboratory Scale Test Piece Figure 6 shows the appearance of the test piece after welding. Both types of weld material were welded to a bar shaped rotor material (1.25Ni-Cr-Mo steel). The weld metal was formed with a thickness of 75 mm (2.95 inches) on the bar by gas tungsten arc welding (GTAW). Optimum welding conditions were selected to avoid defects and these were controlled constantly throughout the welding process to obtain uniform heat input to the bar. Table 5 shows the welding and PWHT conditions for the test pieces.
Deformation
Welding PWHT
Residual stress Reduction of strength on base metal High hardness on HAZ
(Local) PWHT PWHT
BCDE
F C
A: dimension measurement B: Tensile test C: Hardness measurement D: Impact test E: SCC test F: Residual stress measurement G: Ultrasonic test H: Magnetic particle test
Proper welding methods and conditions should be provided to avoid welding defects that would initiate cracks. Proper welding conditions were decided by welding laboratory scale test pieces. An ultrasonic test for the weld metal was performed with the same criterion on minimum defect size as for the base rotor metal. Temperature distribution and gravity during welding and post weld heat treatment (PWHT) cause unfavorable deformation in the rotor. Therefore, the repaired disk was heat treated locally, and the rotor was suspended vertically at the PWHT to prevent bowing of the shaft. The whole circumference of the repaired disk was welded and heat treated uniformly. The welded portion needs to be heat treated to reduce residual stresses due to welding and hardening at the heat affected zone (HAZ), which can cause stress corrosion cracks (SCC) under wet conditions. However, PWHT reduces the strength of base metal and weld metal. Therefore, the repaired disk was heat treated locally to avoid degradation of the base metal. An optimum PWHT temperature satisfying both weld metal strength and HAZ hardness was determined. Selecting Weld Metal Ni-Mn-Mo steel (AWS:ER110S-G) and 2.25Cr-1Mo steel (AWS:ER90S-G) were proposed as the welding material because of their excellent welding characteristics, strength, and toughness. Table 4 shows the comparison of these material properties. Ni-MnMo steel is suitable as weld metal for the disk that requires high strength at lower temperatures. However, 2.25Cr-1Mo steel is suited for higher temperatures because of its creep strength. The latter is preferable because it can be utilized for the entire rotor stage. Thus 2.25Cr-1Mo steel was ultimately selected as the metal for repair welding. Both materials were welded and evaluated in testing conducted on a laboratory scale test. Table 4. Weld Metal for Repair.
Figure 6. Appearance of Welded Test Piece in Laboratory Scale Test. Table 5. Welding and PWHT Conditions for Test Pieces.
(302-482oF)
(2.4-2.8 inch/min)

o
(1112oF)
o o o o o o
(482-572oF)
(36oF/h) (1085oF) (1112oF) (45oF/h) (482oF/h)
Figure 7 shows the relationship between stress and time to SCC initiation reported by Speidel and Bertilsson (1984). Based on these data, it is estimated that the critical stress that causes SCC after 105 hours of operation is 420 MPa in a normal steam environment. Considering the safety factor, the target residual stress on the weld metal was set at 200 MPa. The specimens for evaluation were taken in a welded condition and the heat treated test pieces. Figure 8 indicates the locations of each specimen in the test piece. Tensile, impact, hardness measurement, microstructure observation, and SCC tests were performed. Evaluation of Laboratory Scale Test Pieces Welding Defects Welded test pieces were inspected by ultrasonic testing. Defects exceeding the criteria [1.6 mm (.06 inch) in diameter] were not found. According to the analysis by fracture mechanics, the critical defect size for cracking is much larger then the criterion. Therefore, it can be concluded that the defect in weld was small enough to prevent fracture.
19
(87,000 psi)
Table 6. Mechanical Properties of Specimens (Weld Metal: 2.25Cr1Mo Steel).

5
Base metal Vickers Max hardness, HV Min 324 277 585oC (1085oF), 10h HAZ 370 Weld metal 286 243 793 (115,000) 792 (114,900) 645 (93,500) 642 (93,100) 28.0 28.0 71.3 69.7 201 (148) 153 (113) Target for weld metal Requirement for base metal 350 720 (104,400) 785 (113,900) 600 (87,000) 638 (92,500) 15 15 40 40 39 (29) 39 (29)
a y
(60,900 psi)
Tensile strength MPa (psi) Yield strength MPa (psi) Elongation % Reduction of area % Charpy 2mm Longitudinal direction Thickness direction
927 (134,500) 938 (136,000) 922 (133,700) 971 (140,800) 785 (113,900) 817 (118,500) 785 (113,900) 850 (123,300) 23.2 23.2 66.5 66.5 77 (57) 112 (83) 24.0 23.2 68.9 64.0 191 (141) 201 (148) 87 (64) 98 (73) 67 (50)
(29,000 psi)
a y
Figure 7. Estimated Critical Residual Stress for SCC. (Courtesy Speidel and Bertilsson, 1984)
A - A
75
(0.08inch)
U Notch J (ft-lbf)
600oC (1112oF), 10h Base Metal Vickers Max hardness, HV Min Tensile strength MPa (psi) 297 262 HAZ 339 Weld metal 309 210 722 (104,700) 725 (105,200) 602 (87,300) 602 (87,300) 28.5 28.5 74.3 74.3 229 (169) 197 (145) -
:HOG PHWDO +$= %DVH PHWDO
Target for weld metal Requirement for base metal 350 720 (104,400) 785 (113,900) 600 (87,000) 638 (92,500) 15 15 40 40 39 (29) 39 (29)
932 (135,200) 913 (132,400) 951 (137,900) 921 (133,600) 791 (114,700) 795 (115,300) 812 (117,800) 808 (117,200) 22.5 22.5 66.5 66.5 91 (67) 143 (106) 24.5 24.0 69.7 68.2 83 (61) 99 (73) 83 (61) 94 (69) 99 (73)
,PSDFW WHVW
7HQVLOH WHVW
A B - B
Yield strength MPa (psi) Elongation % Reduction of area % Charpy 2mm

(0.08inch)
C - C
$V ZHOG VDPSOH
+DUGQHVV 0LFURVWUXFWXUH REVHUYDWLRQ
Longitudinal direction Thickness direction
+HDW WUHDWHG VDPSOH
+DUGQHVV 0LFURVWUXFWXUH REVHUYDWLRQ
,PSDFW WHVW
6&& WHVW
U Notch J (ft-lbf)
Figure 8. Locations of Specimens for Evaluation. Mechanical Properties Table 6 shows results of the mechanical tests on test pieces welded using 2.25Cr-1Mo steel and heat treated for 10 hours at 585C and 600C (1085F and 1112F). Figure 9 and Figure 10 indicate distribution of hardness on the welded test piece and the heat treated piece at 600C (1112F). Hardness, tensile properties, and impact value on both heating temperatures satisfied the target for the weld metal and the requirement for base metal, except for hardness heated at 585C (1085F). Figure 11 shows the relationship of maximum hardness and minimum tensile strength with PWHT temperature. Hardness and tensile strength of the weld metal tends to become weaker as the welding temperature becomes higher. On the other hand, tensile strength of base metal is almost constant in the temperature range tested. This tendency comes from the high temperature properties of base metal. The maximum hardness of HAZ satisfies the target at lower temperatures. On the other hand, the minimum tensile strength of the weld metal satisfies the target at higher temperature. Therefore, temperature range that satisfies both targets is between 595C and 600C (1103F and 1112F) when welding with 2.25Cr-1Mo steel. Table 7 shows the results of mechanical tests for a test piece welded with Ni-Mn-Mo steel. The tensile strength of Ni-Mn-Mo weld metal is higher than the one of 2.25Cr-1Mo heat treated at the same time and at the same temperature [10 hours, 600C (1112F)]. This shows that the Ni-Mn-Mo weld metal has an advantage in strength over 2.25Cr-1Mo at low temperatures. Sensitivity for Stress Corrosion Cracking An SCC test was performed in an accelerating NaCl solution for the test piece, for the 2.25Cr-1Mo weld metal. Table 8 shows the test conditions. Figure 8 shows the location of specimens on the
Figure 9. Distribution of Hardness on Welded Test Piece.
Figure 10. Distribution of Hardness on Heat Treated Test Piece. test piece. Stress around the yield point of material was applied by U-bending. The surface of the test pieces was observed up to 1000 hours. Figure 12 indicates the appearance of the HAZ specimen after the 1000 hour test. Although the surface became rough by corrosion, cracks did not occur.
20
(145,000 psi)
Weld metal: 2.25Cr-1Mo steel Holding time: 10h
(87,000 psi)
Tensile strength of Base metal Tensile strength of Weld metal Hardness of Weld metal
(1058oF) o (1130oF)
Figure 11. Relationship Between PWHT Temperature and Mechanical Properties. Table 7. Mechanical Properties of Specimens (Weld metal: Ni-MnMo Steel).
600oC (1112oF), 10h Base metal Vickers Max hardness, HV Min Tensile strength MPa (psi) Yield strength MPa (psi) Elongation % Reduction of area % Charpy 2mm
(0.08inch)
Figure 12. Surface of Specimens after 1000 Hours in NaCl Solution.
HAZ 340
Weld metal 315 280 858 (124,400) 852 (123,600) 766 (111,100) 776 (112,500) 27.8 28.4 67.3 69.7 130 (96) 138 (102) -
Target for weld metal Requirement for base metal 350 720 (104,400) 785 (113,900) 600 (87,000) 638 (92,500) 15 15 40 40 39 (29) 39 (29)
313 292
932 (135,200) 921 (133,600) 932 (135,200) 928 (134,600) 796(115,500) 799(115,900) 24.6 24.0 68.9 68.2 180 (133) 179 (132) 828 (120,100) 826 (119,800) 24.2 23.0 64.8 68.2 196 (145) 178 (131) 96 (71) 97 (71) 77 (57)
Longitudinal direction Thickness direction
Figure 13. Welded Disk on Real Scale Test.
U Notch J (ft-lbf)
Table 8. Conditions of SCC Test.
(1112oF)
Figure 14. Heater Layout on Local PWHT for Welded Disk. Evaluation of the Welded Rotor Deformation
o
(353oF)
Measurement was performed for bowing of the shaft and other deformations due to welding and PWHT. No unfavorable deformation was found. Welding Defects
Welding and Heat Treatment for the Rotor Disk An actual rotor was welded and heat treated for a real scale test. Figure 13 shows the appearance of a welded disk. 2.25Cr-1Mo was welded 60 mm (2.36 inches) in height by GTAW. Welding speed was kept constant to prevent bowing of the rotor shaft. Local heating was applied for the PWHT. Layout of the heater was decided based on the result of a finite element analysis to reduce the thermal stress around the edge of the heating zone due to temperature distribution. Figure 14 shows the schematic view of the heater layout. The rotor was suspended vertically during heating to avoid bowing due to the influence of gravity on the horizontal rotor layout.
The welded disk was inspected by ultrasonic testing. Defects exceeding the criterion [1.6 mm (.06 inch) in diameter] were not found. According to the analysis of fracture mechanics, the critical defect size for cracking is much larger then the criteria. Therefore, it can be concluded that any defects in the weld were small enough to prevent fractures. Residual Stress Figure 15 shows the appearance of the residual stress measurement. A strain gauge was attached to a square weld metal area, and a groove of 1 mm (.04 inch) in depth was cut around this area.
21
(1292oF) (1112oF)
700 600 500 400 300 200 100
A B D E F G, H C
0
(662oF)
10
20 30 Time[h]
40
50
Figure 17. Estimated Temperature Distribution on Local PWHT for Disk. Figure 15. Residual Stress Measurement for Repaired Disk. Measured strain after cutting was translated into the released residual stress. Measured residual stress was around 200 MPa (tension) in a circumferential direction. This corresponds to the target value required to prevent SCC and the value estimated through the finite element analysis. Mechanical Properties Hardness on the surface of the weld metal was measured. Figure 16 indicates the measured hardness. The measured hardness of 240 to 270 HV is almost the same as that of the final bead on the laboratory test piece heated at 600C (1112F) as shown in Figure 10. Therefore, it can be estimated that the hardness of HAZ was also reduced within 350 HV. stress is around 200 MPa on the weld metal. This means that the weld metal is compressed by the disk during heating. This stress is turned into tensile stress after cooling down. It is presumed that the compression stress during heating changes into tensile stress by creep deformation.
(14,500 psi)
(27,700 psi)
(1112oF)
450 0o 180o 350 90o 270o Target

(-29,000 psi)
Measured surface
Weld metal
60 (2.36 inch)
Figure 18. Estimated Circumferential Stress Distribution on Local PWHT for Disk. Temperature and Stress Distribution of Post Weld Heat Treatment for the Shaft Portion Figure 19 shows the temperature distribution at the start of temperature hold and the temperature change at each spot for positioned heaters that satisfy the uniform temperature on the whole surface of the weld metal. The center of the shafts heated portion is more than 500C (932F). This means there is a higher possibility of unfavorable deformation by welding of the shaft and PWHT than that for the disk.
250
60 (2.36 inch)
150 0 10 20 30 40 50 60
(2.36 inch)
Figure 16. Hardness Distribution on Real Scale Disk.
NUMERICAL ANALYSIS
A finite element analysis was performed to set the optimum heater layout at local PWHT. Temperature distribution and thermal stress by elastic analysis were estimated through a heating process that included heating and cooling. Heater layout was adjusted to maintain a target thermal stress of less than 200 MPa. Analysis was performed for the heating of both the disk and the shaft. Temperature and Stress Distribution on the Post Weld Heat Treatment Area of the Disk Figure 17 shows the temperature distribution at the start of temperature hold after heating and the temperature change at each spot for positioned heaters. The temperature at the root of the heated disk and adjacent disks was less than 450C (842F). Therefore, temperature at the root of the disk should be maintained at more than 450C (842F) to satisfy the target residual stress. Figure 18 shows the distribution of circumferential stress around the heated disk at the start of temperature hold. Circumferential
(1292oF) (1112oF)
$ % & ' ( ) *
(392oF)
Figure 19. Estimated Temperature Distribution on Local PWHT for Shaft.
22
Figure 20 shows the distribution of circumferential stress around the heated shaft at the start of temperature hold. The stress around weld metal is low and the compressive stress is 75 MPa maximum in a circumferential direction around the edge of the heater. Therefore, it can be determined that the residual tensile stress on the shaft is lower than that on the disk. Bending residual stress around 80 MPa can be estimated on the disk adjacent to the heated portion. This is almost the same as that on the shaft.
(7,300 psi)
Figure 21. Welding of the Rotor Disk.
(10,700 psi) (-14,500 psi)
Figure 20. Estimated Circumferential Stress Distribution on Local PWHT for Shaft.
TYPICAL REPAIR EXPERIENCE

Experience in overlay welding for repairs is listed in Table 9. These overlay welds were performed in emergency cases. In considering rotor materials and the required strength, overlay welding was applied to a portion of blade root grooves and rotor disks by selecting suitable weld metals. In this paper, two typical repair experiences for mechanical drive turbines are introduced. Table 9. Application of Rotor Overlay Welding.
No. Use Welded Part Root Groove Partially Root Groove (Partially Disk Side Plane Circumference Whole Root Groove Circumference Rotor Disk LP Section (1 Stage) Rotor Disk HP and LP Section (4 Stages) Rotor Materials Ni Cr Mo V Cr Mo V 2.5Ni Cr Mo V Method
Figure 22. Hardness of the Welded Metal for the Actual Rotor. repair all stages. However, it was also necessary to start the plant very quickly, and the time for repair was limited. In discussion with the customer regarding minimum required power in the plant, turbine performance and blade strengths of each stage were decided upon with the customer, and as a result, the number of stages were reduced and heat drop distribution was changed for the LP section. Figure 23 shows the condition of welding repair on the rotor disks. Figure 24 shows rotor residual warpage check after welding was completed. Rotor residual warpage after welding and final machining were within the criteria of 0.005 mm (.0002 inch), and the blades for each stage were assembled as shown in Figure 25.
1 2 3 4
Ship Power Generation Power Generation Power Generation Mechanical Driver Mechanical Driver
Overlay Welding Removal of Cracks Overlay Welding Overlay Welding Removal of the Whole Root Groove Overlay Welding Removal of Damaged Disk Overlay Welding Removal of Damaged Disks Overlay Welding
1.25Ni Cr Mo V
1.25Ni Cr Mo V
Welding Repair for a Large Rotor First, Figure 21 shows the case for a large rotor in a charge gas compressor drive steam turbine at an ethylene plant. The ninth stage disk of this turbine had heavy frictional contact with the diaphragm, and the blades and one side of the disk were damaged. The damaged portion was cut off and repaired successfully based on the results of the above laboratory test and analysis. Figure 22 shows hardness check results in the repair process used to confirm that the welding conditions are the same as in the test results. Welding Repair for High-Speed Rotor Second, as explained in the catastrophic damage of Figure 4 and Figure 5, the case for a high-speed machine is explained in detail. All stages of this rotor were damaged; thus it was necessary to Figure 23. Welding Repair of Rotor Disks. Figure 26 shows the comparison between the turbine performance curves at the original 9555 kW and after repairs at 6700 kW, with the number of stages reduced by cutting off the last fifth,
23
CONCLUSIONS
This paper introduces typical steam turbine engine damage modes and causes, and actual experiences in repairing and reviving catastrophically damaged turbine rotors through special welding procedures, based on risk analysis for rotor repair from a systematic viewpoint. Root cause analysis in the catastrophic failure process is explained, regarding the integrated control system for governor control and backup system. The practical repair techniques are developed based on element tests to find optimized welding conditions and detailed strength calculation to confirm the integrity, and heat transfer analysis for proper heat treatment process conditions. These basic procedures are discussed to show useful data. The case study for this optimization is discussed by showing thermodynamic calculation, performance, and repair schedule. Finally, this turbine was uniquely modified in order to balance required power and repair time, in order to quickly restart the plant in an emergency case, to acquire customer satisfaction. The revived extraction condensing turbine rotor was placed back into the casing and operated successfully at full load.
Figure 24. Rotor Residual Warpage Check.
REFERENCES
Speidel, M. O. and Bertilsson, J. E., 1984, International Brown Boveri Symposium on Corrosion in Power Generating Equipment, p. 331.
ACKNOWLEDGEMENT
The authors gratefully wish to acknowledge the following individuals for their contribution and technical assistance in analyzing and experiments: O. Isumi and M. Wakai of Mitsubishi Heavy Industries Ltd.
Figure 25. Blade Assembly after Welding Repair. sixth, and seventh stages. Blades and diaphragm strengths were checked and exhaust vacuum pressure was limited. This repaired and revived the rotor of the extraction condensing turbine, and it was actually placed back into the casing and operated successfully at a condition with low vibration and full load for three months until the new rotor was prepared.
Figure 26. Actual Operation of Repaired Rotor.
24
ROTORDYNAMIC DESIGN OF CENTRIFUGAL COMPRESSORS IN ACCORDANCE WITH THE NEW API STABILITY SPECIFICATIONS
by John C. Nicholas
Owner, Director, Chief Engineer Rotating Machinery Technology, Inc. Wellsville, New York
and John A. Kocur

Machinery Engineer ExxonMobil Research and Engineering Company Fairfax, Virginia
ABSTRACT
John C. Nicholas is the owner, Director, and Chief Engineer of Rotating Machinery Technology, Incorporated, a company that repairs and services turbomachinery, and manufactures bearings and seals. He has worked in the turbomachinery industry for 28 years in the rotor and bearing dynamics areas, including five years at IngersollRand and five years as the Supervisor of the Rotordynamics Group at the Steam Turbine Division of Dresser-Rand. Dr. Nicholas, a member of ASME, STLE, and the Vibration Institute, has authored over 40 technical papers, concentrating his efforts on tilting pad journal bearing design and application. He received his B.S. degree from the University of Pittsburgh (Mechanical Engineering, 1968) and his Ph.D. degree from the University of Virginia (1977) in rotor and bearing dynamics. Dr. Nicholas holds several patents including one for a spray-bar blocker design for tilting pad journal bearings and another concerning bypass cooling technology for journal and thrust bearings. John A. Kocur, Jr., is a Machinery Engineer in the Plant Engineering Division at ExxonMobil Research & Engineering, in Fairfax, Virginia. He has worked in the turbomachinery field for 20 years. He provides support to the downstream business within ExxonMobil with expertise on vibrations, rotor/aerodynamics, and health monitoring of rotating equipment. Prior to joining EMRE, he held the position of Manager of Product Engineering and Testing at Siemens Demag Delaval Turbomachinery. There Dr. Kocur directed the development, research, engineering, and testing of compressor and steam turbine product lines. Dr. Kocur received his BSME (1978), MSME (1982), and Ph.D. (1991) from the University of Virginia and an MBA (1981) from Tulane University. He has authored papers on rotor instability and bearing dynamics, lectured on hydrostatic bearings, has been a committee chairman for NASA Lewis, and is a member of ASME. Dr. Kocur holds a patent on angled supply injection of hydrostatic bearings.
25
The American Petroleum Institute (API) has recently implemented new rotordynamic stability specifications for centrifugal compressors. The specifications consist of a Level I analysis that approximates the destabilizing effects of the labyrinth seals and aerodynamic excitations. A modified Alfords equation is used to approximate the destabilizing effects. If the compressor fails the Level I specifications, a more sophisticated Level II analysis is required that includes a detailed labyrinth seal analysis. Five modern high-pressure example centrifugal compressors are considered along with a classic instability case, Kaybob. After applying API Level I and Level II stability analyses and reviewing the results, design changes are made to stabilize the compressors, if necessary. For these cases, the API stability specifications are used to identify the component with the greatest impact on rotor stability. Specifically, the balance piston seal and impeller eye seals are analyzed. The suitability of applying the modified Alfords equation to compressors with multiple process stages is examined and compared to the full labyrinth seal analysis. Important aspects of labyrinth seal analyses are discussed, such as seal clearance effects, inlet swirl effects, and converging, diverging clearance effects. Finally, a modal approach to applying the labyrinth seal calculated cross-coupled forces is presented. For all five example compressors, the modified Alfords force was determined to produce the worst case stability level compared to labyrinth calculated forces.
INTRODUCTION
Centrifugal compressor instability became a major problem in the 1960s due to increased speeds and power ratings. Unstable compressors exhibited a high subsynchronous vibration whose vibration frequency coincided with the rotors first fundamental natural frequency. Two famous and classic centrifugal compressor instability cases from the early 1970s are referred to as Kaybob (Smith, 1975; Fowlie and Miles, 1975) and Ekofisk (Geary, et al., 1976). Both problems occurred onsite and the solution involved a costly and time-consuming effort ultimately requiring rotor redesigns. As a result of these experiences, the evaluation of rotor system stability has become an essential part of rotordynamic analyses and rotating machinery design. Most often, the lowest or first mode, corresponding to the rotors first fundamental natural frequency, is the mode that is reexcited causing the subsynchronous vibration and rotor instability. The primary results of a stability or damped
26
natural frequency analysis are the stability prediction from the real part and the predicted instability frequency from the imaginary part of the solution roots or eigenvalues. Some of the earliest stability work includes two classic publications by Gunter where basic stability methodology (Gunter, 1966) and internal friction excitation (Gunter, 1967) are discussed. Initial attempts at developing a stability computer code were made by Lund (1965). Lunds code was not easy to use and it was said that if one knew the approximate answer the code would converge on the solution. Another early computer code was by Ruhl and Booker (1972). They presented both finite element and transfer matrix solution techniques using a Mullers method solver. For lightly damped systems, their transfer matrix solution analysis worked fine but for more heavily damped structures, such as fluid film bearing supported turbomachinery, the programs analysis methodology produced incorrect and false modes. These shortcomings were overcome in Lunds landmark stability paper (1974). Lund not only outlines a detailed transfer matrix solution procedure, but also describes how the stability results may be presented to study machine design parameters. His transfer matrix solution is able to search for the first several modes very efficiently, solving for the most important lowest frequency modes first, although in random order. Ruhls transfer matrix analysis was updated to include flexible supports by Bansal and Kirk (1975), replacing Mullers solver with a Cauchy-Rieman condition finite difference algorithm plus a Newton-Raphson search solution specified by Kirk (1980). With minor differences, this was essentially the same solution as used by Lund (1974). Both procedures work but will occasionally skip modes especially when asymmetric flexible supports are included at the bearing locations. Other transfer matrix computer programs have been developed based on Lunds original analysis such as Barrett, et al. (1976). More recent computer codes are based on a finite element solution that successfully extracts all of the correct modes. These finite element code authors include Nelson and McVaugh (1975), Rouch and Kao (1979), Edney, et al. (1990), Chen (1996), and Ramesh and Kirk (1993). One disadvantage of the finite element analysis was that the problem size increases dramatically with the number of elements used to model the rotor, resulting in longer run times compared to the transfer matrix method. However, this is no longer an issue with the fast processing speed of modern personal computers. Additionally, some methods require that all roots be found, extracting the highest natural frequency first and ending with the lowest, most important mode. Alternate solution techniques are now available that extract the lowest eigenvalues first (Murphy and Vance, 1983). The most recent API Standard 617, Seventh Edition (2002), for centrifugal compressors includes stability acceptance criteria along with analytical procedures. The stability specification is segmented into two parts: a simplified Level I analysis and a detailed Level II analysis. The Level I analysis is meant to be a screening process in which a quick and simple analysis can be conducted to filter out machines that are well away from the instability threshold. Level I utilizes a modified Alfords equation (Alford, 1965) to estimate the destabilizing forces. The Tutorial on Rotordynamics, API Technical Publication 684 (2005), presents a discussion of the modified Alfords equation. The more involved Level II analysis requires that the dynamic properties of labyrinth seals be included implicitly through the use of an appropriate labyrinth seal code. Initial labyrinth seal computer codes based on the Iwatsubo, et al. (1982), solution were developed by Childs and Scharrer (1986b). Kirk (1988a, 1988b, 1990) further extended the work of Childs and Scharrer. First published experimental results for gas labyrinth stiffness coefficients are presented in Benckert and Wachter (1979, 1980). Damping coefficients were not obtained since only static pressure measurements of the individual chambers were made. Thieleke
and Stetter (1990) as well as Kwanka, et al. (1993), have also carried out similar efforts. The research summarized in Childs (1993) provided the first measurements of labyrinth seal damping coefficients. While the results gave the first comprehensive basis for comparison against predictions, Childs and Ramsey (1991) revealed the importance of testing at or near the application conditions. The seal test rig was further extended toward this goal by Childs and Scharrer (1986a) and then again by Elrod, et al. (1995). Wagner and Steff (1996) further expanded the existing experimental knowledge database to geometries and gas conditions matching industrial applications, namely, pressure differential, size, and speed. Pressures of 70 bar (1015 psi) were possible at surface speeds of up to 157 m/s (515 f/s). The main objective of this paper is to examine the stability results for several industrial representative centrifugal compressors. The API Level I modified Alfords cross-coupling force calculation is examined to determine if it is indeed a conservative estimation of the compressors destabilizing forces by comparing it to the API Level II labyrinth seal calculated forces based on Kirk (1988a, 1988b, 1990). Also, specifics of the labyrinth analysis are examined to determine what parameters are key in determining centrifugal compressor stability. Some of the parameters examined include bearing clearance tolerance range, labyrinth seal clearance, and labyrinth seal inlet swirl effects.
THE KAYBOB INSTABILITY

As a historical perspective, a brief summary of the Kaybob instability will be presented (Smith, 1975; Fowlie and Miles, 1975). This nine-stage low-pressure natural gas injection compressor was commissioned in 1971 in Alberta, Canada. Key operating parameters are summarized in Table 1. The maximum continuous speed (MCS) is 11,400 rpm with 18 MW gas at 1150 psi inlet and 3175 psi discharge. The bearing span, Lb, to midshaft diameter, Dms, ratio is 13.2, indicating a very flexible shaft. The compressors cross section is shown in Figure 1. Table 1. Example Centrifugal Compressor Design Data.
Parameter Type # Stages Configuration Speed (rpm) Rotor Weight (lbm) Horse Power (hp) Pin (psi) Pdisch (psi) Mole Weight Bearing Span (in) Dms (in) Lb/Dms Bearings Seals Balance Piston Alfords Qa (lbf/in) Kaybob Example #1 Example #2 Example #3 Example #4 Example #5 Mixed Propane Injection Hydrogen Refrigeration Injection #1 Injection #2 Refrigeration 9 10 3 4 4 5 Back-Back Straight Straight Straight Straight Straight 11,400 10,750 3,000 12,700 9,900 3,000 1,340 540 422 44,730 36,689 12,000 65,000 30,000 11,000 93,000 1,150 800 20 1,200 5,600 60 3,175 1,725 100 3,300 9,000 320 18 6 44 17 25 25 59.7 65 230 59.3 47.8 222 4.5 5.75 22.4 6.69 6.59 21.8 13.2 11.3 10.3 8.86 7.26 10.2 5 Pad Tilt 5 Pad Tilt 5 Pad Tilt 4 Pad Tilt 5 Pad Tilt 5 Pad Tilt Dry Gas Dry Gas Dry Gas Dry Gas Oil Dry Gas Tooth Laby Honeycomb Tooth Laby Tooth Laby Tooth Laby Tooth Laby 57,250 23,046 68,015 72,854 41,907
Figure 1. Kaybob Injection Compressor. (Courtesy, Smith, 1975)
27
The severity of the instability may be seen in the orbit of Figure 2. Note that the outline of the five pad tilting pad bearing is clearly evident in the 6.0 by 9.0 mils peak-to-peak orbit. From Figure 3, the 6.3 mil instability is obviously subsynchronous, reexciting the compressors first critical speed.
Original Rotor Lb = 59.7"
1st Modification Lb = 53.4"
2nd Modification Lb = 53.4" w/ Increased Shaft Diameter
Figure 4. Kaybob Rotor Modifications. (Courtesy, Smith, 1975)
Figure 2. Kaybob Instability Orbit. (Courtesy, Smith, 1975)
Figure 5. Kaybob Impeller Modification. (Courtesy, Fowlie and Miles, 1975) The log dec is defined as the natural logarithm of the ratio of any two successive amplitudes. Referring to Figure 6, the log dec is defined as:
Figure 3. Kaybob Instability. (Courtesy, Fowlie and Miles, 1975) Attempts to eliminate the instability included bearing redesigns, oil seal modifications, labyrinth seal modifications, balance piston modifications, a vaneless diffuser retrofit, a squeeze film damper retrofit, and, finally, at least two rotor redesigns (Figure 4). The second rotor redesign included increasing the midshaft diameter. Initially, existing impeller forgings were used, cutting and welding the impeller hub to increase the impeller inside diameter to accommodate the increase in shaft diameter (Figure 5). Clearly, this effort was extremely costly and time consuming. However, it was instrumental, along with the Ekofisk instability, in providing motivation for improved analytical capabilities, ultimately resulting in the existing stability and labyrinth seal codes as well as the new API stability specification.
= ln
X1 X2
(1)
For stable systems, with a positive rate of decay, the log dec is positive. For unstable systems with a negative rate of decay, the log dec is negative. Stable systems with positive log dec values contain sufficient damping to overcome an initial excitation. The resulting displacements will dissipate over time. Conversely, unstable systems with negative log dec values do not contain sufficient damping to overcome the excitation, resulting in increasing displacements over time. The log dec can also be related to the real, s, and imaginary, d, parts of the eigenvalue as:
LOGARITHMIC DECREMENT
The key parameter in stability analyses and the API stability acceptance criteria is the logarithmic decrement or log dec. The log dec is a measure of the rate of decay of free oscillation and is a convenient way to determine the amount of damping present in the system. Greater damping values produce faster decay rates and more stable systems.
=
=
2 s
d
60 s Nd
(2a)
(2b)
where d and Nd are the damped natural frequency in rad/sec and rpm, respectively.
28
X
1 2
The shaft seals, with a much lower pressure drop, are neglected. Additionally, the seal flow enters the seal from a stationary part. Thus, the inlet swirl value is low compared to the eye seal swirl (Figure 7). High inlet swirl results in high Q values.
High Swirl, High P
Figure 6. Stable Vibration Wave Form.
Low Swirl, Low P
EXAMPLE #1 12,000 HP, 10 STAGE HYDROGEN COMPRESSOR

The first example is a 10 stage 12,000 hp hydrogen centrifugal compressor with a 65 inch bearing span and a Lb/Dms = 11.3. The rotor weighs 1340 lbm and operates at an MCS of 10,750 rpm with dry gas seals and five pad tilting pad journal bearings (Table 1). The inlet pressure is 800 psi with a 1725 psi discharge pressure and a gas mole weight of 6.0. Base log dec values with zero aerodynamic cross-coupling, Q, are 0.26 and 0.40 for minimum and maximum bearing clearances, respectively. From API Specification 617, Seventh Edition (2002), the modified Alfords equation is:
Figure 7. Centrifugal Compressor Labyrinth Seals. With the total eye seal Q lumped at the rotor center to be conservative, the resulting log dec values are 0.08 and 0.28 for minimum and maximum bearing clearances, respectively. Instead of estimating the labyrinth eye seal inlet swirl, it can be calculated by modeling the gap between the impeller face and the stator as in Figure 8 (Kirk, 1990). With the inclusion of impeller gap modeling, the inlet swirl is calculated at 0.52 with a resulting Q value for all eye seals of 7600 lbf/in. With this total eye seal Q lumped at the rotor center, the resulting log dec values are 0.17 and 0.35 for minimum and maximum bearing clearances, respectively. As with the Level I analysis, the API stability acceptance criterion for Level II is a log dec greater than 0.1. Thus, this compressor passes API over the bearing clearance tolerance range.
Q=
where: c = hp = Hc = Dc = N = ratio =
c (63,000)hp
Hc ( Dc ) N
( ratio )
(3)
3.0 12,000 hp (total all stages) Varies stage-to-stage Varies stage-to-stage 10,750 rpm 1.5 (total across compressor)
Q is calculated for each stage using the above values. Since the stage density ratio and horsepower were not available, the stage horsepower was assumed to be one-tenth of the total value shown above. The stage density ratio was assumed to be the total density ratio shown above, raised to the power of 1/10. The impeller discharge width and impeller diameter are stage-to-stage variables. The summation of all 10 Q values is the API Alford calculated or anticipated cross-coupling value of Qa = 57,250 lbf/in. With Qa lumped at the rotor midspan, the resulting log dec values are 0.40 and 0.23 for minimum and maximum bearing clearances, respectively. The API stability acceptance criterion is a log dec greater than 0.1. Thus, a Level II analysis is required. Prior to any analysis, it had already been decided to use a honeycomb seal on the balance piston. Thus, the stability results reported above include the honeycomb seal dynamic properties (Scharrer and Pelletti, 1994). With the labyrinth seal geometry, stage gas properties, and stage pressures used as input, a labyrinth seal analysis (Kirk, 1990), is conducted for each of the 10 impeller eye seals. A gas swirl value at the seal inlet is assumed to be 0.6 (60 percent of rotational speed). To be conservative, minimum eye seal clearances are used considering the machining tolerance range for the seal and the seal sleeve. The resulting total labyrinth calculated Q for all 10 eye seals is 15,700 lbf/in.
Figure 8. Labyrinth Seal Modeling for Inlet Swirl Calculation. These results are summarized in Table 2 and shown in Figures 9 and 10 for minimum and maximum bearing clearances, respectively. Also of note from Figures 9 and 10 are the Q values for zero log dec, Q0. These are 22,500 and 40,000 lbf/in for minimum and maximum bearing clearances, respectively. Clearly for this application, the predicted stability levels using the modified Alfords force is conservative.
29
Table 2. Hydrogen Compressor Stability Summary.

Bearing Clearance Minimum Minimum Minimum Minimum Minimum Maximum Maximum Maximum Maximum Maximum Q (lbf/in) 0 57,250 15,700 7,600 22,500 0 57,250 15,700 7,600 40,000 Q Description Base Qa, Alford Laby Analysis Laby Analysis Q0 (Q for Log Dec = 0) Base Qa, Alford Laby Analysis Laby Analysis Q0 (Q for Log Dec = 0) c 3.0 3.0 Eye Seal Inlet Swirl 0.60 Estimated 0.52 Calculated 0.60 Estimated 0.52 Calculated Nd (cpm) 3,545 3,590 3,551 3,547 3,558 3,361 3,321 3,333 3,348 3,318 Log Dec, 0.26 -0.40 0.08 0.17 0.00 0.40 -0.23 0.28 0.35 0.00
Calculated Swirl = 0.52
Estimated Swirl = 0.6
Min Laby Clearance
Laby Q Swirl = 0.52
Laby Q Swirl = 0.6
Level II Analysis Required a < 0.1 Q0
Figure 11. Hydrogen Compressor Q Versus Laby Eye Seal Inlet Swirl.
Balance Piston Honeycomb Seal Included Min Bearing Clearance w/ Eye Seal Swirl Brakes Min Laby Clearance
Min Laby Clearance Balance Piston Honeycomb Seal Included
a = - 0.40
Alfords Qa
Calculated Swirl = .52 Estimated Swirl = .6
Figure 9. Hydrogen Compressor StabilityMinimum Bearing Clearance.

Laby Q Swirl = 0.52 Laby Q Swirl = 0.6 Level II Analysis Required a < 0.1
Figure 12. Hydrogen Compressor Stability Versus Laby Eye Seal Inlet Swirl.
Q0
Laby Inlet Swirl = 0.6
Max
Min Laby Clearance Balance Piston Honeycomb Seal Included
Alfords Qa a = - 0.23
Nom Min
Figure 10. Hydrogen Compressor StabilityMaximum Bearing Clearance. Eye Seal Inlet Swirl Effects The effect of inlet swirl on total eye seal Q is shown in Figure 11. The corresponding effect on stability is presented in Figure 12. Note that the inclusion of swirl brakes (Childs and Ramsey, 1991; Moore and Hill, 2000), on the eye seals with an estimated inlet swirl value of 0.3 increases the log dec to 0.28 for minimum bearing and eye seal clearances. Eye Seal Clearance Effects The effect of the seal clearance on total eye seal Q is shown in Figure 13. The machining tolerance range is 16 to 20 mils diametral. Assuming that the minimum eye seal clearance decreases by 4 mils diametral due to centrifugal expansion of the impeller, the minimum operating clearance of 12 mils diametral is also shown on Figure 13. The corresponding effect on stability for an inlet swirl of 0.6 and minimum bearing clearance is presented in Figure 14.
Min Operating
Figure 13. Hydrogen Compressor Q Versus Laby Eye Seal Clearance. The effects on stability of converging and diverging eye seal clearances are illustrated in Figure 15. Assuming a constant clearance distribution for all eye seal teeth of 12 mils diametral, the resulting log dec is 0.6. If the clearance converges from seal inlet (16.0 mils diametral) to seal discharge (8.0 mils diametral), the log dec decreases to 0.11. Conversely, a divergent clearance from inlet (8.0 mils diametral) to discharge (16.0 mils diametral) produces a log dec of 0.23. These values assume minimum bearing clearance and an inlet swirl value of 0.6. Clearly, a divergent clearance distribution is preferable for improved stability. While the eye seal principal stiffness is negative
30
Balance Piston Honeycomb Seal Included Min Bearing Clearance Laby Inlet Swirl = 0.6
Nom Min
Min Operating
Figure 14. Hydrogen Compressor Stability Versus Laby Eye Seal Clearance.
Balance Piston Honeycomb Seal Included Min Bearing Clearance Laby Inlet Swirl = 0.6 Diverging Cs
Min Operating Eye Seal Cs = 6.0 mils dia.
Constant Cs
Converging Cs
Figure 15. Hydrogen Compressor StabilityEffect of Clearance Profile. for a diverging clearance, this effect is minimal. Also, the leakage actually decreases for the diverging clearance case compared to a constant clearance. For this compressor, the impeller growth with speed increases from the eye seal discharge location (impeller eye) to the eye seal inlet location resulting in a divergent seal clearance. Balance Piston Honeycomb Seal Cell Clogging and Inlet Swirl Effects One problem with honeycomb seals is that the honeycomb cells may clog or fill with debris. Other nonlabyrinth type seals are also known to clog, such as hole pattern seals. Hole pattern seal test results with plugged holes may be found in Moore and Soulas (2003). This clogged cell effect for honeycomb seals is illustrated in Figure 16 for minimum bearing clearance and three different honeycomb seal inlet swirl values: 0.8, a pessimistic value; 0.6, a more realistic value; and 0.3, simulating the inclusion of a swirl brake. The plot shows that as the honeycomb cells fill or clog, stability decreases. For the 0.6 inlet swirl case, as long as there are less than 48 percent of the cells filled, the log dec is greater than 0.1, the API acceptance value. For inclusion of a swirl brake (0.3 inlet swirl), the log dec is greater than 0.1 as long as there are less than 60 percent filled cells.
EXAMPLE #2 65,000 HP, 3 STAGE PROPANE REFRIGERATION COMPRESSOR

The second example is a three stage, three section 65,000 hp propane compressor in a refrigeration service with a 230 inch
Max
Figure 16. Hydrogen Compressor Stability Versus Percent Filled Balance Piston Honeycomb Cells. bearing span and a Lb/Dms = 10.3. The rotor weighs in excess of 36,000 lbm and operates at 3000 rpm with dry gas seals and five pad tilting pad journal bearings (Table 1). The inlet pressure is 20 psi with a 100 psi discharge pressure and a gas mole weight of 44. Low base log dec and relatively small destabilizing forces characterize large refrigeration compressors. This machine is no exception. The base log dec (rotor and bearings only, Q = 0.0) of this compressor ranges from 0.18 to 0.13 for the range of bearing tolerances. In the past, questions have arisen concerning the applicability of Wachel type equations (Wachel and von Nimitz, 1981) for compressors with multiple process sections. The concerns have been addressed, for the most part, by applying the modified Alfords equation on a stage by stage basis. In this application, each stage represents a process section with side streams added to the main flow prior to the second and third impellers. Figure 17 presents the rotordynamic model of the compressor.
Figure 17. Propane Compressor Cross Section. Applying Equation (3) to this service, the anticipated destabilizing force, Qa, can be calculated. On a per wheel basis, the total anticipated destabilizing force is found to be Qa = 23,046 lbf/in. Applying this to the rotor center yields a log dec of 0.15 and 0.09 for minimum and maximum stiffness bearings, respectively. The stability sensitivity plot is shown on Figure 18. (The tolerance range for the bearing clearances and oil inlet temperatures defines the range of bearing stiffness.) As required by API since the worst case a<0.1, a Level II analysis was performed using the method developed by Kirk (1990) to predict the behavior of the impeller labyrinth seals and balance piston. These destabilizing forces are applied at the physical location of the seal. For the same range of bearing stiffness, the log dec is calculated for the following conditions:
Rotor and bearing only Rotor, bearing, and impeller labyrinth seals Rotor, bearing, impeller labyrinth seals, and balance piston
Table 3 contains the results of the Level II analysis and the estimation of rotor stability using the modified Alfords force. As
0.20
Level II Analysis Required a < 0.1
31
0.15
Minimum Bearing Stiffness a = 0.09 Q0 = 103,000 lbf/in
0.10
0.05
Alfords Qa = 23,046 lbf/in
Q0 = 70,000 lbf/in
0.00
-0.05
Maximum Bearing Stiffness
-0.10 0 20 40 60 80 100
As in the prior examples, the Level I analysis is compared against the Level II analysis to determine conservatism. (Note: A Level II analysis is required since Q0<2*Qa.) The modified Alfords force is calculated to be 68,015 lbf/in reflecting the high horsepower of the application (Figure 20). Table 3 contains the results of the stability analysis for the same conditions as in example 2. For applications in the midrange of pressure, the balance piston effect on rotor stability is typically equivalent to the impeller eye seals if both are labyrinth type with no antiswirl features. This can be seen for this compressor as the decrease in log dec produced by the impeller labyrinth seals is nearly equal to that produced by including the balance piston. As before, the stability level predicted using the modified Alfords force is conservative in relation to the Level II analysis performed.
1.7 1.5
Level II Analysis Required Q0 < 2*Qa
Figure 18. Propane Compressor Stability. expected, the excitation force due to the labyrinth seals (including the balance piston) is small and has a minor impact on the rotor stability. For this application, the predicted stability levels using the modified Alfords force are conservative. Table 3. Propane Refrigeration, Injection #1, and Injection #2 Compressor Stability Summary.
Predicted Log Dec, Example #2 Example #3 Example #4 Propane Refrigeration Injection #1 Injection #2 0.18 0.13 0.15 0.09 0.17 0.12 0.17 0.12 1.55 0.86 0.82 0.24 1.25 0.65 1.00 0.48 0.84 0.39 0.52 0.10 0.72 0.30 0.38 0.12
1.3 1.1
Minimum Bearing Stiffness
0.9 0.7 0.5 0.3 0.1 -0.1 0 20 40 60 80 100

a = 0.24
Q0 = 95,300 lbf/in
Configuration Rotor / Bearing (Base log dec) + Alfords Qa
Bearing Stiffness Minimum Maximum Minimum Maximum
Figure 20. Injection #1 Compressor Stability.
+ Laby Seals Minimum Maximum +Laby Seals & Balance Piston Minimum Maximum
EXAMPLE #4 11,000 HP, 4 STAGE INJECTION COMPRESSOR

The fourth example is a four stage 11,000 hp centrifugal compressor in an injection service with a 48 inch bearing span and a Lb/Dms = 7.26. The rotor weighs 422 lbm and operates at 9900 rpm with dry gas seals and five pad tilting pad journal bearings (Table 1). The inlet pressure is 5600 psi with a 9000 psi discharge pressure on natural gas. In terms of injection service, this compressor would be considered near the higher end of the discharge pressure range and in the midrange of horsepower per weight ratio. In this application, destabilizing forces are expected to be higher due to the elevated gas densities in the compressor. In fact, a Level II analysis is required due to the average gas density of 115 kg/m3. As before, the manufacturer conservatively designed the compressor with a larger central shaft section to counter the expected higher destabilizing forces (Figure 21).
EXAMPLE #3 30,000 HP, 4 STAGE INJECTION COMPRESSOR

The third example is a four stage 30,000 hp centrifugal compressor in an injection service with a 59 inch bearing span and a Lb/Dms = 8.86. The rotor weighs 540 lbm and operates at 12,700 rpm with dry gas seals and four pad tilting pad journal bearings (Table 1). The inlet pressure is 1200 psi with a 3300 psi discharge pressure on natural gas. In terms of injection service, this compressor would be considered near the lower end of the discharge pressure range. However, the high horsepower per rotor weight would place it near the top of that range. Recognizing this fact, the manufacturer conservatively designed the compressor with a larger central shaft section (Figure 19). The stiffer shaft produces bending modes with higher relative bearing motion as compared to the shaft center for the first mode. This permits the bearing damping to be more effective in controlling the shaft center and results in higher log dec values.
Figure 21. Injection #2 Compressor Cross Section. For this service, the modified Alfords force is calculated to be 72,854 lbf/in, roughly equal to the compressor in example 3 (Figure 22). Higher gas densities offset the higher horsepower of the previous example. Thus, the anticipated level of destabilizing force is roughly the same for the two examples. Table 3 contains the results of the stability analysis for Level I and Level II
Figure 19. Injection #1 Compressor Cross Section.
32
analyses. As with the other injection compressor, a Level II analysis was required since Q0<2*Qa, indicating that an insufficient safety margin exists between the anticipated destabilizing force and the amount needed to drive the system unstable.
0.9 0.8 0.7 0.6
Minimum Bearing Stiffness
To produce an equivalent destabilizing force as calculated by the modified Alfords equation, an equivalent cross-coupled stiffness is calculated for each seal using the following relation:
Qeq = k Ccs
(4)
The k and C values are determined from a labyrinth seal analysis (Kirk, 1990). The equivalent modal cross-coupling at the rotor center is defined as:
0.5 0.4 0.3 0.2 0.1 0.0 -0.1 0 20 40 60 80 100

a = 0.103
Qm =
Q0 = 100,200 lbf/in
f Qeq
(5)
Figure 22. Injection #2 Compressor Stability. As noted, the Alfords force approximation of the destabilizing force was nearly equal for the two injection compressors. Table 3 confirms this fact when comparing both Level II analyses. The change in log dec from the base value to the condition including all destabilizing forces is roughly 0.55 and 0.40 for the first injection compressor and 0.45 and 0.30 for the second injection compressor for the range of bearing coefficients. This is considered close given the difference in rotor and bearing geometry. Unlike the previous example, the balance piston is producing the majority of the destabilizing force. This was recognized in the early design stages and a shunted balance piston (Kanki, et al., 1988) was employed in the final configuration. Finally, the Alfords force is shown to be slightly conservative for one bearing condition and somewhat optimistic at the other. However, the conservatism is still seen in the worst case prediction, which would invoke a Level II analysis. Additionally, a Level II analysis was required due to the average gas density in the compressor. This compressor successfully passed a full load test without stability problems.
The modal influence factor, Mf, is determined from the normalized mode shape of the first damped natural frequency and represents the displacement at the seal location. The reduced cross-coupling forces are summed for all seal locations. This modal crosscoupling, Qm, along with the log dec value calculated in the Level II analysis is plotted on the stability sensitivity plots for the two compressor examples at the maximum bearing stiffness only. Figure 24 presents the results for the first injection compressor, example 3. The minimum modal factor for this compressor was only 0.91 reflecting the stiffer shaft operation of the compressor. Given the labyrinth seal and balance piston forces, the modal cross-coupling was calculated to be 44,933 lbf/in. From Table 3, the Level II log dec including all destabilizing forces was 0.48 at the maximum bearing stiffness. Plotting this point on the sensitivity chart, one finds that the point lies almost directly on the line derived by placing a varying amount of cross-coupling at the rotor center. It needs to be emphasized that the log dec plotted, m, was calculated from the Level II analysis with the seal effects located at the physical location of the seal. From this one can conclude the following:
The modal reduction produces a reduced cross-coupling force

directly comparable to the modified Alfords force.
The modal cross-coupling also provides an indication of how
much margin the rotor has based on the actual seal coefficients. (In this case, the 44,933 lbf/in is compared to the Q0 amount of 95,300 lbf/in. A safety margin of roughly two exists or more simply the destabilizing effect of the labyrinth seals could be two times larger than calculated before an unstable condition is predicted.)
1.7 1.5 1.3 1.1
Minimum Bearing Stiffness Maximum Bearing Stiffness
DISTRIBUTED VERSUS LUMPED ANALYSIS AND EXAMPLE #5MIXED REFRIGERATION

In the previous three examples, the conservatism of the modified Alfords equation was determined by comparing the resulting log dec values from the Level I and Level II analyses. In this section, a more direct method of comparing an equivalent lumped destabilizing force is presented similar to Memmott (2000). The example 3 injection compressor is used along with a larger refrigeration compressor. The 93,000 hp mixed refrigeration compressor (example 5) has a 222 inch bearing span and a Lb/Dms = 10.2. The two compressors are compared on Figure 23. The rotor weighs 44,730 lbm and operates at 3000 rpm with dry gas seals and five pad tilting pad journal bearings (Table 1). The inlet pressure is 60 psi with a 320 psi discharge pressure using a 25 MW gas.
0.9 0.7 0.5 0.3 0.1 -0.1 0 20 40 60 80 100

Modal Qm = 44,933 lbf/in Modal m = 0.48 Q0 = 95,300 lbf/in
Figure 24. Injection #1 Compressor Stability with Modal Q. The calculation is repeated for the mixed refrigeration compressor (example 5). For this compressor, the minimum modal factor was 0.56 representing the more flexible bending mode of the shaft. Two configurations of seals are included in the Level II analysis, one with a shunted balance piston and one without a shunt. Both modal cross-couplings and the final Level II log dec values are plotted (Figure 25). As with the injection compressor, both points
Injection & MR drawn to scale.
Figure 23. Mixed Refrigerant Compressor Cross Section.
33
lie closely to the sensitivity line from the Level I analysis. This is true even in the case of the shunted balance piston where the net cross-coupling term is negative (or stabilizing).
0.20
0.15
Shunted BP
0.10
m = 0.106 m = 0.08
Non-Shunted BP
0.05
Q0 = 69,100 lbf/in Qm = -12,403 lbf/in Qm = 23,300 lbf/in
Qeq Qm Q0 s X1,2 a m ratio d cs
= = = = = = = = = = = =
Equivalent cross-coupling (lbf/in) Modal cross-coupling (lbf/in) Aerodynamic cross-coupling for zero log dec (lbf/in) Real part of eigenvalue Amplitude (mils) Efficiency factor Log dec Log dec for Qa Final log dec from the Level II analysis Density ratio Damped natural frequency (rad/sec) Damped first natural frequency (rad/sec)
REFERENCES
API Standard 617, 2002, Axial and Centrifugal Compressors and Expander-Compressors for Petroleum, Chemical and Gas Industry Services, Seventh Edition, American Petroleum Institute, Washington, D.C.
0.00
-0.05 -50 -25 0 25 50 75 100
Figure 25. Mixed Refrigerant Compressor Stability with Modal Q.
API Technical Publication 684, 2005, Tutorial on Rotordynamics: Lateral Critical Speeds, Unbalance Response, Stability, Train Torsional and Rotor Balancing, Second Edition, American Petroleum Institute, Washington, D.C. Alford, J. S., 1965, Protecting Turbomachinery from Self-Excited Rotor Whirl, ASME Journal of Engineering for Power, 87, (4), pp. 333-344. Bansal, P. N. and Kirk, R. G., 1975, Stability and Damped Critical Speeds of Rotor-Bearing Systems, ASME Journal of Engineering for Industry, Series B, 97, (4), pp. 1325-1332. Barrett, L. E., Gunter, E. J., and Allaire, P. E., 1976, The Stability of Rotor-Bearing Systems Using Linear Transfer Functions A Manual for Computer Program ROTSTB, Report No. UVA/643092/MAE81/124, School of Engineering and Applied Science, University of Virginia, Charlottesville, Virginia. Benckert, H. and Wachter, J., 1979, Investigations on the Mass Flow and the Flow Induced Forces in Contactless Seals of Turbomachines, Proceedings of the Sixth Conference on Fluid Machinery, Scientific Society of Mechanical Engineers, Akadmiaki Kikado, Budapest, pp. 57-66. Benckert, H. and Wachter, J., 1980, Flow Induced Spring Coefficients of Labyrinth Seals for Application to Rotordynamics, NASA CP-2133, pp. 189-212. Chen, W. J., 1996, Instability Threshold and Stability Boundaries of Rotor-Bearing Systems, ASME Journal of Engineering for Gas Turbines and Power, 118, pp. 115-121. Childs, D. W. and Scharrer, J. K., 1986a, Experimental Rotordynamic Coefficient Results for Teeth-on-Rotor and Teeth-on-Stator Labyrinth Gas Seals, ASME 86-GT-12. Childs, D. W. and Scharrer, J. K., 1986b, An Iwatsubo-Based Solution for Labyrinth Seals: Comparison to Experimental Results, ASME Journal of Engineering for Gas Turbines and Power, 108, (2), pp. 325-331. Childs, D. W., and Ramsey, C., 1991, Seal RotordynamicCoefficient Test Results for a Model SSME ATD-HPFTP Turbine Interstage Seal With and Without a Swirl Brake, ASME Journal of Tribology, 113, pp. 113-203. Childs, D. W., 1993, Turbomachinery Rotordynamics: Phenomena, Modeling, and Analysis, New York, New York: John Wiley & Sons, Inc. Edney, S. L., Fox, C. H. J., and Williams, E. J., 1990, Tapered Timoshenko Finite Elements for Rotor Dynamics Analysis, Journal of Sound and Vibration, 137, (3). Elrod, D. A., Pelletti, J. M., and Childs, D. W., 1995, Theory Versus Experiment for the Rotordynamic Coefficients of an
CONCLUSIONS
The API Level I modified Alfords cross-coupling force calculation was examined and determined to be indeed a conservative estimation of the compressors destabilizing forces. Several industrial applications were examined including a hydrogen compressor, two high-pressure injection compressors, one with a high horsepower to weight ratio, and two large refrigeration compressors, including a multisection configuration. For all cases, the modified Alfords force was determined to produce the worst case stability level. Also, specifics of the Level II analysis were examined to determine what parameters are key to determining centrifugal compressor stability. Some well-known influences including bearing clearance tolerance and labyrinth seal inlet swirl were shown to have a major impact on the stability level of the 10 stage hydrogen compressor of example 1. Not so well known was the impact of the labyrinth eye clearance profile. Varying the tooth clearance slope from 8 mils diametral convergent to 8 mils diametral divergent at the impeller eye seals only, changed the predicted log dec from 0.11 to 0.23 for a constant inlet swirl of 0.6. This provides another simple tool to increase the stability of marginal centrifugal compressors. Finally, a modal approach was presented to permit direct comparison of the Level I and Level II destabilizing forces. Beyond confirming the conservatism of the modified Alfords force, the modal approach also permits use of the stability sensitivity plot to approximate the safety margin of the labyrinth seal coefficients against a zero log dec threshold.
NOMENCLATURE
Cs C Dc Dms Hc hp k Lb Mf N Nd Pdisch Pin Q Qa = = = = = = = = = = = = = = = Seal diametral clearance (mils) Principle damping (lbf-sec/in) Impeller diameter (inch) Midshaft diameter (inch) Minimum width of the impeller or discharge volute (inch) Horsepower (hp) Cross-coupled stiffness (lbf/in) Bearing span, (inch) Modal influence factor Speed (rpm) Damped natural frequency (cpm) Discharge pressure (psi) Inlet pressure (psi) Aerodynamic cross-coupling (lbf/in) Anticipated aerodynamic cross-coupling (lbf/in)
34
Interlocking Labyrinth Gas Seal, ASME Paper 95-GT-432, Presented at the International Gas Turbine and Aeroengine Congress and Exposition, Houston, Texas. Fowlie, D. W. and Miles, D. D., 1975, Vibration Problems with High Pressure Centrifugal Compressors, ASME 75-Pet-28. Geary, C. H., Damratowski, L. P., and Seyer, C., 1976, Design and Operation of the Worlds Highest Pressure Gas Injection Centrifugal Compressor, Paper Number OTC 2485, Presented at the Eighth Annual Offshore Technology Conference, Houston, Texas. Gunter, E. J., 1966, Dynamic Stability of Rotor-Bearing Systems, NASA SP-113. Gunter, E. J., 1967, The Influence of Internal Friction on the Stability of High Speed Rotors, ASME Journal of Engineering for Industry, Series B, 89, (4), pp. 683-688. Iwatsubo, T., Matooka, N., and Kawai, R., 1982, Spring and Damping Coefficients of the Labyrinth Seal, NASA CP-2250, pp. 205-222. Kanki, H., Katayama, K., Morii, S., Mouri, Y., Umemura, S., Ozawa, U., and Oda, T., 1988, High Stability Design for New Centrifugal Compressor, Rotordynamic Instability Problems in High-Performance Turbomachinery, NASA CP-3026, pp. 445-459. Kirk, R. G., 1980, Stability and Damped Critical Speeds: How to Calculate and Interpret the Results, CAGI Technical Digest, 12, (2). Kirk, R. G., 1988a, Evaluation of Aerodynamic Instability Mechanisms for Centrifugal CompressorsPart I: Current Theory, ASME Journal of Vibration, Acoustics, Stress, and Reliability in Design, 110, (2), pp. 201-206. Kirk, R. G., 1988b, Evaluation of Aerodynamic Instability Mechanisms for Centrifugal CompressorsPart II: Advanced Analysis, ASME Journal of Vibration, Acoustics, Stress, and Reliability in Design, 110, (2), pp. 207-212. Kirk, R. G., 1990, A Method for Calculating Labyrinth Seal Inlet Swirl Velocity, ASME Journal of Vibration and Acoustics, 112, (3), pp. 380-383. Kwanka, K., Ortinger, W., and Steckel, J., 1993, Calculation and Measurement of the Influence of Flow Parameters on Rotordynamic Coefficients in Labyrinth Seals, Rotordynamic Instability Problems in High-Performance Turbomachinery, NASA CP-3239, pp. 209-218. Lund, J. W., 1965, Rotor Bearing Dynamics Design Technology, Part V, AFAPL-TR-65-45, Aero Propulsion Laboratory, Wright-Patterson Air Force Base, Dayton, Ohio. Lund, J. W., 1974, Stability and Damped Critical Speeds of a Flexible Rotor in Fluid Film Bearings, ASME Journal of Engineering for Industry, 96, (2), pp. 509-517. Memmott, E. A., 2000, Empirical Estimation of a Load Related Cross-Coupled Stiffness and the Lateral Stability of Centrifugal Compressors, Presented at the 18th Machinery Dynamics Seminar, Canadian Machinery Vibration Association, Halifax, Nova Scotia, Canada.
Moore, J. J. and Hill, D. L., 2000, Design of Swirl Brakes for High Pressure Centrifugal Compressors Using CFD Techniques, Proceedings of the Eighth International Symposium of Transport Phenomena and Dynamics of Rotating Machinery, (ISROMAC-8), Honolulu, Hawaii, pp. 1124-1132. Moore, J. J. and Soulas, T., 2003, Damper Seal Comparison in a High-Pressure Re-Injection Centrifugal Compressor During Full Load, Full-Pressure Factory Testing Using Direct Rotordynamic Stability Measurement, Proceedings of DETC 03, ASME Design Engineering Technical Conferences and Computers and Information in Engineering Conference, Chicago, Illinois. Murphy, B. T. and Vance, J. M., 1983, An Improved Method for Calculating Critical Speeds and Rotordynamic Stability of Turbomachinery, ASME Journal of Engineering for Power, 105, (3), pp. 591-595. Nelson, H. D. and McVaugh, J. M., 1975, The Dynamics of Rotor-Bearing Systems Using Finite Elements, ASME Journal of Engineering for Industry, 98, (2), pp. 593-600. Ramesh, K. and Kirk, R. G., 1993, Stability and Response of Rotors Supported on Active Magnetic Bearings, Proceedings of the 14th ASME Vibrations and Noise Conference, DE- 60, Vibration of Rotating Systems, pp. 289-296. Rouch, K. E. and Kao, J. S., 1979, A Tapered Beam Finite Element for Rotor Dynamics Analysis, ASME Journal of Sound and Vibration, 66, pp. 119-140. Ruhl, R. L. and Booker, J. F., 1972, A Finite Element Model for Distributed Parameter Turborotor Systems, ASME Journal of Engineering for Industry, Series B, 94, (1), pp. 126-132. Scharrer, J. K. and Pelletti J. M., 1994, Commercial Applications of Space Propulsion Turbomachinery Component Technology, SAE Paper Number 941197, Presented at the SAE International Aerospace Atlantic Conference and Exposition, Dayton, Ohio. Smith, K. J., 1975, An Operational History of Fractional Frequency Whirl, Proceedings of the Fourth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 115-125. Thieleke, G. and Stetter, H., 1990, Experimental Investigations of Exciting Forces Caused by Flow in Labyrinth Seals, Rotordynamic Instability Problems in High-Performance Turbomachinery, NASA CP-3122, pp. 109-134. Wagner, N. G. and Steff, K., 1996, Dynamic Labyrinth Coefficients from a High-Pressure Full-Scale Test Rig Using Magnetic Bearings, Rotordynamic Instability Problems in High-Performance Turbomachinery, NASA CP-3344, pp. 95111. Wachel, J. C. and von Nimitz, W. W., 1981, Ensuring the Reliability of Offshore Gas Compression Systems, Journal of Petroleum Technology, pp. 2252-2260.
REMAINING LIFE ASSESSMENT OF STEAM TURBINE AND HOT GAS EXPANDER COMPONENTS
by Phillip Dowson
General Manager, Materials Engineering
Wenchao Wang
Senior Materials Engineer Elliott Company Jeannette, Pennsylvania
and Agustin Alija

Maintenance Engineer Dow Chemical PBBPolisur Bahia Blanca, Argentina
ABSTRACT
Phillip Dowson is General Manager, Materials Engineering, with Elliott Company, in Jeannette, Pennsylvania, and has 34 years of experience in the turbomachinery industry. He is responsible for the metallurgical and welding engineering for the various Elliott product lines within the company. He is the author/coauthor of a number of technical articles related to topics such as abradable seals, high temperature corrosion, fracture mechanics, and welding/brazing of impellers. Mr. Dowson graduated from Newcastle Polytechnic in Metallurgy and did his postgraduate work (M.S. degree) in Welding Engineering. He is a member of ASM, ASTM, and TWI. Wenchao Wang is a Senior Materials Engineer with Elliott Company, in Jeannette, Pennsylvania. He joined the company in 1997, and has been involved in materials related R&D projects, production supports, and aftermarket services. He is experienced in turbomachinery failure analysis, protective/functional coatings, and remaining life assessment. Dr. Wang received a B.S. degree (Metallurgical Engineering, 1984) from the University of Science & Technology Beijing, and M.S. (1994) and Ph.D. (1997) degrees (both in Materials Science & Engineering) from the University of Cincinnati. He is a member of ASM. Agustin Alija is a Maintenance Engineer of the Hydrocarbon and Energy Business of PBBPolisur (a Dow Chemical owned company), in Bahia Blanca, Argentina. He joined the company in June 2003. Mr. Alija has a B.S. degree (Mechanical and Aeronautical Engineering, 2000) from the National University of Cordoba in Argentina. In todays marketplace, a large percentage of oil refinery, petrochemical, and power generation plants throughout the world have been trying to reduce their operation cost by extending the service life of their critical machines, such as steam turbines and hot gas expanders, beyond the design life criteria. The key ingredient in plant life extension is remaining life assessment technology. This paper will outline the remaining life assessment procedures, and review the various damage mechanisms such as creep, fatigue, creep-fatigue, and various embrittlement mechanisms that can occur in these machines. Also highlighted will be the various testing methods for determining remaining life or life extension of components such as high precision stress relaxation (STR) test, which determines creep strength, and constant displacement rate (CDR) test, which evaluates fracture resistance. Other tests such as replication/microstructure analysis and toughness tests will also be reviewed for calculating the remaining life or life extension of the components. Use of the latest computer software will also be highlighted showing how creep-life, fatigue-life, and creep/fatigue-life calculations can be performed. Also shown will be actual life extension examples of steam turbines and hot gas expander components performed in the field.
INTRODUCTION
In recent years, from oil refinery to petrochemical and power generation industries, more and more plants throughout the world are facing a common issueaging turbines, usually over 30 years old. Questions bearing in managers minds are what is the machine condition and whether they can be continually operated (if yes, how long). The answer is significant not only for safety concerns but also for cost reduction, especially with todays limited budgets. Therefore, there is an increasingly strong desire for the engineering aftermarket service to perform remaining life assessment of steam turbines and hot gas expanders. Remaining life assessment is to use metallurgical and fracture mechanics methodologies to predict the remaining life of structures and components that have been in service for an extended period of time, usually close to or beyond the designed life. Traditionally, if parts are found with material degradations or damages during an overhaul, they might be scrapped and replaced for risk-free consideration; even though they might have some useful life. Remaining life assessment offers a possible tool to estimate the useful remaining lifetime and avoid premature scrapping of the parts. So remaining life assessment is considered
77
78
to be an attractive method/process for cost reduction and reduction downtime. Remaining life assessment has often been improperly referred to as life extension. Actually this analysis cannot extend the lifetime of the components. It can only assess the useful remaining lifetime, based on the metallurgical examinations and theoretical (fracture mechanics) calculations. If such assessments indicate the need for extensive replacements and refurbishments, life extension may not prove to be a viable option. Above and beyond this objective, remaining life assessment technology serves many other purposes. It helps in setting up proper inspection schedules, maintenance procedures, and operating procedures. It should, therefore, be recognized at the outset that development of techniques for remaining life assessment is more enduring in value and broader in purpose than simply the extension of plant life. For instance, it has been possible to extend the inspections from six to 10 years for modern rotors, on the basis of assessments based on fracture mechanics, resulting in considerable savings. In implementing remaining life assessment procedures, the appropriate failure definition applicable to a given situation must be determined at the outset, and the purpose for which the assessment is being carried out must be kept in mind. While determining the feasibility of extended plant life may be one objective, a more common objective is the setting of appropriate intervals for inspection, repair, and maintenance. In this context, remaining life assessment procedures are used only to ascertain that failures will not occur between such intervals. It should never be assumed that having performed a remaining life assessment study for a 20-year life extension, one could then wait for 20 years without interim monitoring. Periodic checks to ensure the validity of the initial approach are essential. In this sense, remaining life assessment should be viewed as an ongoing task, rather than a one-time activity. A phased approach, in which the initial level includes nonincursive techniques followed by other levels of actual plant monitoring, then followed by nondestructive inspections and destructive tests would be the most logical and cost-effective approach. In Level I, assessments are performed using plant records, design stresses and temperatures, and minimum values of material properties from the original equipment manufacturer (OEM). Level II involves actual measurements of dimensions, temperatures, simplified stress calculations, and inspections coupled with the use of minimum material properties from the OEM. Level III involves indepth inspection, stress levels, plant monitoring, and generation of actual material data from samples removed from the component (destructive testing). The degree of the detail and accuracy of the results increases from Level I to Level III, but at the same time, the cost of the assessment also increases. Depending on the extent of the information available and the results obtained, the analysis may stop at any level or proceed to the next level as necessary. In evaluating the failure criteria or remaining life, one needs to understand the various failure mechanisms that can occur. In turbomachinery components, the failure criteria can be governed by one or a combination of the following failure mechanisms:
However, in remaining life assessment, usually only those mechanisms depending on temperature and time are taken into account. For example, for turbine casing, engineers usually focus on thermal stress-induced low cycle fatigue, creep rupture, and tempering embrittlement cracking. These failures usually are slow processes, therefore, they can be assessed and forecasted by examining the warning evidences in the material. Countless works have been done to study the behaviors of fatigue crack initiation/propagation and creep or embrittlement rupture in steels and alloys. Scientists and engineers have reached such a level that, by knowing the flaw size or microstructure deterioration/damage, one can theoretically calculate and predict the remaining lifetime of the parts, based on the knowledge of the material properties and understanding of the stress distributions.
FATIGUE
Failures that occur under cyclic loading are termed fatigue failures. These can be vibration stresses on blades, alternating bending loads on shafts, fluctuating thermal stresses during startstop cycles, etc. There are two types of fatigue: low cycle fatigue (LCF), high cycle fatigue (HCF). Traditionally, low cycle fatigue failure is classified occurring below 104 cycles, and high cycle fatigue is above that number. An important distinction between HCF and LCF is that in HCF most of the fatigue life is spent in crack initiation, whereas in LCF most of the life is spent in crack propagation because cracks are found to initiate within three to 10 percent of the fatigue life. HCF is usually associated with lower stress, while LCF usually occurs under higher stress. Remaining life of casings or rotating components is generally based upon crack growth consideration. Fracture mechanics is the mathematical tool that is employed. It provides the concepts and equations used to determine how cracks grow and their effect on the strength of the structure. At the authors company fracture mechanics is utilized in analyzing the structural integrity of components that have been in operation to determine whether the component is suitable for further operation. Based upon crack growth analysis one considers a number of scenarios. From an initial defect size ao one must determine critical flaw size ac for fast fracture.
ao ac KI KIC
where: KI = Applied tensile mode I stress intensity factor KIC = Plane strain fracture toughness of material
(1)
KI = Stress intensity factor range ksi inch Kth = Threshold stress intensity factor range below which fatigue crack growth (or corrosion fatigue crack growth) does not occur
For LCF, determine how many cycles for ao to grow to ac. For HCF, one must prevent crack growth. Consequently for HCF, KI < Kth, where:
Fatiguehigh cycle or low cycle Corrosion/corrosion fatigue Stress corrosion cracking (SCC) Erosionsolid particle or liquid impingement Erosion corrosion Embrittlement Creep rupture/creep fatigue High temperature corrosion/embrittlement Mechanical (foreign objective) damage
Further discussion of fracture mechanic concepts can be found elsewhere (Dowson, 1995, 1994).
CREEP RUPTURE AND STRESS RUPTURE

Evidence of creep damage in the high temperature regions of blade attachment areas of rotors has been observed in some instances (Bush, 1982). The rim stresses and metal temperature at these locations are assessed against the creep rupture data for that particular grade of steel/or material. Traditionally one has used a Larson-Miller (LM) plot of the type shown in Figure 1. The degree of safety margin depends on the user and what lower bound design curve is applied. Since these curves are based upon the chemistry, variation in chemistry for a particular grade can have an effect on the Larson-Miller curve. Also, Larson-Miller
100
79
Life fraction expended, ti/tri, at 1000F 42,500 = = 0.19 220,000 Life fraction expended, ti/tri, at 1025F 42,500 = = 0.516 82,380
Stress [ksi]
The total life fraction expended is 0.71.

10
This rule was found to work well for small changes in stress and temperature especially for CrMoV rotor steel. However, for stress variations, the actual rupture lives were lower than the predicted values. Consequently, the LFR is generally valid for variable-temperature conditions as long as changing creep mechanisms and environmental interaction do not interfere with test results. However, the possible effect of material ductility on the applicability of the LFR needs to be investigated. Nondestructive Techniques
1 25 27 29 31 33 35 37 39 41 43 45 (T+460)(logt+20)x10^-3
Figure 1. Larson-Miller Curve of Cr-Mo-V Alloy Steel (ASTM A470 Class 8). curves are generally based upon creep rupture tests done for 104 to 3 104 hours and very few data at 105 hours. Consequently, the data for longer hours are generally extrapolated. Since most of the creep rupture data is done with smooth bar specimens, the effect of notch ductility at long-term service has not been done. Short-term notched bar tests may fail to predict the onset of notch sensitivity. Notch sensitively is not an inherent property but depends on the temperature, stress, stress state, and strain rate. In assessing remaining life of the components due to creep, such as blade attachments, crack initiation is used as the criterion. However, with the emergence of cleaner steel and fracture mechanics and an increasing need to extend the life of a component, application of crack growth techniques have become common in the past decade. For crack initiation as the fracture criterion, history-based calculation methods are often used to estimate life. Methods for Crack Initiation Due to Creep For the analytical method, one must have accurate operating history of the components, which may consist of temperature, applied loads, changes in operation, such as shut downs or variation in speed or pressure. A simplistic estimation of the creep life expended can be made by assessing the relaxed long-term bore stresses and rim stresses against the standard rupture data using the life fraction rule. The life fraction rule (LFR) states that at failure:
Conventional nondestructive evaluation (NDE) techniques fail to detect incipient damage that can be a precursor to crack initiation and subsequent rapid failure. However, there are other NDE techniques that have been developed for estimating the life consumption. These include microstructural techniques and hardness based techniques. Metallographic Examination Metallographic techniques have been developed that can correlate changes in the microstructure and the onset of incipient creep damage, such as triple point cavitation at the grain boundaries. For this technique, measurements by replication technique are taken on crack sensitive areas that are subjected to the higher temperatures and stresses. These areas are generally indicated by experience and analysis of previous damages. The creep damage measured by replication is classified into four damage stages:
Isolated cavities (A) Oriented cavities (B) Macrocracks (linking of cavities) (C) Formation of macrocracks (D)
Figure 2 shows the location of the four stages on the creep strain/exposure time curve (Neubauer and Wadel, 1983).
t
Example:
ti
ri
=1
(2)
where ti is the time spent at a given stress and temperature and tri is the rupture life for the same test conditions. The purpose of this example is to illustrate the use of the lifefraction rule. A steam turbine piping system, made of 1.25Cr-0.5Mo steel designed for a hoop stress of 7 ksi, was operated at 1000F (538C) for 42,500 hours and at 1025F (552C) for the next 42,500 hours. Calculate the life fraction expended using the life fraction rule. From the Larson-Miller parameter curve of the steel, it is found that, at = 7 ksi, tri at 1000F = 220,000 hours tri at 1025F = 82,380 hours
Figure 2. Replicas for Remaining Life Assessment. In applying this approach Neubauer and Wadel (1983) classified the stages into five stages, which are Undamaged, Stage A, Stage B, Stage C, and Stage D. These stages corresponded roughly to expended life fracture (t/tr) values of 0.27, 0.46, 0.65, 0.84, and 1, respectively, using the conservative lower bound curve.
80
Consequently, the remaining life can be calculated using the relationship as shown in Equation (3):
t trem = t r 1 t
where t is the service life expended and tr is the rupture life.
(3)
For undamaged material and damaged stages A, B, and C, the remaining life was found to be approximately 2.7t, 1.17t, 0.54t, and 0.19t, respectively. Then by applying a safety factor of 3 to the calculations, the safe reinspection intervals will become 0.9t, 0.4t, 0.18t, and 0.06t, respectively. This approach has been developed and implemented in the power generation industry (Viswanathan and Gehl, 1991). It was found to give increased inspection intervals as compared to the Neubauer and Wadel (1983) approach, as shown in Table 1. Table 1. Suggested Reinspection Intervals for a Plant with 30 Years of Prior Service.
Inspection Interval (Years) Damage Classification Wedel-Neubauer EPRI-APTECH Undamaged 5 27 A. Isolated Cavities 3 12 B. Oriented Cavities 1.5 5.4 C. Linked Cavities (Microcracks) 0.5 1.8 D. Macrocracks Repair Immediately Based on fracture mechanics
Figure 3. Correlation Between Post-Exposure Rupture Time in the Standard Test at 538C (1000F) and 240 MPa and Room Temperature Hardness for Cr-Mo-V Rotor Steel.
This approach has been applied by several utilities and realized significant savings in inspection costs. Other investigations indicate that there are wide variations in behavior due to differences in grain size, ductility, and impurity control (Carlton, et al., 1967). For conservatism, the authors company adapted the Neubauer and Wadel (1983) approach and classified the five stages as follows: 1. Undamaged materialEquipment can run and be reinspected at next shutdown. 2. Class AReinspection would be three to five years. 3. Class BReinspection would be one and one-half to three years. 4. Class CReplacement or repair would be needed within six months. 5. Class DImmediate replacement or repair would be required. Hardness Measurement The first attempt to develop hardness as a technique to determine creep damage was by Goldhoff and Woodford (1972). In their study a good correlation was observed between room temperature hardness measured on exposed creep specimens and the post exposure rupture life (Figure 3). If similar calibration could be established between prior creep life expended or the remaining life fraction in the post exposure test and the hardness values for a range of CrMoV steels, this method could be applied to estimation of remaining life. However, data of this nature are not available in sufficient quantity. Other work done by Viswanathan and Gehl (1992) showed a lot of promise where they attempted to use the hardness technique as a stress indication. They observed that the application of stress accelerated the softening process and shifted the hardness to lower parameter values compared with the case of simple thermal softening on a plot of hardness versus a modified Larson-Miller parameter (Figure 4). Destructive Techniques Newer tests to ascertain the useful life of used and/or repaired components have been utilized by the authors company. Design-forperformance is a recently developed methodology for evaluating the creep strength and fracture resistance of high temperature materials. Whereas the traditional approach to creep design involves
Figure 4. Plot of Hardness Ratio Versus G Parameter for LongTerm Heating and Creep of Cr-Mo-V Rotor Steel. long-term testing and attempts to incorporate microstructural evolution in the test measurements, the new approach aims to exclude these changes in a short time high-precision test. The test may also be used to evaluate consequences of such changes in service-exposed samples. The new methodology recognizes that separate tests are necessary to measure creep strength and fracture resistance. For creep strength, a stress versus creep rate response is determined from a stress relaxation test (SRT), and for fracture resistance a constant displacement rate (CDR) test of a notched temple specimen is performed at a temperature where the part is most vulnerable to fracture (Woodford, 1993). Constant Displacement Rate Test A description of the standardized CDR test is found elsewhere (Pope and Genyen, 1989). The data from the CDR test are tabulated in a curve similar to the load displacement curve for an
81
ordinary elevated tensile test. For a typical tensile test fracture becomes unstable after the peak load is reached. On the other hand, in the CDR test, since the deformation is controlled at a constant rate and the notch is midway between the controlling extensometer, fracture rarely becomes unstable. For a valid CDR test, the criterion for failure was considered to be the value of displacement at fracture, defined as the point of intersection of the 100 pound load line and the descending load displacement curve. The displacement at failure is measured from the start of the test to the point where the load displacement curve decreases below 100 pounds (Figure 5).
. Where e is the elastic strain, e is the elastic . strain rate, I is the inelastic strain (principally creep strain), I is the inelastic strain rate, t is the total strain, is the stress, and E is the elastic modulus measured during loading. Using this procedure, stress versus strain-rate curves were generated covering up to five orders of magnitude in strain-rate in a test lasting less than 5 hours. An example of the data generated in such tests is provided in Figure 7 for Waspaloy material. This shows stress versus predicted time to 1 percent creep for Waspaloy. By utilizing these data one can plot a stress versus Larson-Miller parameter for 1 percent predicted creep of Waspaloy compared to rupture data (Figure 8). From the data shown on the curve, the stress relaxation test can generate creepstress rupture data in less than a few weeks as compared to the traditional approach that incorporates long time testing.
Figure 5. Example of Load Displacement Curve from CDR Tests at 1200F and 2 mils/in/hr. An example of how the environment can affect the notch sensitivity of the material is indicated by Figure 6. This example illustrates the effect of air exposure on IN738. Figure 7. Stress Versus Predicted Times to 1 Percent Creep for Standard Waspaloy.
Figure 6. Constant Displacement Rate Tests Comparing Crack Growth Resistance in Heat Treated and Oxygen Embrittled Specimens. Stress Relaxation Test Specially designed samples were tested on an electromechanical test system fitted with self-aligning grips, a 1500C (2732F) short furnace, and a capacitive extensometer. Details of the specimen geometry and extensometer sensitivity are provided elsewhere (Woodford, et al., 1992). The standard test procedure involved loading the specimen at a fast rate of 10 MPa/sec (1450 psi/sec) to a prescribed stress and then switching to strain control on the specimen and monitoring the relaxation stress. The inelastic (principally creep) strain-rate is calculated from the following equations, Equations (4) and (5).
Figure 8. Stress Versus Larson-Miller Parameter for 1 Percent Predicted Creep of Standard Waspaloy Compared with Rupture Data. One major objective to this framework has been that effects of very long time exposures that could influence stress rupture life will not be accounted for. However, Woodford believes that such effects, i.e., precipitation of embrittling phases and grain boundary segregation of harmful elements, are expected to influence the fracture resistance rather than creep resistance. The authors company has utilized this methodology to generate data for high temperature materials and weldments. Current methods are being developed for miniature specimens taken from serviced blades. From these data, it is envisioned that establishment of a set of minimum performance criteria will enable repair/rejuvenation/ replacement decisions to be made.
e + I = t = Cons tan t
I = e = 1 d dt
(4)
CREEP/FATIGUE INTERACTION
For components that operate at higher temperature where creep growth can occur, one must take into account the creep crack growth at intervals during the fatigue life of the component. The following is an example of a high temperature steam turbine rotor
(5)
82
that failed catastrophically at a power plant in Tennessee (Saxena, 1998). The authors utilized the latest computer software to demonstrate how creep-life, fatigue life, and creep/fatigue-life calculations can be performed, and how inaccurate the calculation would be without accounting for the creep life. The power plant rotor was operated for 106,000 hours and had incurred 105 cold starts and 183 hot starts. The material was 1CR1Mo-0.25V forging and had been operating at a temperature of 800F. The cracks originated from several majority node set (MnS) clusters with the original flaw size of 0.254 inch 5.51 inch and 0.7 inch from the bore of the rotor (Figure 9).
Figure 11. Computer Software ModelElliptical Subsurface Cracked Plate under Membrane and Bending Stresses. By computing the information a plot of stress intensity factor versus crack depth was done. Based on the plane strain fracture toughness of the material, the critical crack size 0.42 inch for cold start and 0.48 inch for hot start were determined (Figure 12).
Figure 9. Schematic of the Intermediate Pressure (IP) Section of the Rotary Showing the Size and Location of the Primary and Secondary Flaws Beneath the Seventh Row (7-R) of Blades. Step 1Assessment of Low Cycle Fatigue Life The principle is that fatigue crack growth follows equations such as the Paris law:
da n = C f K f dN
(6) Figure 12. Critical Crack Size CalculationThe Critical Crack Size ac is 0.42 Inch for Cold Start and 0.48 Inch for Hot Start. Fatigue crack growth using the Paris law was computerized (Figure 13, Figure 14) and the low cycle fatigue crack growth was determined for both cold and hot starts. Figure 15 shows the low cycle fatigue crack growth, which does not compare very well with that of the real life.
Cf and nf are constants that depend on the material and environment. The stress intensity factor range K depends on the stress level at the crack tip. The life assessment criterion is that critical crack size ac is not to be exceeded. In other words,
a ac
(7)
Figure 10 and Figure 11 show the stress intensity factor calculation together with the crack model that was used.
Figure 13. Fatigue Crack Growth. Figure 10. Computer Software Module. Stress Intensity Factor Calculation. The reason for the calculated life being much longer than the real life is that the hold time effect (or creep cracking effect) is not
83
Figure 16. Creep/Fatigue Model.
Figure 14. Fatigue Crack Growth Calculation Model.

0.22 Crack Depth [inch] 0.2 0.18 0.16 0.14 0.12 0 500 1000 1500 2000 2500 Cycles Cold Start Hot Start Real Life
Figure 15. Shows Low Cycle Fatigue Crack Growth for Cold and Hot Starts Compared to Real Life. taken into account. Consequently, one must run a creep-fatigue remaining life assessment. The principle is that high temperature crack propagation is the summation of high temperature fatigue plus primary creep plus secondary creep. Creep-fatigue crack growth: Figure 17. Creep/Fatigue Material Data.
Cf da q = Cc Ct (t ) + ( K )n f
h dt

(8)
Crack Depth [inch]
0.22 0.2 0.18 0.16 0.14 0.12 0 500 1000 1500 Cycles 2000 2500
Creep
Fatigue
where h is the hold time in each cycle (which is 368 hours in this case). The creep crack driving force consists of two parts:
Ct (t )= C*
[ ]
n + p +1 * 1 + + Ch C* , ( n + 1)t n + 1)( p + 1) t ( Secondary Creep

Pr imaryCreep
2 / (1+ p ) (m1)
1 ] (
)K
1 2 / ( 1+ p ) (m 1)
p / ( p +1)
(9)
The data that are input into the code are shown in Figure 16 and Figure 17. The calculation shows that, accounting for creep effect, the creep-fatigue growth is much closer to the real life (Figure 18). This example demonstrates that if the material/component is operating in the creep mode, one must perform a creep-fatigue analysis instead of fatigue only. Generally a rule of thumb is that if only low cycle fatigue crack growth is counted and creep is not, then the calculated lifetime is about 10 times longer than the real life. By utilizing this software program, more accurate remaining life assessment can be achieved for materials operating in the creep regime under cyclic loading. Also, the lesson being learned is that this rotor should have been examined by ultrasonic inspection every five years.
Low Cycle Fatigue Real Life Creep-Fatigue
Figure 18. Compares Low Cycle Fatigue and Creep-Fatigue Crack Growth to the Real Crack Growth.
EMBRITTLEMENT
Trends toward increasing size and operating stresses in components, such as large turbine-generator rotors, require higher hardenability steels with increased strength and fracture toughness. However, higher hardenability steels especially those containing
84
nickel and chromium are usually much more susceptible to a phenomenon called temper embrittlement. The term temper embrittlement refers to a shift in the brittle-to-ductile transition temperature when steels/rotor shafts are heated or cooled slowly through the temperature range of 660F to 1060F. This shift in the brittle-to-ductile transition temperature can be reversed by heating at a temperature of 1100F or higher then fast cooled. Consequently, when examining rotor or casings that have been in service and operated within this temper embrittlement temperature range, the property toughness becomes an important criterion. Evaluation of Toughness Due to the advancement of fracture mechanics, it has now become possible to characterize toughness in terms of critical flaw size ac. The definition of ac depends on the conditions under which final rapid fracture occurs following the initial phase of subcritical crack growth. At rotor grooves and rotor bores where final fracture is likely to occur at low temperatures during start-stop transients, ac is dictated by linear elastic fracture mechanics. ac is given by an expression of the form:
periphery of a rotor and converting the FATT into KIC using a correlation such as the one shown in Figure 20 (Schwant and Timo, 1985).
KIC = M ac
(10)
where KIC is the fracture toughness of the material or critical stress intensity for fracture, M is a constant to a given flaw size and geometry, and is the nominal applied stress. A typical loading sequence for turbine rotors shows variation in temperature and stress and their effect on critical flaw size ac for cold start sequence (Viswanathan and Jaffee, 1983). Figure 19 shows the cold start sequence and associated variations in stress (), temperature (T), and critical flaw size (ac) as a function of time for the power plant rotor, which failed catastrophically.
Figure 20. Turbine Rotor and Wheel Toughness Data. Viswanathan and Gehl (1991) evaluated a number of data from numerous exposed CrMoV rotors and defined a lower threshold band for low alloy steels. The lower limit line for CrMoV steel is defined by the equation (Viswanathan and Wells, 1995) below:
2 KIC = 95.042 + 0.5872T + 0.00168T + 0.00000163
(11)
Figure 19. Illustration of Cold-Start Sequence and Associated Variations in Stress (), Temperature (T), and Critical Flaw Size (ac) as Functions of Time from Start. Region A consists of a warmup period after which the rotor was gradually brought up to speed (Region B). Once continuous operating speed was reached (Region C) approximately 3 hours after the warm up period, maximum loading was applied. Analysis of the failure location (seventh row) showed that stresses reached a peak value of 74 ksi (520 MPa), 11/2 hours after the maximum continuous speed had been attained. The critical crack size reached its lowest value of 0.27 inch (6.9 mm) at a temperature of 270F (132C) and 74 ksi (510 MPa). Since variation in temperature, stress, and material fracture toughness at the defect location can dictate the ac value for the rotor, one must calculate for the worst combination of these variables to prevent failure. This can be done by using lower scatter band values of KIC. However, to determine KIC for rotors, large specimens need to be taken to satisfy the plane strain conditions required for a valid test. A more common practice is to determine the ductile-brittle fracture at transition temperature (FATT) using samples extracted from the
where KIC is expressed in ksi-(in) 1/2 and TE is the excess temperature (T-FATT) expressed in degrees Fahrenheit. Once the FATT is known, a KIC versus temperature T curve can be established and used to determine ac versus T. However, such procedures tended to be conservative and there have been various other nondestructive and relatively destructive tests involving removal of very small samples to determine toughness of rotors. These techniques investigated include eddycurrent examination, analytical electron microscope, secondary ion mass spectroscopy (SIMS), compositional correlations, Auger electron spectroscopy, chemical etching, use of single Charpy specimen, and small punch tests. The techniques that show the most promise and have currently been applied in service application are:
Correlation based on composition. Small punch testing. Chemical etching.

Correlation Based on Composition An American Society for Testing and Materials (ASTM) special task force on large turbine generator rotors of Subcommittee VI of ASTM Committee A-1 on steel has conducted a systematic study of the isothermal embrittlement at 750F of vacuum carbon deoxidized (VCD) NiCrMoV rotor steels. Elements, such as P, Sn, As, Sb, and Mo were varied in a controlled fashion and the shifts in FATT, () FATT were measured after 10,000 hours of exposure. From the results, the following correlations were observed in Equation (12):
FATT = 13544 P + 12950Sn + 2100 As 93 Mo 810,000( P Sn) (12)
where FATT is expressed in degrees Fahrenheit and the correlation of all the elements are expressed in weight percent. According
85
to this correlation, the elements P, Sn, and As increase temper embrittlement of steels, while Mo, P, and Sn interaction decrease the temper embrittlement susceptibility. All available 10,000 hour embrittlement data are plotted in Figure 21 as a function of calculated FATT using Equation (11), (Newhouse, et al., 1972). A good correlation is observed between calculated and experimental FATT. The scatter for these data is approximately 30F for 750F exposure and 15F for the 650F exposure.
Figure 22. Correlation Between Compositional Factor J and the Shift in FATT of NiCrMoV Steels Following Exposure at 650F and 750F for 8800 Hours.
Figure 21. Correlation Between Compositional Parameter N and the Shift in FATT of NiCrMoV Steels Following Exposure at 650F and 750F for 8800 Hours. Other correlations for determining the temper embrittlement susceptibility of steel, such_ as the J factor proposed by Watanabe and Murakami (1981) and X factor proposed by Bruscato (1970), are widely used. These factors are given by:
J = (Si + Mn)( P + Sn)104
(13) _ Figure 23. Relationship Between Increase of FATT and X. (14)
X = (10 P + 5Sb + 4 Sn + As)102
The Figure 22 and 23 _ show the relationship between increase of FATT and J factor and X factor at 399C (750.2F) for a 3.5 percent NiCrMoV steel. Small Punch Testing Small punch testing of small disk-like specimens subjected to bending loads have found good correlation for determining the ductile-brittle transition temperature (Baik, et al., 1983). The procedure consists of thin plate 0.4 0.4 0.02 inch subjected to a punch deformation with a 0.09 inch diameter steel ball in a specially designed specimen holder. The test is performed at various temperatures and from the load deflector curves obtained at various temperatures, the fracture energy is calculated. The fracture energy is plotted as a function of test temperature to determine the ductile-to-brittle transition temperature. The area under the deflector curve denotes the energy absorbed during the test. This test procedure has been used successfully on a number of retired rotor samples to determine the TSP (ductile-to-brittle transition) and found to correlate well with the Charpy FATT values (Foulds, et al., 1991) (Figure 24).
Figure 24. Correlation Between Charpy FATT and Small Punch Transition Temperature.
86
Chemical Etching The grain boundaries of embrittled steel are attacked preferentially by picric acid solutions consisting of saturated picric acid solution with an addition of 1 gram of tridecyl/benzene sulfonate (per 100 ml of aqueous picric acid). Correlations have been made between the grain boundary groove depth as measured metallographically even from plastic replicas and the FATT of the sample due to prior temper embrittlement. By using plastic replicas this technique becomes very attractive for field use (Kadoya, et al., 1991). Also Kadoya, et al., (1991) were able to correlate the width of the grain boundary groove measured from plastic replicas using a scanning electronic microscopes (SEM), with the FATT of the samples and actual rotors. They also correlated the following equation based on regression analysis of a number of material variables. This equation was found to predict the FATT within a scatter of 20C.
FATT = 99.12W + 1609 . HV + 816.4 Si + 6520 Mn + 3320 P + 310.4Cr + 3404 Sn + 0.282 J + 325.6
where: J is defined as (Si + Mn) (P + Sn) 104 W = Width of grain boundary groove HV = Vicker hardness of sample/rotor
flue gas composition, regenerated catalyst, and the nature and quality of additions injected into the FCC process. When evaluating remaining life assessment of hot gas expanders especially the rotating components, one must consider the effect of the environment such as high temperature corrosion. The authors company has developed a fracture mechanics model that incorporated both the effect of oxide wedge formation and the apparent reduction in fracture toughness of Waspaloy in contact with the catalyst residue. By utilizing this model one can predict whether fracture will occur under various environmental/operating conditions of the hot gas expander. The authors company periodically tests catalysts from end-users hot gas expanders to determine if oxide wedge can occur and what life span to reach the critical size for failure. Generally, if the catalyst is active, then high temperature corrosion will occur. Consequently, a blade will be removed from the unit and examined metallographically to determine the oxide wedge depth. Based upon the depth and time of operation, the remaining life can be estimated (Figure 26).
(15)
By utilizing this equation, the calculated FATT values were compared with critical FATT values and the results were very attractive (Figure 25).
Figure 26. Stress Intensity Profile Versus Oxide Wedge Depth for Unit A. Critical Oxide Wedge Depth for Failure Was Defined as ac. In the Failed Blade, a2 Was Found to Exceed ac. In an Intact Blade, a1 Was Less Than ac.
CASE STUDY 1 REMAINING LIFE ASSESSMENT OF STEAM TURBINE CASING

Background Information of the Turbines The subject ethylene plant has three steam turbines that drive three trains of compressors on the main deck. The units are listed in the following. The three turbines were commissioned in August 1981. They have been using the same steam source and are operated in similar conditions (as summarized in Table 2). This case study only discusses the remaining life assessment of one of them, a seven-stage 14 MW steam turbine. According to the plant logbook, in the past 23 years, the turbine was started/stopped 24 times, which are plotted in Figure 27. During the last 23 years of operation, only two failure incidents were recorded. The first one was a damage on the rotor, due to condensate inlet during the startup of the turbine in the late 1980s. The other was found in the latest turnaround in 2004, in which a blade failure and multiple cracks at steam balance holes were found on the sixth stage disk that was subject to repair. The casing never experienced any problem in the past. During the turnaround in 1995, a remaining life assessment was performed by Wiegand (1995), and the casing was found to be in a very good condition, as shown later in the top photo of Figure 30. The turbine component materials are listed in Table 3.
Figure 25. Comparison Between Measured FATT and Predicted FATT Using the Method of Etching.
HIGH TEMPERATURE CORROSION

Fluid catalytic cracking (FCC) hot gas expanders operate in environments that can be both corrosive and erosive. Although it is well documented that the source of erosion comes from the regenerated catalyst that is carried with the hot flue gas from the FCC, its effect on high temperature corrosion has only begun to be understood by the authors company. Papers published by the author outline the relationship of stress and temperature on the high temperature corrosion/fracture mechanics of Waspaloy in various catalyst environments (Dowson, et al., 1995; Dowson and Stinner, 2000). The nature of the corrosion attack is primarily influenced by the type of crude oil stock, which in time has a bearing on the resulting
87
Table 2. Turbine Operating Conditions.

Designed Conditions Temperature, Inlet 806 (430) F ( C) Inlet Pressure, psig (kg/cm2g) 696 (48.9) Actual Conditions Inlet Temperature, F ( C) 797 (425) Inlet Pressure, psig (kg/cm2g) 710 (50)
1st Stage Disk 692 (367)
12 11 10 9 8 7 6 5 4 3 2 1
19 82 19 83 19 84 19 85 19 86 19 87 19 88 19 89 19 90 19 91 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03 20 04
Month
Year
Figure 27. Turbine Start/Stop Record. Table 3. Turbine Component Materials.

Steam-End Casing
1.25Cr-0.5Mo alloy steel casting, ASTM A217 Grade WC6 (limited to 950 F service)
Figure 28. The Turbine Casing and the Replication/Cutoff Locations. F Represented the Hottest Spot, While E Represented the Coolest Spot. D Was at the Transition Radius of Steam Chest to Casing Barrel, Which Was Invisible From this Photo.
Diaphragms
Alloy steel plate ASTM A517 Grade F & A516 Grade 60 for 2nd to 5th stages. Gray cast iron ASTM A278 Class 40 for 6th & 7th stages.
Rotor Forging
Ni-Mo-V alloy steel, ASTM A470 Class 4
Blades
AISI 403 stainless steel for all stages
Field Inspection The turbine upper casing was visually examined. No cracks or reportable indications were found. The split-line bolt-holes were examined with dye-penetrant inspection, which did not reveal any cracks. The casing was polished at six different spots shown in Figure 28. The microstructure was observed with a portable optical microscope, and replicated for documentation (Figure 29). Microstructure deteriorations, such as carbide precipitation at grain boundaries and bainite degradation, were evident. However, the casing did not show any creep voids at any of the spots being examined. Hardness was measured at the same spots and the results are listed in Table 4. Microstructure Deterioration of the Casings Material In order to study the microstructure deterioration, a fragment of the sample being cut off from the casing was reheat treated to restore the original microstructure for comparison. The reheat treatment was done in two steps: normalized at 1685F (918C) for 1 hour and controlled air-cool, then tempered at 1250F (677C) for 2 hours and air-cool. The controlled air-cool from austenite temperature was to obtain a similar percentage ratio of proeutectoid ferrite and bainite as of the original heat treatment. Assuming the casting was 2 3/4 inch (70 mm) thick, a cooling rate of 54F per minute (30C/m) was employed and the resultant ferrite/bainite ratio simulated the original heat treatment very well. The reheat treated and service-exposed samples were mounted and polished and examined with an optical and microscope SEM. Their hardness was also measured. The findings are summarized in the following.
Figure 29. Micrograph of a Replica Taken from Spot D. Carbides Precipitated and Partially Networked at Grain Boundaries. However, No Creep Voids Were Observed There. Five Percent Nital Etched. Table 4. Casing Hardness Was Fairly Consistent from Spot to Spot.
Brinell Hardness [HB] (converted from Equotip data) Spot ID# A B C D E F 1st Reading 152 151 154 159 137 154 2nd Reading 150 151 151 158 141 154 3rd Reading 155 156 153 159 140 154 Average 152 153 153 159 139 154
Comparing
with the restored microstructure, the serviceexposed samples showed profound carbide precipitation at grain boundaries, and the extent of the precipitation increased with increasing the service exposure time (Figures 30, 31, 32).
88
Figure 31. Microstructure of the 23 Years of Service-Exposed Casing Material. Note Deteriorated Bainite and Carbides Precipitated at Grain Boundaries and in Ferrite Phase.
Figure 32. After Being Reheat Treated, the Bainite Phase Was Restored, and the Carbides Were Generally Dissolved.
Service-exposed
samples also showed carbide precipitation inside the proeutectoid ferrite phase.
Evidence of microstructure degradation of the bainite phase was

noticed in the service-exposed sample, due to the transformation of cementite (M3C) into different carbides (M2C and M7C3).
No creep voids were found in either sample. Hardness was measured as HRB 91 (HB 160) for the service-
exposed sample and HRB 93.5 (HB 169) for the reheat treated sample, indicating a slight hardness decrease after a long-term exposure at elevated temperature. Discussion and Conclusions Figure 30. The Extent of Carbide Precipitations at Grain Boundaries Increased with Increasing the Service-Exposure Time. These Precipitations Were the Results of Long-Term Diffusion at Elevated Temperature. A Reheat Treatment (Normalizing Plus Tempering) Generally Restored the Microstructure by Dissolving the Carbides. Five Percent Nital Etched. Turbine casings usually are normalized and tempered after being cast. This heat treatment leads to a desirable combination of strength/hardness and toughness/ductility for the casing material, 1.25Cr-0.5Mo alloy steel in this case. It also gives an excellent thermal stability for the casings. The typical treatment process contains normalizing at 1685F (918C) followed by air-cool then tempering at 1250F (693C) followed by air-cool.
89
The microstructure after the heat treatment consists of proeutectoid ferrite plus bainite. Bainite is a metastable aggregate of ferrite and cementite, which may degrade after long-term exposure at elevated temperature. The degradation was observed as a transformation of cementite (M3C) into M2C and M7C3 carbides (Biss and Wada, 1985). This transformation will lead to less carbides (in volume percentage) and coarser carbide particles, which will result in lower strength and hardness, which may reduce fatigue and rupture strengths of the material. Carbon will also diffuse along the grain boundaries and into the proeutectoid ferrite phase to form carbides. Carbides at grain boundaries, especially in networked form, are considered detrimental because they can result in embrittlement and reduce rupture ductility of the material. Another grain boundary damage that is also related to long-term diffusion but was not seen in this case is creep voids. Creep voids are considered as a higher-degree of microstructure damage than the carbide precipitation, and a warning sign of component remaining life. In the subject turbine casing, no creep voids were evident yet, so the microstructure deterioration was considered moderate and not immediately harmful to the casing life. It was concluded that the casing was in class A condition and could be safely used for the following service period (about five years). However, the microstructure deterioration imposed a concern of the integrity of the casing material and consequent potential risks on the creep strength and rupture ductility of the casings. It was reported that 1.25Cr0.5Mo low alloy steel could lose its rupture ductility after long-term exposure at elevated temperature (Demirkol, 1999). This behavior is so called rupture ductility trough, which is related to carbides precipitation and may occur before creep voids are developed. Therefore, in addition to continually monitoring the microstructure with nondestructive replica techniques in the next turnaround, the authors company suggested that the ethylene plant conduct a Level III remaining life assessment, which requires destructive material tests, in order to evaluate the extent of damage on creep strength and rupture ductility of the casing material. The material tests mainly consist of SRT and CDR test, which are described in a previous section. These two mechanical testings can supply information that microstructure examination (replica) cannot.
Figure 33. Diagram Indicating Location of Disk Replicas.
CASE STUDY 2 REMAINING LIFE ASSESSMENT OF HOT GAS EXPANDER DISK

A life assessment was performed on a disk from a hot gas expander that had been in operation for approximately 51,000 hours of service. The present life assessment was necessary to ensure that the properties of the disk had not degraded with time. This life assessment consisted of a microstructural evaluation as well as mechanical testing. Replicas of the microstructure of the disk were taken from five locations shown in Figure 33 and subsequently examined via optical microscope. A blade root from the disk that had seen the same accumulation of time was also sectioned, metallographically prepared using the standard techniques, and examined. Specifically the microstructure was examined in the area of the first, third, and fourth landings. Also specimens were prepared to examine corrosion products. Representative micrographs of the microstructure of the root of the blade and of the disk may be seen in Figure 34. The structure was found to be uniform and typical for A286 with an ASTM grain size ranging from four to six. No creep voids were observed. Several areas of high temperature corrosion attack were evident in the area of the first and third root lines of the blade that was sectioned. A micrograph of this corrosion may be seen in Figure 35. The maximum depth of penetration of corrosion product was found to be approximately 100 micrometers. No corrosion product was observed in areas removed from this accelerated attack, which suggests that a thin Cr2O3 layer had formed over the surface of the
Figure 34. Optical Micrograph of Typical Microstructure of Root of Blade #38. (15 ml HCI, 10 ml HN03, 10 ml Acetic Acid, 100).
Figure 35. SEM Micrographs of Cross Section of Corrosion Seen in Radius of First Root Landing of Blade #38. majority of the blade root and remained protective. This type of corrosive attack is expected for an iron chromium based alloy exposed to an atmosphere consisting of a mixture of sulfur and oxygen. Many high temperature alloys including A286 rely on the formation of the compact slow growing chrome oxide scale for protection from the environment. Exposure to sulfur may lead to the breakdown of this protective scale providing the necessary
90
conditions of temperature and pressure exist. The nature of the corrosive products depends on both thermodynamic and kinetic factors, the details of which are beyond the scope of this paper. A more detailed description of the mixed oxidants is given by Dowson, et al. (1995), and Birks and Meier (1983). It is not uncommon to see corrosion in the areas of blade root described in Figure 36. One explanation of this is that in these areas there is a gap between the blade root and the disks that allows catalyst into the blade root area. The presence of this catalyst may lead to high temperature corrosion as described by Dowson, et al. (1995), and Dowson and Stinner (2000). Some corrosion was also seen on the context surface of the land that is being proposed that corrosion in these areas is exacerbated by fretting. However, in this investigation this type of corrosion occurred to a much lesser extent than seen in the areas mentioned previously.
The stress rupture specimen from the blade was found to have an elongation of 10 percent and a reduction of area of 14 percent. The fracture occurred in the smooth section of the bar after 195.5 hours. This meets the authors company specification, which requires that the specimen last at least 30 hours with rupture occurring in the smooth section and possess an elongation of at least 10 percent. The mill certification of the original blade material gives an elongation and reduction in area of 11.7 and 19 percent, respectively, which is comparable to the present results. Thus, the rupture properties of the disk have not been significantly degraded as a result of operation of the expander. Modified Stress Rupture Testing This test developed by Dowson, et al. (1995), is used to determine the stress to cause fracture of the disk material in the presence of spent catalyst. The stress may be converted to a stress intensity that may be compared to the stress intensity at the tip of the wedge created by the corrosive product. A determination may then be made as to whether cracking in the disk will occur. It was found that the presence of spent FCC catalyst may greater reduce the rupture strength of Waspaloy. For example, a sample test of 1200F and a stress of 95 ksi lasted over 900 hours when tested in air, while the same temperature and stress produced failure in only 3 hours when the sample was tested in the presence of catalyst. In the present investigation, an initial test of 1200F and a stress of 54 ksi confirmed that A286 is also susceptible to this type of degradation. The sample failed in only 4.25 hours as opposed to 500 hours as predicted by the authors company material property database for stress rupture in air. Therefore, in order to get an accurate assessment of the rupture properties of the disk in the area of the root in question, the modified stress rupture test was used by Dowson and Stinner (2000). In the case at hand a temperature of 1065 F was chosen for the test based on the temperature of the disk rim at the point of maximum stress, which was determined by finite element analysis to be in the radius of the fourth root landing. The stress used in the tests range from 62 to 102 ksi. Each test consists of an exposure of stress and temperature for at least 20 hours. If the specimen did not break it was unloaded and the catalyst replaced and then retested at a stress that was increased by 10 ksi. This process, which was repeated until fracture occurred at a stress of 102 ksi, corresponds to a stress intensity of 24.27 ksi inch. It should be noted that the catalyst that was used is kept at the authors company for testing purposes, i.e., it was not from the actual end users unit. It is likely that this test catalyst was much more contaminated than the end users catalyst, which will result in lower values of rupture stress than if the actual end users catalyst was used. In order to determine the stress intensity of the tip of the corrosion wedge it was first necessary to determine the stress profile. The (X) of the area of the root in question, the stress intensity may be calculated from Equation (16):
Figure 36. Diagram Indicating Sectioning of Blade Root and Location of Corrosion. Because destructive testing could not be preformed on the actual disk material, samples for stress rupture testing were taken from the root and lower airfoil of the blade as shown in Figure 37. It was assumed that the properties of the sample are similar to those of the disk roots. Modified stress rupture tests as developed by Dowson, et al. (1995), and Dowson and Stinner (2000) were also performed in order to determine if the corrosion products seen in the blade root area would cause cracking of the disk.
KI =
(x)m(x, a)dx
0
(16)
where a is the wedge depth and the X is the coordinate dimension. The term M (x, a) is the weight function that is dependent on the geometry of the cracked body. When considering the corrosion, wedge formed within the area of a disk rim, the geometry may be approximated as an edge crack in a semi-infinite plate, Equation (17):
m( x , a )=
Figure 37. Photograph Illustrating Location from Which Rupture Specimen Was Taken Relative to the Blade Root/Lower Airfoil.
x 2 x 1+ .5693 1 + .279375 1 (17) a a 2 ( a x ) 2
A linearized stress field of the fourth root landing of the disk rim of the end users expander was determined by a finite element
91
analysis. From these data an expression of (X) was derived and the integral of Equation (16) evaluated to give the stress intensity at the tip of the corrosion wedge. The results of these calculations are summarized in Figure 38, which is a plot of stress intensity versus corrosion wedge depth.
Dowson, P., 1994, Fitness for Service of Rotating Turbomachinery Equipment, 10th Annual North American Welding Research Conference, Columbus, Ohio. Dowson, P., 1995, Fracture Mechanics Methodology Applied to Rotating Components of Steam Turbines and Centrifugal Compressors, Third International Charles Parson Turbine Conference, Materials Engineering in Turbine and Compressors, 2, pp. 363-375. Dowson, P. and Rishel, D. M., and Bornstein, N. S., 1995, Factors and Preventive Measures Relative to the High Temperature Corrosion of Blade/Disk Components in FCC Power Recovery Turbines, Proceedings of the Twenty-Fourth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 11-26. Dowson, P. and Stinner, C., 2000, The Relationship of Stress and Temperature on High-Temperature Corrosion Fracture Mechanics of Waspaloy in Various Catalyst Environments, Proceedings of High Temperature Corrosion and Protection, Materials at High Temperatures, 18, (2), pp. 107-118. Foulds, J. R., Jewett, C. W., and Viswanathan, R., 1991, Miniature Specimen Test Technique for FATT, ASME/1EEE Joint Power Generator Conference, Paper No. 91-JPGC-PWR-38, ASME, New York, New York. Goldhoff, R. M. and Woodford, D. A., 1972, The Evaluation of Creep Damage in CrMoV Steel in Testing for Prediction of Material Performance in Structure and Components, STP 515 American Society for Testing and Materials, Philadelphia, Pennsylvania, pp. 89-106. Kadoya, Y., et al., 1991, Nondestructive Evaluation of Temper Embrittlement in CrMoV Steel, ASME/JEEE Joint Power Generator Conference, Paper PWR-Vol. 13, New York, New York. Neubauer, E. and Wadel, V., 1983, Rest Life Estimation of Creep Components by Means of Replicas, Advances in Life Prediction Methods, Editors, Woodford, D. A., and Whitehead, J. E., (New York, New York: ASME), pp. 307-314. Newhouse D. L., et al., 1972, Temper Embrittlement of Alloy Steels, ASTM STP 499, pp. 3-36. Pope, J. J. and Genyen, D. D., 1989, International Conference on Fossil Power Plant Rehabilitation, ASM International, pp. 3945. Saxena, A., 1998, Nonlinear Fracture Mechanics for Engineers, Boca Raton, Florida: CRC Press, p. 449. Schwant, R. C. and Timo, D. P., 1985, Life Assessment of General Electric Large Steam Turbine Rotors in Life Assessment and Improvement of Turbogenerator Rotors for Fossil Plants, Viswanathan, R., Editor, New York, New York: Pergamon Press, pp. 3.25-3.40. Viswanathan, R. and Gehl, S. M., 1991, A Method for Estimation of the Fracture Toughness of CrMoV Rotor Steels Based on Composition, ASME Transaction, Journal of Engineering Mat. and Tech., 113, p. 263. Viswanathan, R. and Gehl, S. M., 1992, Life-Assessment Technology for Power Plant Components, JOM, February, pp. 34-42. Viswanathan, R. and Jaffee, R. J., 1983, Toughness of Cr-Mo-V Steels for Steam Turbine Rotors, ASME Journal of Engineering Mat. and Tech., 105, pp. 286-294. Viswanthan R. and Wells, C. H., 1995, Life Prediction of Turbine Generator Rotors, Third International Charles Parsons Turbine Conference, Materials Engineering in Turbines and Compressors, 1, pp. 229-264.
Figure 38. Stress Intensity Versus Corrosion Wedge Depth for Fourth Root Landing of Disk. Based on this plot, a corrosion wedge will have to penetrate to an approximate depth of .021 inches before the critical stress intensity is exceeded and catastrophic fracture occurs. As mentioned, the maximum penetration seen was approximately .004 inches. Because the growth of the corrosion product is parabolic, e.g., the growth slows with time, it is unlikely that the corrosion would cause a failure of the disk providing the temperature and atmosphere of the expander are not changed significantly. Conclusions The microstructure in the ruptured product has not been significantly affected by the operation of the expanders. Some high temperature corrosion was seen in the root area; however, it has been determined that this corrosion will not result in the failure of the disk. Therefore, the disk will be used without significant risk of failure. It is recommended that this be reevaluated after an operational period of four to five years.
REFERENCES
Baik, J. M., Kameda, J., and Buck, O., 1983, Scripta Met., 17, p. 1143-1147. Birks, N. and Meier, G. H., 1983, Introduction to High Temperature Oxidation of Metals, London, England: Edward Arnold. Biss, V. A. and Wada, T., 1985, Microstructural Changes in 1Cr0.5Mo Steel after 20 Years of Service, Metallurgical Transactions A, 16A, January, pp. 100-115. Bruscato, R. M., 1970, Temper Embrittlement and Creep Embrittlement of 2 Cr-1 Mo Shielded Metal-Arc Weld Deposits, Welding Journal, 35, p. 148s. Bush, S. H., 1982, Failures in Large Steam Turbine Rotors, in Rotor Forgings for Turbines and Generators, R.I. Jaffe, Editor, New York, New York: Pergamon Press, pp. I-1 to I-27. Carlton, R. G., Gooch, D. J., and Hawkes, B. M., 1967, The Central Electrical Generating Board Approach to The Determination of Remanent Life of High Temperature Turbine Rotors, I. Mech E. Paper C300/87. Demirkol, M., 1999, On the Creep Strength-Rupture Ductility Behavior of 1.25Cr-0.5Mo Low Alloy Steel, Journal of Engineering and Environmental Science, 23, pp. 389-401.
92
Watanabe, J. and Murakami, Y., 1981, Prevention of Temper Embrittlement of CrMo Steel Vessels by the Use of Low Si Forged Steels, American Petroleum Institute, Chicago, Illinois, p. 216. Wiegand, R., 1995, Petroquimica Bahia Blanca Turbine Casing Assessment, An Internal Report. Woodford, D. A., 1993, Materials and Design, 14, (4), p. 231. Woodford, D. A., Von Steele, D. R., Amberge, K., and Stiles, D., 1992, Creep Strength Evaluation for IN738 Based on Stress Relaxation from Superalloys, The Minerals, Metals and Material Society, Editors, Antolovich, S. D., Stusned, R. W., Mackay, R. A., Anton, D. L., Khan, T., Kissinger, R. D., and Klarstrom, D. L., pp. 657-663.
ACKNOWLEDGEMENT
The authors are grateful for the support from the Materials Engineering Department at Elliott Company, and recognize Elliott Company and Dow Chemical PBBPolisur for permission to publish this paper.
AXIAL ALIGNMENT AND THERMAL GROWTH EFFECTS ON TURBOMACHINERY TRAINS WITH DOUBLE-HELICAL GEARING
by Thom Eldridge
Manager of Aero/Thermodynamics Engineering Design Dresser-Rand Company Olean, New York
Greg Elliott
Senior Project Engineer
Ed Martin
Project Engineer Lufkin Industries Lufkin, Texas
and M. Shukri Hitam

Turbomachinery Group Superintendent Petronas Carigali Terengganu, Malaysia
Thom Eldridge is Manager of the Aero/Thermodynamics Engineering Design Department at Dresser-Rand Company in Olean, New York. His previous responsibilities as supervisor of the Rotordynamics Group included managing the analysis of turbomachinery and guiding rotordynamic development activities. Before joining Dresser-Rand, he worked on bearing development, turbomachinery modeling, data acquisition systems, and high-temperature transducer design as a senior engineer and project team leader for Bently Nevada. Mr. Eldridge received a B.S. and an M.S. degree (Mechanical Engineering, 1991, 1994) from Washington State University, and an MBA degree (2002) from the University of Nevada. He is a registered Professional Engineer in the State of California. Mr. Eldridge holds two patents and has written several papers on damper seals and hydrostatic bearing applications. Greg Elliott is a Senior Project Engineer in the Power Transmission Division of Lufkin Industries in Lufkin, Texas. His primary responsibilities include product development and standardization, as well as engineering systems development. He also supports Lufkin engineering groups with finite element analysis, fatigue analysis, and other applications of engineering mechanics in machinery design and problem solving. Before joining Lufkin, he worked in research and teaching at Texas A&M University. Mr. Elliott received a B.S. and an M.S. degree (Agricultural Engineering, 1982, 1990) from Texas A&M University.
93
Ed Martin is a Project Engineer in the Power Transmission Division of Lufkin Industries, in Lufkin, Texas. He is currently responsible for the engineering functions of Lufkins high-speed gear product line. He also performs lateral rotordynamic analyses and has been heavily involved in troubleshooting gear vibration, noise, and temperature problems during his 14-year career with Lufkin. Mr. Martin received a B.S. degree (Mechanical Engineering, 1990) from the University of Texas at Austin. He has given numerous presentations on gearing principles and vibration issues, and is also an active member of the AGMA 6011 committee on high-speed helical gearing.
ABSTRACT
Turbomachinery shafting and casings typically experience axial growth during thermal transients occurring at startup, shutdown, and changes of load. This growth does not usually present difficulties during operation. However, this paper discusses four recent turbomachinery trains that experienced radial subsynchronous vibration resulting from axial misalignment. In each example, the train included a gearbox with a double-helical gearset that presented a lateral vibration response at system torsional critical frequencies while operating at low load. Thermal growth resulted in axial misalignment that produced a zero-backlash condition when one helix contacted on the back side of the teeth. Data collected with an axial displacement probe verified the appearance of lateral subsynchronous vibration was associated with excessive axial misalignment of the gearset. Data are presented from these experiences, including a discussion of other vibration phenomena that could be mistaken for this behavior. Finally, design and analysis recommendations are provided for preventing such occurrences.
94
INTRODUCTION
The authors have encountered several instances of unacceptable gearbox lateral vibration arising from forces external to the gearbox caused by axial misalignment. Most of the lessons learned from these experiences build upon previous findings (Zirkelback, 1979; Mancuso, 1986; Carter, et al., 1994). However, most emphasis has been placed on parallel or angular alignment problems. Although axial alignment problems may occur less frequently, the delays and expense caused by such problems make it worthwhile to share this information with others in the rotating equipment community. This paper presents necessary background, examples, and recommendations for problem diagnosis and prevention. Helical Gearing Background Parallel-shaft, double-helical gearing is widely used in high-speed, high-power turbomachinery applications. In these applications, double-helical designs usually are the best choice because of their high power density, smooth meshing, reduced thermal distortion, high-energy efficiency, and high reliability. The term doublehelical refers to the use of two sets of helical teeth arranged with opposing helix directions. Figures 1 and 2 show typical doublehelical gearsets used in turbomachinery trains. Figures 3 and 4 illustrate basic gear terminology (Dudley, 1984; Townsend, 1991). Figure 3. Gear Terminology.
Figure 4. Gear Mesh Centered.
FMESH = ( 2T / DP ) tan( H )
where: FMESH T DP H Figure 1. Double-Helical Gearset Used in a MotorCompressor Drive. = = = = Mesh centering force (lbf) Transmitted torque (inch-lbf), per helix Pitch diameter (inch) Helix angle (degrees)
(1)
The force can be quite strong. For example, in a typical doublehelical gearset with a 12 inch (305 mm) pitch diameter and a 28-degree helix angle transmitting 40,000 horsepower (30,000 kW) from a gas turbine at 5000 rpm to a centrifugal compressor at 10,000 rpm, each helix generates an axial force of about 11,000 lbf (49,000 N). When applying the formula, one must use the transmitted torque and pitch diameter together for the same element. When applying it to double-helical gearing, the actual transmitted torque per helix should be used. Under normal circumstances, the mesh forces act equally on each element in opposite directions. This causes the axial mesh forces to cancel internally within a double-helical gearset. Canceling the axial force eliminates the need for high-capacity thrust bearings that would otherwise be required in the gearbox to absorb the axial mesh force. Typically, a gearbox with doublehelical gearing will have one thrust bearing on the low-speed gear. If the connected equipment permits, a double-helical gearbox may have no thrust bearing. Additional information on helical gearing can be found in numerous references (Dudley, 1984; Drago, 1988; Dudleys Gear Handbook, 1991). Double-Helical Mesh Axial Alignment The axial alignment problem discussed in this paper pertains to double-helical gearing. If the external axial forces on the gear elements are less than the mesh centering force, the mesh holds the pinion centered on the gear as shown in Figure 4. If the external axial force is zero, then the torque load on each helix is equal.
Figure 2. Double-Helical Gearset Used in a Gas Turbine Compressor Drive. Axial thrust induced by gear mesh force is a characteristic of helical gearing that is particularly significant to the topic of axial alignment. The axial mesh force of a set of helical gear teeth is calculated (Dudley, 1984) as:
95
Backlash (clearance) between the teeth is provided in the gearset to avoid contact of the nondriving (back) sides of the teeth. In double-helical gearing, backlash can be defined in the circumferential and axial directions. Circumferential backlash is illustrated in Figure 3. The axial backlash is equal to the circumferential backlash divided by the tangent of the tooth helix angle. For typical turbomachinery applications, a small double-helical gearset might have 0.015 inch (0.4 mm) circumferential backlash and 0.028 inch (0.7 mm) axial. Large gearsets may have three or four times those amounts. However, it is important to note that the axial backlash so defined includes the total potential axial displacement of the pinion relative to the gear in both directions. The pinion can move off center of the gear in each direction only by a distance equal to half the axial backlash. For a gearset with 0.028 inch (0.4 mm) of axial backlash, the teeth will bind if one attempts to shift the pinion more than 0.014 inch (0.4 mm) off center in either direction relative to the gear. This is illustrated in Figures 4, 5, and 6.
in Figure 5). This can occur, for example, when a flexible coupling is involved. If axial backlash is sufficient, the coupling axial force decreases as the pinion moves off center until the coupling axial force equals the mesh axial force. In Figure 5, the backlash is sufficient to maintain clearance at the nondriving sides of the teeth and permit the gearset to run smoothly. Note that when the gearset is running on one helix, the total transmitted torque is carried on that helix alone. Typically, when service torque load is applied, the mesh force increases and the gearset centers itself axially. If the external axial force is significant compared to the axial mesh force at service torque, it may cause unequal loading of the helices and decrease load carrying capacity. If the external force is greater than the axial mesh force, it will push the gearset into a bound condition as shown in Figure 6. In this state there is still contact on the normally loaded (driving) tooth flanks on one helix. On the other helix there is contact on the back (nondriving) flanks of the teeth (which should not normally be in contact). Running with contact on the driving side of one helix and the nondriving side of the other helix is similar to operating with zero backlash. It is well known that high-speed, high-power gearing should not be run without backlash (Dudley, 1984). Quantifying External Axial Forces As previously identified, external axial forces can originate from a number of sources. Frequently such a force is generated by the axial displacement of a flexible element coupling (in compression or tension). Many turbomachinery trains include either a disc or diaphragm coupling between the gearbox and the connected equipment shaft ends. Axial displacement of coupling flexible elements can originate from prestretch of the coupling, axial alignment inaccuracies, and from thermal growth of equipment (both steady-state and transient conditions). The main purpose of coupling prestretch is to cancel out the steady-state thermal growth of the coupling and shaft ends. In this paper, transient thermal growth and axial alignment errors are shown to result in excess axial forces that are capable of forcing the pinion out of center with the gear of a minimally loaded gearset. A simple estimate of the axial force generated by a flexible element coupling can be obtained through the axial stiffness of the coupling:
Figure 5. Gear Mesh Offset Axially.
FAXIAL = DSE K
(2)
where: FAXIAL = External axial force (lbf) DSE = Relative displacement of shaft ends attached to the coupling (inch) K = Axial stiffness of the flexible element coupling (lbf/in) The expected displacement of the shaft ends can be calculated. Commonly it will be a function of axial alignment error and thermal growth. Evaluation of the mesh centering force [Equation (1)] defines an upper bound to the allowable external axial force [Equation (2)]. This establishes a maximum allowable error in axial alignment. It also provides a guide to the maximum allowable transient thermal growth, before disengagement of the gear elements should be anticipated. In instances of large thermal transients, it may be necessary to modify the steady-state axial alignment to accommodate the transient thermal growth condition. To put this in perspective, consider the hypothetical 40,000 hp (30,000 kW) compressor drive mentioned previously. Assume an external axial force is applied to the pinion by axial displacement of a typical flexible-disc type coupling, K = 5000 lbf/in (875 kN/m). At steady-state operating conditions with reasonably loose alignment accuracy, less than 0.050 inch (1.3 mm) error, the axial coupling force would be 250 lbf (1100 N). For the gearset of this example, a 250 lbf axial mesh centering force would be generated by a helix transmitting 450 hp (335 kW). The mesh will remain centered if the transmitted power is greater than 1 percent of
Figure 6. Gear Mesh Bound Axially. In turbomachinery trains, often there will be axial forces caused by axial misalignment, thermal effects, thrust bearings, magnetic centering forces from electrical machines, gravity loads during ship pitch, and so on. As previously described, when an external axial force is applied to the double-helical gearset, it offsets part of the axial mesh force. This causes no harm (or axial displacement) if the external force is small compared to the axial mesh force. That is typically the condition while service torque is applied. If the external force is larger than the mesh axial force, the two meshing gear elements can be pushed axially out of proper engagement as shown in Figure 5. This is most likely to occur when torque load is small, as may be encountered with a no-load string test, standby operation of a generator set, or during startup or shutdown transients. If the source of the external force reaches equilibrium with the axial mesh force, the gearset may run satisfactorily on one helix (as
96
normal load. If a typical diaphragm-type coupling was used (K = 20,000 lbf/in), the load necessary to keep the mesh centered could be closer to 3 or 4 percent of normal transmitted load.
INTRODUCTION TO EXAMPLES
Based on experiences such as those described in this paper, the authors believe the resulting nonuniform transmission of motion from a bound mesh can supply broad-spectrum angular or lateral excitation to the system. Experience has shown this can lead to excessive vibration by directly stimulating lateral displacement or through interaction of lateral and torsional vibration in the gearset (Hudson, 1992). Additionally, low bearing stiffness (such as encountered in a gearbox with pressure-dam bearings operating under light load) has been reported to increase the interaction between torsional and lateral response (Pennacchi and Vania, 2004). Finally, the combination of running on both the loaded and unloaded faces of the gears produces high relative variation in gear mesh stiffness. Variation in gear mesh stiffness has been discussed as a source of torsional excitation capable of producing noticeable lateral response (Iwatsubo, et al., 1984). If the axial force is extremely high, the wedging effect can cause tooth overload and failure, and thrust bearing wear can be accelerated. The following examples illustrate practical problems that occurred because of thermal transients and axial misalignment. Four examples are described with emphasis on their vibration signatures. They highlight some common causes of excessive axial force in turbomachinery trains. Additionally, they provide lessons for designing trouble-free systems and for diagnosing problems. The first example will be discussed in considerable detail to illustrate the diagnostic process and illuminate how similar phenomena can be mistakenly blamed for the appearance of this subsynchronous vibration. The mechanics of the thermal transient growth and gear binding in Example 1 are presented in APPENDIX A. The second through fourth examples are included to present other train configurations that have experienced these phenomena and to illustrate some points not covered in the first example.
As shown in Figure 8, string test results typically showed a variety of subsynchronous vibration (SSV) peaks in the gearbox low-speed (input) shaft displacements with most notable components near 30 to 40 and 70 to 80 Hz. The highest SSV peaks were approximately 0.0005 inch (0.013 mm) near 40 Hz. Amplitude at 1 was about 0.0005 inch (0.013 mm) and overall levels approached 0.002 inch (0.05 mm). Significant SSV appeared only after operating at MCOS for 12 to 30 minutes, disappearing after returning to idle speed. Only the gearbox low-speed shaft exhibited the SSV. All gearbox journal and thrust bearing temperatures were as expected. Review of relevant frequencies identified a small SSV peak near 35 Hz in the gearbox shop test. It also was noted that the predicted first torsional resonant frequency of the train was 41.8 Hz and the second was 75.5 Hz (refer to Figure 9). Preliminary diagnosis indicated axial alignment and journal bearing instability as possible contributors to the SSV problem.
Figure 8. Radial Vibration of Gearbox Input Shaft During Typical String Test of Example 1.
EXAMPLE 1 GAS TURBINE, GEARBOX, CENTRIFUGAL COMPRESSOR

This problem arose during a no-load string test of a 15,000 hp gas turbine, speed-increasing gearbox, and centrifugal compressor shown schematically in Figure 7. Maximum continuous speeds (MCOS) were 9300 rpm (155 Hz) at the turbine and 11,840 rpm (197 Hz) at the compressor. The original string test setup was open-air, the compressor had inlet and outlet silencers and absorbed only an estimated 250 hp. This train was very similar to an existing design that had performed well in field operation.
Figure 9. Torsional Model of Train from Example 1. Diagnostic Run without Compressor Figure 7. Layout of Gas Turbine, Gearbox, Centrifugal Compressor Train in Example 1. A diagnostic test was run with the compressor uncoupled from the gearbox output shaft. If the issue was journal bearing instability, the problem would get worse as the minimal 250 hp
97
compressor load was removed. Decreasing bearing load by decreasing torque should make the cylindrical bore, pressure-dam bearings in the gearbox less stable. If the issue was an axial alignment error, eliminating the compressor (and its thrust bearing) would eliminate the problem. Performance of the unit during the uncoupled test showed the SSV had been eliminated (refer to Figure 10). The SSV peak just below 40 Hz in the original vibration spectrum was gone. A smaller peak appeared at 60 Hz, very close to the calculated 59.5 Hz first torsional resonance with the compressor removed from the train. This was a strong indication that there was some level of torsional/lateral interaction occurring in the train and that interaction was contributing to the SSV. This validated the theory that the source of the vibration was alignment related, most likely axial alignment.
bearing design could eliminate bearing stability as a possible factor and reduce vibration levels (Nicholas, et al., 1980). The reduced vibration level would also avoid machine protection trips and should provide a longer diagnostic run to help clarify the effect of other potential causes. The gear supplier developed a modified journal bearing design that was optimized for stability at no load and could be implemented by modifying the bearings that were in the machine. The low-speed (input) shaft journal bearings were removed from the gearbox, modified, and reinstalled. At the same time the journal bearings were being modified, a provision was added to the gearbox for installing an axial proximity probe at the high-speed pinion. The intent was not to measure axial vibration of the pinion, but to monitor the axial position of the pinion and provide insight on axial alignment. It was expected the pinion axial probe would enable detection of gear mesh axial misalignment similar to Figure 5 or 6 if such a condition was present. Diagnostic Run with Axial Probe on Pinion and Modified Journal Bearings This run revealed that the subsynchronous vibration was correlated with axial misalignment of the gear mesh. Simultaneously monitoring the low-speed shaft radial vibration and pinion axial movement was the key to solving the problem. Verifying the pinion position and subsynchronous vibration levels changed in concert during the test illuminated the solution. As shown in Figure 11, subsynchronous vibration appeared much the same as it had in previous runs. At about 8:42 the speed was increased from 9000 rpm to MCOS of 11,844 rpm. The lateral vibration probes showed that until 8:57:36 a clean subsynchronous spectrum was present with low amplitudes and no significant discreet components (light data in Figures 11 to 13). After that time, two significant SSV frequencies appear at 38 and 82 Hz (roughly corresponding to the first and second torsional resonances). This SSV remained consistent for 65 minutes (dark data in Figures 11 to 13) and then for a brief period the SSV presented varying frequencies. After that, a clean subsynchronous spectrum was present (second light data in Figures 11 to 13). During the deceleration at 11:00, the SSV reappeared momentarily, but subsided once the new speed of 9000 rpm was maintained. The modified (stiffer) bearings appeared to reduce subsynchronous discrete amplitudes and eliminated much of the nondiscreet SSV (noise or hash). However, they did not prevent the SSV from occurring.
Figure 10. Radial Vibration During Diagnostic Test Run of Turbine and Gearbox Alone in Example 1. Diagnostic Run with Decreased High-Speed Coupling Prestretch An additional 0.020 inch (0.51 mm) shim was added to the highspeed coupling to decrease the prestretch to 0.018 inch. This did not solve the problem. Results at MCOS were similar to the initial runs, except that the SSV peaks were higher in amplitude and less vibration hash appeared between the peaks. SSV appeared approximately 10 to 15 minutes into the run, and discreet components reached 0.0011 inch (0.028 mm) levels. Frequencies were much more distinct at 39, 78, 116, and 155 Hz, all of which are related to the 155 Hz running speed. It appeared decreasing prestretch made vibration worse during shutdown. When the unit was manually tripped, the lateral vibration spiked to higher levels. This result focused attention on nonalignment issues. Diagnostic Run with Light Torque Load At this point attention turned to load. There was consensus that if sufficient torque load was placed on the train, the problems being experienced would be entirely eliminated, regardless of whether the problem was caused by axial misalignment or journal bearing instability. However, the only way to increase load on the train during the string test was to install an orifice on the compressor output, which was expected to add 250 hp additional load over the unthrottled configuration. This was done, but the resulting 500 hp load was not enough to make a significant difference, except that the time to onset of SSV was increased slightly. Gearbox Modifications Focus shifted to the gearbox journal bearing design as a means of decreasing vibration. Although rotordynamic analysis of the gearbox journal bearing design indicated there should be stable operation (even at zero load), it was decided that a different
1X
Figure 11. Radial Vibration after Installing Pinion Axial Probe in Example 1. Figures 12 and 13 show high-speed pinion and low-speed gear axial positions. In these axial position plots 200 mV output voltage change corresponds to 0.001 inch (0.025 mm) axial position change (200 mV/mil). For example, during the period from 8:42
98
Figure 12. High-Speed Pinion Position Plot for Example 1.
Figure 14. Vibration with Increased High-Speed Coupling Prestretch in Example 1.
Figure 13. Low-Speed Gear Position Plot for Example 1. until 8:56 the pinion probe voltage changed 3.4 volts. That represents a movement of 0.017 inch (0.43 mm). Not only was the pinion undergoing substantial axial displacement, comparing pinion and gear positions revealed that the pinion moved considerably further axially than the gear. This measurement verified the pinion and gear were not remaining axially centered on one another (similar to the condition depicted in Figures 5 and 6). Further evaluation of the data showed that the SSV began when the mesh became tightly bound (transition from light to dark data). Details of the evaluation are presented in APPENDIX A. Final Qualification Run with Increased Prestretch in High-Speed Coupling Based on results of the previous run, the coupling prestretch was increased an additional 0.040 inch (1 mm) to 0.058 inch (1.5 mm) and another run was completed. The modified bearings from the previous run were used. The gearbox was able to run through the thermal transients and be subjected to numerous speed cycles without the SSV appearing (refer to Figures 14 and 15). The axial probe indicated the pinion axial position leveled off smoothly at the maximum displacement, with no indication of mesh binding. Based on these results it was concluded that the problem was solved. The unit has subsequently entered service. No SSV has been detected during startup, steady-state, transient/load change, or shutdown.
Figure 15. Pinion Axial Position Leveling Off Gradually in Example 1. steam turbine is the driver. It is coupled to the high-speed pinion through a flexible element coupling. The low-speed shaft of the gearbox carried the thrust bearing and was connected to an electrical generator rotating at 1800 rpm. The train layout is shown in Figure 16.
Figure 16. Layout of Turbine, Gear, Generator Train of Example 2. During commissioning of the train, one of the checks performed was to unload the steam turbine at full speed by removing the generator load. This produces an unloaded condition in the gearbox. The sudden unloading of the steam turbine resulted in rapid cooling (and shortening) of the turbine rotor as the steam rate fell. During the first unloading cycle a subsynchronous vibration was measured
EXAMPLE 2 STEAM TURBINE, GEARBOX, GENERATOR

Similar vibration behavior was experienced on a steam turbinedriven generator with a speed-reducing gearbox. In this train, the
99
on the gearbox pinion shaft as shown in Figure 17 with no other excessive vibration appearing elsewhere in the train. The vibration appeared discreetly at 750 cpm and 2100 cpm, with magnitude greater than the synchronous component. These SSV frequencies corresponded to the predicted first and second torsional natural frequencies of the train, 755 cpm and 2230 cpm, respectively.
1X
1X
Figure 18. Radial Vibration after Installing Pinion Axial Probe in Example 2.
Figure 17. Radial Vibration of Pinion in Example 2. Diagnostic Run with Increased Coupling Prestretch The axial alignment of the train was modified after these characterization runs by increasing the prestretch of the high-speed coupling by 0.023 inch (0.6 mm). With this modification, the SSV continued. It was hypothesized that the source of the vibration could be a thermal transient, causing axial displacement of the shaft end and shifting of the pinion to gear alignment. The sudden cooling of the steam turbine rotor had the potential to produce 0.10 inch (2.5 mm) of shaft-end movement. Unlike the situation described in Example 1 (when the growth of the compressor shaft tended to push the pinion shaft end toward the gear), in this example the motion of the turbine shaft end would be to pull the pinion shaft end away from the gearbox. In Examples 1 and 2, the axial force applied to the pinion is sufficient to overcome the mesh centering force of the double-helical gear set and drive the pinion out of axial alignment with the gear. With sufficient axial motion, the axial backlash of the gears would be consumed and the gearset would become bound. Diagnostic Run with Axial Probe on Pinion To verify the hypothesized behavior, an axial probe was installed in the gearbox to measure motion of the pinion. Figures 18 and 19 present similar behavior to that observed in Example 1. A sudden discontinuity in the pinion axial position [approximately 0.020 inch (0.5 mm) motion] trace corresponded to the rise of subsynchronous vibration. This 0.020 inch (0.5 mm) displacement during operation agreed with the 0.018 inch (0.45 mm) of axial end play measured during installation checks. Diagnostic Run with Modified Thrust Bearing Clearance In order to remove the constraint on pinion motion imposed by the low-speed gear, the axial clearance of the thrust bearing was increased by 0.040 inch (1 mm) with the modification of the shim pack. The shims were intentionally changed to move the axial running position of the gear closer to the steam turbine. In effect, this decreased the prestretch of the high-speed coupling, decreased the pulling force on the pinion during the cooling transient (and increased the prestretch of the low-speed coupling). The behavior of the gearbox improved significantly. Limited vibration was seen during subsequent unloading and deceleration runs. The unit entered service to the satisfaction of the customer. Figure 20 shows the much cleaner waterfall spectrum. Prior to the changes, the 750 cpm component was equal to the 1 vibration level, while the 2100 cpm had been up to three times the 1 vibration level. Figure 20. Radial Vibration after Increasing Thrust Bearing Clearance in Example 2.
Figure 19. High-Speed Pinion Axial Position in Example 2.
EXAMPLE 3 STEAM TURBINE, GEARBOX, GENERATOR

Before the experiences described in Examples 1 and 2, a similar vibration behavior was encountered on another steam turbinedriven generator with a speed-reducing gearbox (layout shown in Figure 21). Although it was not understood at the time of the incident, it later became apparent that this was another instance of axial misalignment. As in Example 2, the steam turbine is the driver and is coupled to the high-speed pinion through a flexible element coupling. In this example, there were two separate but identical trains, and only one of the trains experienced the high vibration.
100
EXAMPLE 4 GAS TURBINE, GEARBOX, GENERATOR

This train contained a gas turbine driving a generator through a speed-reducing gearbox. The turbine was connected to the gearbox pinion by a diaphragm-type coupling. The gearbox low-speed shaft flange was bolted directly to the generator shaft flange without a coupling. As shown in Figure 23, there were thrust bearings at the nondrive end of the single-shaft gas turbine and on each side of the low-speed gear. The subsynchronous vibration occurred during commissioning while running at full speed and no load.
-
Figure 21. Layout of Turbine, Gear, Generator Train of Example 3. During commissioning of one of the trains, the gear unit repeatedly tripped on high radial vibration of the pinion during startup. An examination of the pinion shaft vibration cascade plot (Figure 22) during startup revealed the main component of radial vibration to be subsynchronous, occurring at approximately 10 Hz once the train reached a generator speed of 950 rpm (full speed = 1800 rpm). It was later realized that this SSV frequency corresponds to the trains calculated first torsional natural frequency of 10 Hz. Figure 23. Layout of Gas Turbine, Gear, Generator Train of Example 4. The train was started and brought to full speed at no load. During the first hour radial vibration was low. After running at no load about an hour, the high-speed pinion radial vibration began increasing and continued to increase during a period of several minutes until the machinery protection system tripped on high pinion radial vibration. This trip occurred shortly before the turbine was expected to complete its startup thermal transient. Limited vibration data were available from a condition monitoring system. The pinion radial vibration had a significant spectral component in the general vicinity of lowspeed gear design rotating speed of 1800 cpm. Initially it was thought the pinion SSV was a forced response at the low-speed gear rotating frequency. Following a test run to obtain better vibration data, it was determined the dominant SSV peak did not track gear rotating frequency. Further, the vibration could be observed in casing acceleration and pinion displacements. Figure 24 is a representative casing velocity spectrum showing frequencies involved. This sample was taken at rotating frequencies of 5109 cpm (85.1 Hz) at the turbine and 1800 cpm at the generator. The vibration frequency peak was fixed near 1875 cpm (31.2 Hz). Once again there was circumstantial evidence that torsional effects could be a factor because the predicted first torsional natural frequency was 1894 cpm, very close to the response frequency.
Figure 22. Radial Vibration of Pinion of Example 3. At the time of the incident, the focus of the investigation centered on a journal bearing oil whirl. Accordingly, the pinion bearings were redesigned in an effort to suppress whirl. Before the installation of the modified bearings, an error was discovered on the installed prestretch of the high-speed coupling. The installed prestretch was 0.030 inch (0.8 mm) less than the calculated required value. This mistake was not made on the second train, which operated well and showed no signs of SSV. The decision was made to correct the prestretch by removing 0.030 inch (0.8 mm) of shim in the high-speed coupling at the same time the modified pinion bearings were being installed. After the modifications, the pinion was able to reach full speed while showing no signs of SSV, and the train was commissioned successfully. An interesting point about this example is that there were two identical trains, but only one experienced a problem. The train with the correctly installed prestretch and original bearing design did not exhibit SSV.
Figure 24. Original Vibration Spectrum for Example 4.
101
Axial alignment of the turbine to the gearbox pinion was reviewed. It was found that coupling axial thermal growth was not included in the original high-speed coupling prestretch calculation. The new analysis indicated that before steady-state conditions were reached there would be enough thermal growth in the turbine, coupling, and pinion to push the pinion away from the turbine into a bound condition as shown in Figure 6. Further, the growth was sufficient to push the low speed shaft through its thrust bearing clearance. Once the mesh was bound and the gear was pushed against its thrust bearing the remaining thermal growth had to be absorbed by axial deflection of the high-speed coupling. The relatively high coupling axial stiffness enabled even a small amount of coupling deflection to create sufficient force to bind the mesh. Diagnostic Run with Increased Coupling Prestretch A test run was made with high-speed coupling prestretch increased from 0.110 to 0.130 inch (2.8 to 3.3 mm) and with an axial position probe installed at the high-speed pinion. This change was not sufficient to solve the problem. However, the axial probe confirmed that the pinion was moving axially as predicted. The machine appeared to run acceptably on one helix, but the SSV picked up when the pinion reached its maximum axial displacement relative to the gear. Further increase of coupling prestretch was considered undesirable because of a turbine shaft axial excursion of 0.16 inch (4.1 mm) away from the gearbox early in the run. Thus another means of reducing axial misalignment was sought. Diagnostic Run with Modified Thrust Bearing Clearance Gearbox thrust bearing clearance was increased for the next run. The resulting combination of gear mesh and thrust-bearing clearance in the gearbox allowed 0.06 inch (1.5 mm) of pinion axial motion. Based on the axial alignment review, this was believed to be sufficient to absorb the thermal transients. Following the modification, the machinery protection system did not trip. Furthermore, the axial probes showed both the pinion and low-speed gear were pushed axially away from the turbine. Vibration began to increase at roughly the same time after the start of the run as it had previously. However, before the vibration reached an undesirable level, the pinion and gear were both observed to axially shift away from the turbine and the vibration decreased. This axial shift was possible only after the gearbox thrust bearing clearance was increased. The shift allowed the pinion to move away from the turbine without tightly binding the mesh. This was additional evidence that transient axial misalignment had caused the SSV problem. The startup trip did not reappear in subsequent runs either with or without load. The modification permitted the commissioning process to continue.
Evaluate the possibility of oil whirl or oil whip if fixed geometry
bearings are used. These are complicated phenomena beyond the scope of this paper, but for most gearboxes a few basic observations using a waterfall plot are sufficient for a first evaluation. Oil whirl usually can be ruled out if the frequency of shaft motion does not remain at about 40 to 45 percent of running speed as speed increases. On a waterfall plot oil whip looks like oil whirl initially but instead of continuing to track running speed it locks onto a resonant frequency and remains at that frequency as speed increases.
Monitor pinion and gear relative axial positions in conjunction
with vibration during the test run. If there is no existing axial probe provision, one can be added. In Examples 1, 2, and 4, an eddycurrent proximity probe at the pinion allowed direct verification of pinion axial movement. It is not necessary to have a high-quality axial probe target area since the purpose is to measure relatively large rotor movements using a direct current (DC) signal. Pinion and gear motion should be compared to the axial backlash in the gearset. Timing of these motions can be compared with changes in vibration.
Verify coupling (or enclosure) temperature rise and compare to

the expected value.
Machinery protection system trips can hinder diagnosis. The
equipment suppliers involved can be consulted for ways to reduce response or for other options to prevent unnecessary trips that hinder diagnosis. For instance, in Example 1 a temporary gearbox bearing change reduced response during the diagnostic process.
Perform test runs with varied loading. Lighter load reduces the Perform a test run with the driven equipment uncoupled. In
mesh centering force and may make the problem worse. Heavier load increases mesh centering force and makes the gearset more resistant to external axial forces. addition to removing load, this can remove one source of axial force on the gearset. This test is most productive if the gearbox output shaft does not have a thrust bearing.
DESIGN AND ANALYSIS TO PREVENT AXIAL ALIGNMENT PROBLEMS

The axial alignment of a turbomachinery train including a double-helical gear unit should address all operating modes of the train. This requires consideration of both steady-state equilibrium thermal growth and transient thermal growth during startup, shutdown, or unloading cycles that may be encountered in field operation. Test conditions must also be considered, particularly if testing includes running at essentially no torque load. If properly applied at the design stage the following observations should help prevent future problems. They may also be helpful in correcting problems in existing trains.
DIAGNOSIS OF AXIAL ALIGNMENT PROBLEMS

A few main points relevant to diagnosing axial alignment problems are summarized here. Although some may not be practical in a particular situation, it would be prudent to consider these checks if the usual diagnostic process does not identify the problem. At a minimum, steps such as reviewing axial alignment and checking predicted torsional frequencies should be taken. If an axial alignment issue is found, the options described in the examples and in the following section should be helpful.
Coupling prestretch can help accommodate thermal growth in
Review
axial alignment calculations, installation results, and related factors. This includes thrust bearing clearances, gearset backlash, coupling stiffness, coupling prestretch, and distance between shafts. In particular, ensure coupling axial thermal growth is included in axial alignment calculations.
the train. However, care must be taken to ensure the flexible elements of the coupling will not be damaged. There may be axial position excursions in both directions during startup thermal transients. A compromise coupling prestretch value may be required to accommodate both transient and steady-state axial alignment conditions.
Coupling thermal growth should be included in prestretch calculations, as well as thermal movements of the shaft ends and casings of connected machines.
Review predicted torsional natural frequencies for a match with

observed response frequencies. As seen in the examples, there is often a correlation between measured gearbox response frequencies and predicted torsional frequencies. Spectral data with good resolution taken during the vibration event are required.
Coupling Coupling
enclosures should be properly designed to prevent excessive coupling heating and thermal growth. flexible element axial stiffness can be lowered to accommodate more axial growth without generating large axial forces.
102
Thermal growth analysis of involved components must include
consideration of maximum and minimum transient thermal states, such as when rotor components are hot and casings are cool. This will allow a comprehensive axial alignment chart to be produced.
A thrust bearing can be situated in the gearbox on the shaft to
which the largest external force is applied. This can prevent the axial force from reaching the mesh. However, the thrust bearing must be capable of absorbing the axial force. It should be kept in mind that purchasing specifications often dictate gearbox thrust bearing location. For example, API 613, Fifth Edition (2003), states that a double-helical gearbox shall have a thrust bearing on the low-speed shaft unless specified otherwise by the purchaser.
bearing location, thrust bearing clearance, gearset backlash, and coupling prestretch can help prevent axial alignment problems. System designers, gear unit suppliers, and others such as coupling suppliers should work together to ensure that the equipment will operate satisfactorily. Ultimately, concerns about transient thermal growth and subsequent impact on operating alignment between gear elements are a system-level issue. A comprehensive analysis should be performed to ensure the design respects the constraints of each system component. Good communication and cooperation at the design stage are crucial, particularly if there are large thermal effects or other special circumstances.
The
gearbox thrust bearing clearance can be increased. The gearset can even be allowed to float axially without a thrust bearing to absorb motion of connected equipment. However, this option may not be possible for some train layouts. Also, there can still be a problem if the connected machines can exert axial force on both gearbox shafts.
APPENDIX A SUPPLEMENT TO EXAMPLE 1

Discussion of Physical Events During Diagnostic Run with Axial Probe on Pinion The following paragraphs provide a physical interpretation of the problem deduced from position measurements, vibration data, and mechanics of the machine components. Refer to the section titled Diagnostic Run with Axial Probe on Pinion. Figure A-1 depicts the location of the axial probe on the pinion blind end, access panel.
Increased backlash can accommodate more mesh axial offset,
but excessive backlash can reduce the load capacity of the gearing by excessively thinning the teeth. It can also increase the severity of transient torsional vibration in some trains.
Casing fixation locations of connected machines can be altered

to reduce shaft end excursions.
Changes
of startup sequences or other thermal management schemes can be considered.
SUMMARY
A double-helical gearbox relies on transmitted torque to generate mesh axial forces that hold the pinion centered on the gear. If an external axial force is larger than the mesh force, it can push the pinion and gear off center with one another. This is most likely to occur in unloaded trains such as a generator drive prior to application of electrical load or a compressor drive in a no-load string test. The external force is most commonly caused by axial misalignment of the connected equipment. This misalignment may be due to an error in design or installation, or it may be a temporary effect caused by transient thermal growth. Axial alignment errors and thermal transients in turbomachinery trains can produce axial displacements, which must be absorbed by the coupling flexible elements. If the axial displacements are large, and the flexible element stiffness is high, the resulting forces and axial motion of the pinion relative to the gear can be sufficient to consume all the axial clearance in the gearset and bind the mesh, essentially producing a gearset running without backlash. Four field examples have been presented where axially bound gearsets have excited subsynchronous torsional resonances. The mechanical coupling of torsional and lateral motion in a gearbox results in the torsional vibration producing a lateral response. Depending on the stiffness of the bearings, varying levels of lateral response are measured, but as shown in the examples these vibrations can be sufficient to trip the machinery protection system. Several diagnostic checks are available to help distinguish axial alignment problems from other sources of vibration. They range from the relatively easy tasks of reviewing axial alignment and predicted torsional frequencies to more difficult and expensive checks such as special test runs. The diagnostic value of an axial probe on the pinion has been well demonstrated. In Examples 1, 2, and 4, this probe allowed direct verification of the phenomena. The verification allowed the involved parties to agree on a solution that would allow successful completion of the testing or installation of the equipment without compromising the operational performance of the machinery. Fortunately, options are available at the design stage to mitigate the effects of axial thermal growth. Careful axial alignment analysis and judicious use of design parameters such as thrust
Figure A-1. Photograph of the Axial Probe Installation in Example 1. During axial alignment of the train, the bull gear was centered in its thrust bearing clearance and the pinion was centered relative to the gear. At that point all components were at ambient temperature and there was a small prestretch in the coupling flex elements [nominally 0.018 inch (0.5 mm)]. The train was started and idled as usual as the equipment began to heat up. At 8:42 the compressor speed was increased from 9000 to 11,844 rpm. This caused two things to happen. First, as speed increased, the torque absorbed by the compressor increased slightly and so did the mesh centering force. This caused the pinion to move axially, changing coupling displacement to maintain equilibrium between the mesh centering force and coupling compression force. This is shown in Figure 12 as a downward movement of the position trace of about 1 V, indicating the pinion moved about 0.005 inch (0.13 mm) toward the compressor (the probes provide 200 mV/mil displacement). Because higher torque (and therefore an increased mesh centering force) caused the pinion to move closer to the compressor, it can be inferred that thermal growth prior to 8:42 had pushed the pinion away from the compressor and moved it off center relative to the gear. Figure 13 shows that the low-speed gear simultaneously moved about 0.0015 inch (0.038 mm) away from the compressor. Because the axial probe on the low-speed gear was oriented in the opposite direction, downward motion of the gear position voltage indicates motion away from the compressor. The second occurrence at 8:42 was the temperature of the compressor rotor and case began rising substantially again. As the
103
compressor came to MCOS, the various parts of the high-speed shaft line began to quickly heat and grow axially. The thermally massive case did not heat up as fast, and took longer to grow axially. The net result was that the compressor shaft end moved axially toward the gearbox. The thermal differential growth of the high-speed rotor string from the mesh center to the compressor thrust bearing was 0.066 inch (1.7 mm) (0.066 inch = 0.042 inch compressor shaft + 0.015 inch coupling + 0.009 pinion shaft). Furthermore, the low-speed gear was undergoing a similar thermal transient that was pushing the gear away from the gas turbine and against the gear thrust bearing. This also served to compress the high-speed coupling and increase the axial load on the pinion. Normally (i.e., when under load), the torque transferred and the double-helix centering force would have kept the two gear elements in axial alignment, and the transient axial growth would have been absorbed by the flexible elements of the coupling. But that did not happen in this test since torque load was near zero. Recalling the discussion related to Figures 5 and 6, under lightload conditions, the maximum mesh axial centering force was much less than at normal operating torque, approximately 100 lbf (445 N) in this instance. Based on the axial stiffness of the coupling flex elements, K = 4000 lbf/in (700 kN/m), only 0.025 inch (0.6 mm) of displacement was needed to produce in excess of 100 lbf (445 N). The calculated 0.066 inch (1.7 mm) thermal growth and the 0.018 inch (0.5 mm) prestretch of the coupling result in 0.048 inch (1.2 mm) of compression in the flex element, which should produce 192 lbf (855 N) axial load if the mesh remained centered. This was sufficient to slide the pinion axially along only one helix while the teeth of the opposite helix would not be making contact (as shown in Figure 5). From 8:42 to 8:57 the pinion position could be seen moving smoothly away from the compressor as shown by the steady rise in the pinion position voltage in Figure 12. During that time, speed and torque were constant and the pinion was moving axially along with the compressor shaft end. Because the pinion was moving axially relative to the gear, it was clearly running on one helix as shown in Figure 5. Note that there was practically no SSV at this time, indicating that the gearset can run smoothly on one helix. At 8:57:36 (where the data transition from light to dark), the moment significant SSV first appeared as shown in Figure 11, the pinion axial motion abruptly stopped as shown by the sharp discontinuity in Figure 12. This suggests that an obstacle had presented itself to restrict that motion. Based on distance traveled and the fact that there were no other obstructions to stop the pinion, it was clear that the pinion stopped moving axially because it reached a bound condition with the gear (as shown in Figure 6). For about an hour after 8:57:36 the data show the mesh was bound tightly. When the pinion stopped moving axially, the compressor thermal transient had not been completed and the compressor shaft end position had not reached its full excursion toward the gearbox. For the next hour the pinion axial position was very nearly constant even though the compressor shaft end would have still been moving, initially growing toward the gearbox, then pulling away as compressor case growth caught up with the rotor. Once the pinion reached the end of its travel in the mesh clearance, any additional compressor shaft growth had to be accommodated by increasing axial compression of the high-speed coupling. Therefore during the hour that the pinion did not move, it must have been forced tightly in the mesh. The SSV continued undiminished throughout this hour period. After one hour, corresponding to the time when the SSV began to change and the data transition from dark to light in Figures 11 to 13, the pinion began to return to its initial axial position. This change corresponds to the compressor casing warming and pulling the compressor shaft end away from the gearbox. At thermal equilibrium, the compressor case provides 0.030 mil (0.8 mm) of axial growth in the opposite direction from the initial shaft growth. Even though the compressor shaft had begun moving back away from the gearbox before the end of the hour, the pinion did not move
axially relative to the gear until the coupling compression decreased sufficiently for the coupling axial force to reach equilibrium with the axial mesh force. When this equilibrium was reached the pinion began moving with the compressor shaft end as it moved back toward the compressor. At this point the pinion was becoming unbound and returning to the condition shown in Figure 5. Shortly after the pinion began to move back toward the compressor, the discrete SSV components vanished, leaving the subsynchronous spectrum clean. When speed was decreased from 11,840 to 9000 rpm (at 11:00), the pinion position moved abruptly away from the compressor, momentarily returning back to the same axial position where it stopped moving axially during initial startup. This happened because the speed decrease reduced the torque transmitted across the gearset and the corresponding gear mesh centering force, allowing the remaining coupling axial force to push the pinion back to its bound position. During this period, SSV briefly reappeared and then subsided. It is important to note that numerous additional runs and stop cycles had produced many data sets for review. Careful observation of the shutdown behavior indicated that the vibration spikes (at 40 and 80 Hz) were not tracking with the 1 speed as the train decreased in speed. Instead, as train speed decreased, the 40 and 80 Hz components remained at a fixed frequency as shown in several of the preceding vibration plots. This behavior is not consistent with an oil whirl. However, it does appear consistent with an excitation of the first and second torsional frequencies.
REFERENCES
API Standard 613, 2003, Special-Purpose Gear Units for Petroleum, Chemical and Gas Industry Services, Fifth Edition, American Petroleum Institute, Washington, D.C. Carter, D. R., Garvey, M., and Corcoran, J. P., 1994, The Baffling and Temperature Prediction of Coupling Enclosures, Proceedings of the Twenty-Third Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 115-123. Drago, R. J., 1988, Fundamentals of Gear Design, Boston, Massachusetts: Butterworth-Heinemann. Dudley, D. W., 1984, Handbook of Practical Gear Design, New York, New York: McGraw-Hill. Dudleys Gear Handbook, Second Edition, 1991, Townsend, D. P., Editor, New York, New York: McGraw-Hill. Hudson, J. H., 1992, Lateral Vibration Created by Torsional Coupling of a Centrifugal Compressor System Driven by a Current Source Drive for a Variable Speed Induction Motor, Proceedings of the Twenty-First Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 113-123. Iwatsubo, T., Arii, S., and Kawai, R., 1984, The Coupled Lateral Torsional Vibration of a Geared Rotor System, Proceedings of the Third International Conference on Vibrations in Rotating Machinery, IMechE, C265, pp. 59-66. Mancuso, J., July 24, 1986, Disc vs Diaphragm Couplings, Machine Design, pp. 95-98. Nicholas, J. C., Barrett, L. E., and Leader, M. E., 1980, Experimental-Theoretical Comparison of Instability Onset Speeds for a Three Mass Rotor Supported by Step Journal Bearings, Journal of Mechanical Design, ASME, 102, pp. 344-351. Pennacchi, P. and Vania, A., 2004, Model-Based Analysis of Torsional and Transverse Vibrations of Geared Rotating Machines, Proceedings of the Eighth International Conference on Vibrations in Rotating Machinery, ImechE, C623, pp. 251-260.
104
Zirkelback, C., 1979, CouplingsA Users Point of View, Proceedings of the Eighth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 77-81.
ACKNOWLEDGEMENTS
The authors would like to thank the management of DresserRand, Lufkin, and Petronas Carigali for their support in publishing this paper. The authors also would like to acknowledge the support of Mr. Ed Turner (Dresser-Rand) and Mr. Craig Kujawa (Solar Turbines) in collecting data, Mr. Rob Kunselman (Dresser-Rand) for creating graphics, and Mr. Richard Pilsbury (Dresser-Rand) and Mr. Mark Winthrop (Lufkin) for assisting in the modifications required to perform the diagnostic testing.
BIBLIOGRAPHY
Eshleman, R. L., 1997, Torsional Vibration of Machine Systems, Proceedings of the Sixth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 13-22.
PERFORMANCE EVALUATION OF A CENTRIFUGAL COMPRESSOR OPERATING UNDER WET GAS CONDITIONS

by Lars Brenne
Staff Engineer
Tor Bjrge
Staff Engineer Statoil ASA Trondheim, Norway
Jos L. Gilarranz
Senior Aero/Thermodynamics Engineer
Jay M. Koch
Staff Engineer, Aero/Thermodynamics
and Harry Miller

Product Manager, Marketing Dresser-Rand Company Olean, New York
Lars Brenne is currently a Staff Engineer at the R&D Department (Rotating Equipment) of Statoil ASA, in Trondheim, Norway. He has been involved in wet gas compression technology studies, tests, and development. Dr. Brenne began his career with Aker Kvrner where he worked in the Mechanical Engineering Department (Rotating Equipment). His primary responsibility was pump systems for the Jotun FPSO. Upon completion of his graduate studies in 2002, he joined Statoils Research Department. Dr. Brenne received his M.S. degree (Mechanical Engineering, 1997) from the Norwegian University of Science and Technology, and his Ph.D. degree (Thermal Energy, 2004) from the Norwegian University of Science and Technology. Tor Bjrge is currently a Staff Engineer at the R&D Department (Rotating Equipment) of Statoil ASA, in Trondheim, Norway. He has been involved in activities covering wet gas compression, compressor transient response, and NOx emissions from gas turbines. Dr. Bjrge received his M.S. degree (Mechanical Engineering, 1981) from the Norwegian University of Science and Technology. Upon graduation, he joined the Norwegian University of Science and Technology where he worked at the Engineering Thermodynamics Department. He received his Ph.D. degree (Engineering Thermodynamics, 1988) from the Norwegian University of Science and Technology. Upon completion of his graduate studies, Dr. Bjrge worked as an Associate Professor at the Engineering Thermodynamics Department. His primary responsibility was as a lecturer within Thermodynamics, Heat and Mass transfer, and within research in
111
the same area. Dr. Bjrge then joined Statoils Research Department. He is a member of ASME. Jos L. Gilarranz is currently a Senior Aero/Thermodynamics Engineer with Dresser-Rand Company, in Olean, New York. He has been heavily involved in the design, specification, and use of advanced instrumentation for development testing of new centrifugal compressor components. He began his career with Lagoven (now PDVSA) and worked for three and a half years as a Rotating Equipment Engineer in PDVSAs Western Division. His primary responsibility was the evaluation and prediction of the aerothermal performance of multistage centrifugal compressor packages utilized by Lagoven in Lake Maracaibo. Upon completion of his graduate studies, he joined Dresser-Rands Development Engineering Group. Dr. Gilarranz received his B.S. degree (Mechanical Engineering, 1993) from the Universidad Simn Bolvar (Venezuela). He received his M.S. degree (Aerospace Engineering, 1998) and his Ph.D. degree (Aerospace Engineering, 2001) from Texas A&M University. He is a member of ASME and AIAA.
ABSTRACT
This paper presents the results of performance testing of a single-stage centrifugal compressor operating under wet gas conditions. The test was performed at an oil and gas operators test facility and was executed at full-load and full-pressure conditions using a mixture of hydrocarbon gas and hydrocarbon condensate. The effect of liquid was investigated by changing the gas-volume fraction between 1.0 and 0.97, which covers the range encountered by the operator during regular gas/condensate field production in the North Sea. Other parameters that were evaluated include the
112
compressor test speed, the suction pressure, and two different liquid injection patterns. During the tests, the machine flowrate was varied from near surge to choke conditions; hence, the evaluation covered the entire operating range of the machine. Although the test was primarily intended to evaluate the effects of the wet gas on the thermodynamic performance of the machine, the mechanical performance was also investigated by measuring the machine vibration levels and noise signature during the baseline dry gas tests as well as during the tests with liquid injection.
INTRODUCTION
Centrifugal compressor packages utilized for upstream gas processing often must operate under wet gas conditions in which the fluid handled by the compression package contains a mixture of liquid and gaseous phases. Typically, the liquid components of the mixture are separated from the gas stream before they enter the compressor by the use of scrubbers and separators located upstream of the compressor inlet. These devices are very large and heavy, requiring a large footprint (amount of floor space) as compared to the gas compression package. A compressor with the ability to directly handle wet gas without the need for separation equipment is very attractive from an economic standpoint, as it would drastically reduce the size, weight, and cost of the gas compression package. For the case of future subsea compression systems, this capability is even more attractive because of the high costs of deploying a compressor train and all of its associated equipment under water. Wet gas compression (WGC) technology represents new opportunities for enhanced, cost-effective production from existing and future gas/condensate fields. Many oil and gas operators face future challenges in tail-end production, unmanned operation, and improved recovery from topside and subsea wells. This emphasizes the need to develop more robust compression systems, which can be designed for remote operation in unmanned topside installations, or could be designed for subsea operation for reinjection and/or transport boosting. The use of this technology for subsea boosting represents a new and exciting application for rotating equipment, which will allow new gas/condensate field production opportunities as well as enhanced recovery of existing gas/condensate fields and cost-effective production from marginal gas fields. As mentioned above, these wet gas compression systems could be based on the use of a liquid tolerant dry gas compressor, which could boost a coarsely separated (via a scrubber) well-stream, however, an even more attractive solution would be the development of compression systems that can boost the well-stream directly. Many research projects and product qualification programs are currently underway to develop such a system either by modifying existing multiphase pump technology or by the adaptation of currently available gas compression technologies (Scott, 2004). Regardless of the choice of concept, the compressor solution should be able to tolerate liquid ingestion for an extended time without failure. For the case of subsea applications, the high cost associated with the retrieval of the compressor from the sea floor accentuates the importance of a reliable design. The work presented herein served as an initial test to verify the multiphase boosting capabilities of a centrifugal compressor as well as to provide an oil and gas operator with data to compare the performance of this technology with other available wet gas compression concepts. It is important to state that the test compressor used for this investigation was not originally designed for wet gas boosting, nonetheless it provided an economically viable test bed for centrifugal compressor technology.
0.02380. The compressor was originally designed to handle an inlet flow of 4332 Kg/min [2167 Am3/hr (76,526.88 ft3/hr)] of dry hydrocarbon gas (molecular weight of 18.49), with an inlet pressure of 130.2 bar (1888.4 psi)and a discharge pressure of 161.8 bar (2346.7 psi). Figure 1 shows a cross-section of the test compressor; the inlet and discharge nozzles are located at a 45 degree angle with respect to the top dead center of the machine. The original design of this machine, which dates to 1986, was not intended for wet gas service, and hence the internal geometry was not optimal. Nevertheless, in order to increase the reliability of the machine, the original rotor design was modified to accommodate an electron-beam welded and vacuum furnace brazed impeller with a shrink fit to the shaft. The rest of the machine remained the same (i.e., casing and stationary components). This compressor was equipped with a vaneless diffuser configuration.
Figure 1. Cross-Section of the Test Compressor. The compressor was driven by a 2.8 MW synchronous electric motor, through a speed increasing gearbox, with a gear ratio of 6.607. A variable speed drive permitted the operation of the compressor within its speed range of 6000 to 13,000 rpm. The test compressor is utilized in the coauthors closed loop test facility, and was equipped to simulate the conditions expected for a centrifugal compressor operating under wet gas conditions. Figure 2 shows a schematic diagram of the test loop that was used for the evaluations. The major components of the test loop included a scrubber, the test compressor, a pump, a cooler, and a liquid injection module (mixer). The scrubber, here called guard separator, was used to separate the dry gas (saturated hydrocarbon mixture) from the liquid (hydrocarbon condensate) in order to permit accurate measurement of the massflow of each stream (liquid and gas). The liquid stream was measured with a Coriolis flowmeter while the gas stream was measured with a calibrated orifice plate.
Gas Flow
p, T, Pdyn, Q
p, T, Q Filter
Cooler/Heater
Condensate
Variable Speed Electric Motor (MW)
Gearbox
Liquid Injection Module
DESCRIPTION OF TEST VEHICLE

The test vehicle used for this work was a barrel-type, singlestage compressor, manufactured by the coauthors company. Said compressor was equipped with a high-head impeller, with a diameter of 0.384 m (1.26 ft), and a design flow coefficient of
Compressor (rpm)
p, T, Pdyn
2 Phase Flow
Guard Separator
Figure 2. Schematic Diagram of the Wet Gas Test Loop.
113
A screw pump was used to handle the hydrocarbon condensate exiting the guard separator in order to increase its pressure to adequate values for injection into the gas stream. The liquid injection module permitted the introduction of the condensate into the gas upstream of the compressor inlet. The liquid could be introduced into the gas with two different patterns: as a uniformly distributed droplet mist or as a liquid film uniformly coating the wetted surface of the inlet pipe. This was obtained by injecting the liquid through specially designed nozzles for the case of droplet flow or through a circumferential slit for the case of liquid film flow. This required pressurization of the liquid phase and the use of a cooler/heater to regulate the temperature of the condensate to assure that it was injected into the gas stream at the same temperature as the gas. The liquid temperature was measured using calibrated PT-100 elements and static pressure measurements were made using calibrated pressure transducers. The temperature and static pressure of the gas stream entering and exiting the compressor was measured using calibrated thermocouples and pressure transducers installed in the pipeline in accordance with the recommendations of ASME PTC 10 (1997). The measurements made on the gas exiting the compressor corresponded to that of the wet gas mixture, while the measurements of the liquid and dry gas streams at the inlet were made independently for each phase. The compressor discharge piping had a 45 degree slope upwards and also a diameter change from 0.203 to 0.305 m (.666 to 1.0 ft). Due to the risk of liquid accumulation in this piping, liquid hold up was monitored using a gamma ray densitometer. The same measurement was also performed upstream of the antisurge valve to detect any liquid accumulation. If this was detected, the test would have to be stopped due to the risk of injecting a liquid slug into the compressor. The gas composition of the hydrocarbon mixtures utilized as test gas and liquid are shown in Table 1. The gas corresponds to an export quality lean hydrocarbon mixture (composed mostly of methane), which is typically commercialized for the European market, while the liquid corresponds to the condensate received from the Sleipner field, which lies in the Norwegian North Sea. Table 1. Gas and Liquid Compositions (Compositions Shown as Molar Percentages).
Component Export gas Condensate Composition in loop at 70 bar and 35 oC Gas phase Liquid phase Total
the inlet and discharge of the compressor were obtained and displayed online for each one of the measurement series and stored together with all of the measured and calculated parameters in one data file. In addition to the instrumentation described above, the test loop was also equipped with dynamic pressure transducers, installed at the inlet and discharge piping of the compressor. These transducers were utilized to measure the fluctuating pressure components at the inlet and discharge of the machine. The signals from these instruments were displayed and recorded during the test in the form of frequency spectra, which enabled the test engineers to monitor the pressure signals in the process loop directly upstream and downstream of the compressor. These measurements were correlated to the noise level of the machine and permitted the comparison of this parameter while the machine operated under dry and wet gas conditions. In addition, the probes could be used to assist in correlating any possible subsynchronous rotor vibrations with pulsations in the gas stream (Marshall and Sorokes, 2000). In order to minimize the complexity of the instrumentation for the initial test, and since the primary mission was the study of the thermodynamic performance change with liquids introduced to the gas stream, it was decided to forego installation of additional instrumentation, which would have given more insight into the mechanical reactions taking place in response to the various liquid loadings. As such, the installation of strain gauges on the impellers with their attendant installation complexities, as well as converting to active magnetic journal and thrust bearings, were held off for future test programs. The installation of a magnetic bearing shaft exciter (Moore, et al., 2002) onto the free-end of the compressor shaft would have provided a means to assess any variation of the rotor natural frequencies, as well as to determine any change to the rotordynamic stability of the compressor due to the addition of liquids into the gas stream. The use of this device was also left to a future test program.
THEORETICAL FOUNDATION
Single-Phase (Dry Gas) Performance For a centrifugal compressor the primary variables of interest are the amount of flow delivered, the pressure rise produced, and the required power. The pressure rise and the efficiency of the gas compression are normally nondimensionalized to allow comparison of different geometries and operating conditions (Stahley, 2000; Colby, 2004). The polytropic compression process is selected for industrial compressors as it is better suited to handle the wide range of gases used in industry (Schultz, 1962). The equations for polytropic head coefficient, flow coefficient, efficiency, and power are shown below (ASME PTC-10, 1997).
N2 CO2 C1 C2 C3 I-C4 N-C4 I-C5 N-C5 C6 C7 C8 C9 C10+ Mole weight
0.756 1.828 90.373 6.074 0.844 0.045 0.064 0.006 0.006 0.004
17.77
0.024 1.059 7.690 10.373 12.015 20.387 18.616 9.487 4.392 15.957 98.222
0.854 1.524 90.933 4.103 0.341 0.124 0.654 0.481 0.454 0.348 0.137 0.035 0.008 0.005 18.483
0.089 0.956 25.920 4.489 0.955 0.651 4.632 6.662 7.856 13.897 12.932 6.637 3.084 11.239 73.52
0.531 1.284 63.474 4.266 0.600 0.347 2.334 3.091 3.581 6.071 5.541 2.824 1.307 4.750 41.728
p =
U = Wp =
Wp U2
60
(1)
D N
(2)
Based on the volumes of gas and liquid and the filling temperature and pressure of the test loop, the composition of the gas and liquid streams and their associated thermodynamic states were evaluated using a thermodynamic property package in combination with the measured pressure and temperature. The thermodynamic property package is a precursor of a commercially available gas property package, which allows the combination of reliable fluid characterization procedures with robust and efficient algorithms to match fluid descriptions to experimental pressure, volume, and temperature (PVT) data. To increase the data accuracy, the gas and liquid densities were determined with a commercially available thermodynamic calculation software. The thermodynamic data at
n ( p2 2 p1 1 ) n1
(3)
n=
1n
p2 p1
1 1n
2
(4)
N 2 60 D3
Q 1
(5)
114
p =
Wp h2 h1
(6)
TP
= GVF1
g1 Wp TP TP1 U 2
(14)
( h2 h1 ) P=m
(7)
This formulation assumes a single-phase gas. If the compressor inlet stream contains both gas and liquid (i.e., wet gas), these equations must be modified. The primary area of interest is the definition of polytropic work, which impacts the polytropic efficiency and polytropic pressure coefficient. Two-Phase (Wet Gas) Performance The calculation procedure to estimate the performance of a machine operating under wet gas conditions is not described in any standard, as is the case for dry gas. However, the thermodynamic approach used for a single-phase gas, as stated above, can still be applied to a two-phase fluid. For the case of the single fluid model, the required modifications are shown below:
TP =
N GVF1 2 60 D3
Q Tot 1
(15)
The efficiency for both cases is then expressed as:
TP =
Wp
TP
Pcs
(16)
nTP =
1n
p2 p1
TP 2
TP1 1n
(8)
Wp
TP =
nTP ( p2 TP 2 p1 TP1 ) nTP 1
where Pcs is the specific compressor shaft power, defined as the power consumed by the compressor per unit mass of wet gas. The compressor two-phase efficiency calculated with the use of the single fluid model was found to be virtually equal to the one calculated via the two-phase fluid model (with a maximum deviation of 0.8 percent), so the results described in this work will be based on the single fluid model. The performance of a wet gas compressor must be compared to the alternative, which would involve the separation of the fluid stream into individual phases (dry gas and condensate), and the subsequent boosting of the streams in separate compressor and pump units.
(9)
TEST PURPOSE AND VARIABLES

The wet gas testing presented in this work had several objectives. The first objective was to investigate the heat transfer rate between the gas and liquid condensate through the compressor and determine the state of thermal equilibrium. Another objective was to evaluate the compressor performance (power consumption, pressure ratio, and temperature ratio) and determine the effects of directly handling a wet gas mixture as opposed to dry gas compression. The impact of liquid ingestion on the compressor mechanical behavior and the pressure pulsations in the loop was also of interest, as was the liquid tolerance capacity and robustness of the compressor. Finally, the testing would create a foundation to evaluate the benefits and/or drawbacks of centrifugal compression technology as opposed to other multiphase boosting concepts. To achieve the test goals, the performance of the machine was evaluated under several combinations of key parameters such as suction pressure, flowrate, rotational speed, gas-volume fraction, and liquid injection pattern following the data presented in Tables 2 and 3. The test program was completed in a time frame of about four weeks, during which the machine accumulated about 300 hours of operation under wet gas conditions. Table 2. Key Test Parameters with Range of Variation.
Key Test Parameter Values 30 70 9651 10723 1600 1800 2000 2200 2400 1.0000 0.9994 0.9950 0.9900 0.9800 0.9700 Uniform Droplet Fluid Film Units bar rpm
where the two-phase specific volume is based on the homogeneous model:
TP =
1 GVF g + (1 GVF ) 1
(10)
The gas-volume fraction (GVF) is defined by:
Q g GVF = Qg + Q 1
(11)
The fluid power was derived from electric power readings, using adequate calibration curves available from previous testing. A different approach would be to consider a two fluid model where each phase is treated individually. The polytropic head is then calculated as:
n 1 n n R0 p2 Wp TP = x1 Z1T1 1 + (1 x1 ) l1 ( p2 p1 ) (12) n 1 MWg1 p1
where the fluid quality is defined as:
x=
g m g + m l m
(13)
Suction Pressure Machine Speed
For the case presented in this work, the phase transition component was small due to a low pressure ratio through the machine and stable fluids. However, for higher pressure ratio multistage compressors, the phase transition contribution cannot be neglected. For the case at hand however, the effect of phase transition is only accounted for in the above expressions by changes in the value of the polytropic exponent due to a lower discharge temperature. For this work and both of the performance calculation models presented above, the two-phase head coefficient and two-phase flow coefficient may be expressed as:
Gas Volumetric Flowrate
Am3/hr
Gas-volume Fraction
N/A
Liquid Injection Pattern
N/A
115
1.20
Table 3. Test Matrix Agenda.

Normalized Polytropic Head Coefficient
10273 RPM
Test Number 1 2 3 4
Suction Pressure 30 70 70 70
Machine Speed 9651 9651 10723 9651
Gas Flowrate All All All All
Gas-Volume Fraction
Liquid Injection Pattern
0.80
1.20 1.00 0.80 0.60 0.40 0.20 0.00
0.60 0.40
1.0 N/A 1.0 N/A 1.0 N/A All Droplet 1.0, 0.995, 0.98 5 70 9651 All Fluid Film 2200 Am3/hr :All 1.0, 0.995, 0.98 6 70 10723 All Droplet 3 2200 Am /hr :All 7 30 9651 All All (*) Droplet 8 30 9651 All 1.0 N/A (*) The test point at 2400 Am3/hr and GVF = 0.97 was not completed due to test loop limitations.
9651 RPM
Head Coefficient
0.20 0.00
Head Coefficient - Predicted Head Coefficient - After Test Efficiency - Baseline
Head Coefficient - Baseline Efficiency - Predicted Efficiency - After Test
0.60
0.70
0.80
0.90
1.00
1.10
1.20
1.30
1.40
1.50
1.60
The quality (x) of the wet gas mixture being injected into the compressor was dependent on the predefined gas-volume fraction as well as the suction pressure at which the test was being executed. Table 4 presents the values of quality for each GVF used for the tests for suction pressures of 30 and 70 bar (435.1 and 1015.3 psi). Table 4. Quality of the Wet Gas at the Compressor Inlet.
Normalized Flow Coefficient
Figure 3. Single-Phase Thermodynamic Performance Evaluation. Baseline Versus After-Test Conditions. (435.1 and 1015.3 psi)], while operating at the same speed (9651 rpm) and with the same liquid injection pattern (uniform droplet). Figure 6 shows a comparison of the machine performance for both liquid injection patterns (droplet and fluid film), with the machine operating at the same suction pressure [70 bar (1015.3 psi)] and the same speed (9651 rpm).
10723 RPM
GVF 1.0000 0.9994 0.9950 0.9900 0.9800 0.9700
Quality at 70 bar 1.0000 0.9931 0.9454 0.8959 0.8100 0.7377
Quality at 30 bar 1.0000 0.9824 0.8699 0.7706 0.6254 0.5218
1.20 1.00 0.80
1.20 1.00 0.80 0.60 0.40 0.20 0.00 0.7

0.8 0.9 1.0 1.1 1.2 1.3
9651 RPM
0.60 0.40 0.20 0.00
RESULTS AND DISCUSSION

Single-Phase (Dry Gas) Performance Prior to the introduction of liquids into the test loop, the thermodynamic performance of the machine was evaluated to establish the baseline performance while it was operating under single-phase (dry gas) conditions (tests 1, 2, and 3 of Table 3). This baseline would be used for comparison with the results obtained during the operation of the machine under wet gas conditions. In addition, the baseline performance would be used to evaluate if the injection of liquids during the multiphase testing had produced any noticeable effects (performance changes) after the tests were concluded. This would be done by running another series of dry gas performance tests and comparing the results to the initial baseline. Figure 3 shows the results of the single-phase performance tests that were executed before and after the evaluation with two-phase (wet gas) flow. As seen in the figure, the performance levels of the machine (polytropic head coefficient and efficiency) have remained unchanged, that is, there is no evidence to suggest that the ingestion of liquid produced any significant variation in the machines performance levels. This implies that the compressor flowpath was not subjected to any significant damage during the wet gas tests. A boroscopic inspection of the inlet and impeller eye areas executed after the tests confirmed that there was no evidence of internal damage. Two-Phase (Wet Gas) Performance Figures 4, 5, and 6 show the performance of the compressor exposed to liquid with up to 3 percent of the inlet volume flow (GVF of 0.97). The wet gas tests are shown together with dry gas results for comparison. Figure 4 presents a comparison of the compressor performance at two different speeds, while operating at a suction pressure of 70 bar, with the liquid being injected with a uniform droplet pattern. Figure 5 shows the performance of the compressor at two different suction pressures [p1 = 30 and 70 bar
Normalized Two-Phase Flow Coefficient

9651 RPM 10723 RPM
GVF_1.00 GVF_1.00 GVF_0.995 GVF_0.995 GVF_0.99 GVF_0.99 GVF_0.98 GVF_0.98 GVF_0.97 GVF_0.97
Figure 4. Two-Phase Thermodynamic Performance Evaluation. Effects of Test Speed (p1 = 70 bar, Droplet Injection Pattern).
1.20 1.00 0.80 1.20 1.00 0.80 0.60 0.40 0.20 0.00 0.7
0.8 0.9 1.0 1.1 1.2 1.3
0.60 0.40 0.20 0.00

p1 = 70 bar p1 = 30 bar
Figure 5. Two-Phase Thermodynamic Performance Evaluation. Effects of Suction Pressure (9651 rpm, Droplet Injection Pattern).
Normalized Polytropic Efficiency
Efficiency
1.00
116

1.20 1.00 0.80 1.20 1.00 0.80 0.60 0.40 0.20 0.00 0.7
0.8 0.9 1.0 1.1 1.2 1.3
0.60 0.40 0.20 0.00
these data have to be compared with the data that would be obtained if the same amount of liquid and gas were to be transported between the same two pressures (as independent streams). A separate gas and liquid boosting case is included in the figure assuming a GVF of 0.97. As can be seen the specific power consumption is lower than the values obtained for wet gas compression. The separate boosting data were based on the same pressure difference. However, when wet gas compression is utilized, the pressure drop in the scrubber may be avoided and the required pressure boost is less than the one depicted in Figure 7. Furthermore, the possibility of simplifying the compressor system by avoiding the scrubber and the appurtenant instrumentation must also be considered when the system is evaluated.
1.10 1.05

Droplet Film
1.00 0.95 0.90 0.85 0.80
Figure 6. Two-Phase Thermodynamic Performance Evaluation. Effects of Liquid Injection Pattern (9651 rpm, p1 = 70 bar). As seen in Figures 4 through 6, the redefined polytropic head and flow coefficients [Equations (14) and (15)], valid for twophase flow, are capable of merging the data from the various wet gas operating conditions with those corresponding to the dry gas operation. Furthermore, these figures show that the efficiency drops when the amount of liquid is increased and that this effect is much more pronounced at lower pressures. The more pronounced effect at lower pressures is due to the increasing density difference between the gas and the condensate when the suction pressure is reduced while maintaining a constant GVF. The increasing density difference leads to a considerable increase in the mass fraction of liquid entering the compressor. As shown in Table 4, at a GVF of 0.97, the mass fraction of liquid (1-x) increases from 0.2623 at 70 bar (1015.3 psi) to 0.4782 at 30 bar (435.1 psi). The reduction in the machine efficiency as the mass fraction of liquid increased is due to larger internal losses in the compressor. The test vehicle was not instrumented internally, so the available data were insufficient to identify the source of the increased losses. The compressor manufacturer plans to evaluate this issue by performing two-phase computational fluid dynamics (CFD) simulations of the compressor at the test conditions. Another way to provide insight into this phenomenon would be to run additional tests with internal instrumentation. Figure 6 shows very little difference between the performance of the machine when subjected to uniform droplet and fluid film injection patterns. Consequently, it is thought that the compressor inlet serves as a mixing element and makes the flow pattern inside the impeller relatively independent of the injection method. This effect will also be evaluated via two-phase CFD calculations. The distance between the point of liquid injection and the center of the impeller was limited to approximately three times the internal diameter of the compressor suction nozzle. This was done in an effort to ensure that the two-phase flow pattern was maintained from the point of injection up to the impeller inlet. In general, for the figures discussed above, there is a tendency of a larger departure from a common head coefficient curve at 30 bar (435.1 psi) when the deviation between the operating flowrate and the impeller design flowrate increases (GVF < 0.99). The compressor specific power consumption is shown in Figure 7. The data shown in the figure correspond to the tests with the compressor operating at 9651 rpm, with a suction pressure of 70 bar (1015.3 psi), and the liquid being injected in a uniform droplet pattern; nevertheless, the same behavior was observed for the other wet gas test conditions. As shown in the figure, when liquid is injected, the required specific power is reduced. To properly evaluate the specific power associated to wet gas compression,
0.75 0.70 0.65 0.60 0.8 0.9 1.0 1.1 1.2 1.3

GVF_1.00 GVF_0.995 GVF_0.99 GVF_0.98 GVF_0.97
Separate Boosting, GVF = 0.97
Figure 7. Two-Phase Thermodynamic Performance Evaluation. Compressor Specific Power Consumption as a Function of GVF (p1 = 70 bar, 9651 rpm, Droplet Injection Pattern). Figures 8 and 9 show the effects of liquid injection on the pressure and the temperature ratios (discharge/inlet) through the compressor. Once again, the data shown in the figures correspond to the tests with the compressor operating at 9651 rpm, with a suction pressure of 70 bar (1015.3 psi), and the liquid being injected in a uniform droplet pattern; nevertheless, the same behavior was observed for the other wet gas test conditions. The pressure ratio increased due to the increased density (and molecular weight) of the fluid processed by the compressor. The temperature ratio slightly decreased for the case of liquid injection. This observation is explained by two mechanisms:
Increased internal energy of the liquid phase, and A certain degree of liquid evaporation has occurred as the liquid
passed through the machine. These results will also be evaluated by the compressor manufacturer via two-phase analytical simulations, as they will be of great importance when designing multistage machines for wet gas operation. The change of phase of the liquids inside the compressor may cause a mismatch between the subsequent stages of the machine, which may lead to performance shortfalls. Dynamic Pressure Measurement Figures 10 through 15 show the frequency spectrum of the dynamic pressure signals measured close to the inlet and discharge flanges of the machine under three different test conditions (note that the scales on all plots are the same). Figures 10 and 11 correspond to the machine operating at 9651 rpm with a suction pressure of 70 bar (1015.3 psi) and show the effects of the liquid being injected in a uniform droplet pattern. Figures 12 and 13 correspond to the same compressor operating condition but with the

1.03 1.02 1.01 1.00 0.99 0.98 0.97 0.96 0.8 0.9 1.0 1.1 1.2 1.3
117
TP-70-D-9651-2200 GVF = 1 and 0.97; X = 1 and 0.738

Droplet Injection, p1 = 70 bar,
Flowrate = 2200 Am3/hr Signal Amplitude 1X Frequency = 161 Hz Blade-Pass Frequency = 3056 Hz
0
Blade-pass frequency
GVF = 0.97
GVF = 1.00
0
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000

Frequency [Hz]
Figure 8. Two-Phase Thermodynamic Performance Evaluation. Compressor Pressure Ratio as a Function of GVF (p1 = 70 bar, 9651 rpm, Droplet Injection Pattern).
1.004
Figure 11. Effects of Two-Phase Flow over the Dynamic Pressure Measurements at the Machine Discharge (p1 = 70 bar, 9651 rpm, Droplet Injection Pattern).
TP-70-F-9651-2200 GVF = 1 and 0.98, X = 1 and 0.810
Film Injection, p1 = 70 bar,
Flowrate = 2200 Am3/hr Signal Amplitude 1X Frequency = 161 Hz Blade-Pass Frequency = 3056 Hz GVF = 0.97 Blade-pass frequency
1.002 1.000 0.998 0.996 0.994 0.992
GVF = 1.00
0
0.990 0.988 0.8 0.9 1.0 1.1 1.2 1.3
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
Frequency [Hz]
Figure 12. Effects of Two-Phase Flow over the Dynamic Pressure Measurements at the Machine Inlet (p1 = 70 bar, 9651 rpm, Film Injection Pattern).
TP-70-F-9651-2200 GVF = 1 and 0.98, X = 1 and 0.810
Film Injection, p1 = 70 bar,
Figure 9. Two-Phase Thermodynamic Performance Evaluation. Compressor Temperature Ratio as a Function of GVF (p1 = 70 bar, 9651 rpm, Droplet Injection Pattern). liquid injection being performed as a fluid film in the periphery of the pipe. Finally, Figures 14 and 15 correspond to the same compressor speed, with a suction pressure of 30 bar (435.1 psi) and the liquid being injected in a uniform droplet pattern.
TP-70-D-9651-2200 GVF = 1 and 0.97; X = 1 and 0.738
Signal Amplitude
Flowrate = 2200 Am3/hr 1X Frequency = 161 Hz Blade-Pass Frequency = 3056 Hz
GVF = 0.97
0
0
GVF = 1.00
0
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000
GVF = 0.97
Frequency [Hz]
Figure 13. Effects of Two-Phase Flow over the Dynamic Pressure Measurements at the Machine Discharge (p1 = 70 bar, 9651 rpm, Film Injection Pattern).
GVF = 1.00
0
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Frequency [Hz]
Figure 10. Effects of Two-Phase Flow over the Dynamic Pressure Measurements at the Machine Inlet (p1 = 70 bar, 9651 rpm, Droplet Injection Pattern).
It is important to state that the low frequency amplitudes (noise) that are evident on the spectrum plots are due to the fact that the data acquisition and display system that was used to capture the data, and to generate these figures, did not have the capability to average several fast Fourier transform (FFT) samples. This hindered the ability to reduce the random noise components of the spectra. Furthermore, the pressure sensors were not installed flush to the pipe wall. They had a small recess [about 25 mm (.98
118

TP-30-D-9651-2200 GVF = 1 and 0.97, X = 1 and 0.53
Flowrate = 2200 Am3/hr Blade-pass frequency
Signal Amplitude
1X Frequency = 161 Hz Blade-Pass Frequency = 3056 Hz GVF = 0.97

0
GVF = 1.00
0
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Frequency [Hz]
Figure 14. Effects of Two-Phase Flow over the Dynamic Pressure Measurements at the Machine Inlet (p1 = 30 bar, 9651 rpm, Droplet Injection Pattern).
TP-30-D-9651-2200 GVF = 1 and 0.97, X = 1 and 0.53
GVF = 0.97
GVF = 1.00
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
Frequency [Hz]
Figure 15. Effects of Two-Phase Flow over the Dynamic Pressure Measurements at the Machine Discharge (p1 = 30 bar, 9651 rpm, Droplet Injection Pattern). inches)] that may have contributed to the generation of some noise components in the FFTs. As may be seen in the figures, for the case in which the compressor is handling dry gas (GVF of 1.0), the frequency spectrum plots show a variety of peaks in the vicinity of the impeller blade passing frequency. On the other hand, when liquids are injected into the flow (GVF = 0.98 or 0.97), the high-frequency components of the spectrum vanish. This behavior suggests that the liquids injected into the process gas dissipate the acoustic signals (evident by a reduction in the audible noise level) and dampen the pressure fluctuations. This phenomenon was observed for all of the test conditions that were evaluated as shown in the test matrix above (refer to Tables 2 and 3). Rotordynamic Behavior Although the main objective of the testing described in this paper was to establish the effects of wet gas conditions over the aero/thermodynamic performance of a centrifugal compressor, another objective of similar importance was to evaluate the effects of liquid ingestion over the rotordynamic behavior of the machine. For this, the shaft vibration was measured via eddy current proximity probes installed at both ends of the machine (i.e., at the driven and nondriven ends). The test compressor was originally supplied with a pair of proximity probes at each journal bearing to measure the shaft vibration in the horizontal and vertical directions. In addition, the axial position of the shaft was also measured via proximity probes at the free end of the machine. The signals from these probes were displayed in the control room and were
linked to the process control system to provide appropriate machinery protection. Said signals were also connected to a proprietary inhouse data acquisition and reduction system, which permitted the frequency spectrum of the shaft vibration signals to be displayed during the tests. These frequency spectra were utilized in the evaluation of the rotordynamic behavior of the machine by comparison of the spectra obtained under dry and wet gas conditions. The test compressor had a bearing span of 0.727 m (2.39 ft), with an impeller bore of 0.132 m (5.2 inches), and a journal diameter of 0.076 m (2.99 inches). The shaft was mounted on tiltpad journal bearings, each of which had five pads, in the load-on pad configuration. The first natural frequency of the rotorbearing system is near 9800 cpm. This mode is well damped and is not a critical speed. The first critical speed of the compressor was determined (analytically) to be between 20,300 to 21,130 cpm. This value is larger than the maximum speed at which the compressor would be tested (10,723 rpm), so smooth operation was expected within the operating envelope that would be used for the tests (9651 to 10,723 rpm). Figure 16 presents the frequency spectra (FFT) of the shaft horizontal vibration component, measured at the driven end of the machine, while it was operating at 9651 rpm, with a suction pressure of 70 bar (1015.3 psi) and handling a volumetric flow of 2200 Am3/hr (77,692.27 ft3/hr). The bottom spectrum corresponds to operation with a GVF of 1.0 (dry gas); while the top spectrum shows the behavior of the machine while handling a two-phase gas mixture, with a gas-volume fraction of 0.97, which represents a gas quality (x) of 0.738. For this case, the liquid was being injected into the gas with a uniform droplet pattern. As may be seen in the figure, the vibration spectra for both the dry gas and the wet gas compression are virtually the same. This suggests that the rotordynamics of the machine remain unaffected by the liquid injection for the case of the condensate being injected with the uniform droplet pattern. This is due to the fact that if the liquids are uniformly distributed throughout the gas, they do not produce any significant source of rotor excitation as they pass through the impeller, nor do they affect the rotor unbalance levels (1 vibration component). The above similarity was also encountered when comparing the vibration spectra corresponding to GVF values of 0.98 and 0.99. The behavior of the machine while it was operating at 10,723 rpm, under the same suction pressure and liquid injection mechanism, presented similar characteristics and hence will not be shown.
TP-70-D-9651-2200 GVF = 1 and 0.97; X = 1 and 0.738
Droplet Injection, p1 = 70 bar, Flowrate = 2200 Am3/hr 1X frequency (161 Hz) GVF = 0.97
Signal Amplitude
G VF = 1.00
0
100
200
300
400
500
600
700
800
900
1000
Frequency [Hz]
Figure 16. Effects of Two-Phase Flow over the Vibration Response of the Machine, Measured in the Horizontal Direction at the Driven End (p1 = 70 bar, 9651 rpm, Droplet Injection Pattern). Figure 17 presents the frequency spectra of the shaft horizontal vibration component, measured at the driven end of the machine, while it was operating at 9651 rpm, with a suction pressure of 70
119
bar (1015.3 psi) and handling a volumetric flow of 2200 Am3/hr (77,692.27 ft3/hr). The bottom spectrum corresponds to operation with a single-phase (dry) gas, while the top spectrum represents the behavior of the machine while handling a two-phase gas mixture, with a gas-volume fraction of 0.98 (gas quality of 0.810). For this case, the condensate was being injected into the gas with a liquid film pattern, which was uniformly distributed around the wetted surface of the inlet pipe. As may be seen in the figure, the vibration spectra for the dry and the wet gas cases are also very similar. This supports the belief that when the liquids are ingested in a uniform manner by the compressor, they do not provide a sufficiently strong source of excitation or unbalance to disturb the rotordynamic behavior of the machine. The above similarity was also encountered when comparing the vibration spectra corresponding to GVF values of 0.97 and 0.99.
TP-70-F-9651-2200 GVF = 1 and 0.98, X = 1 and 0.810
Film Injection, p1 = 70 bar, Flowrate = 2200 Am3 /hr
1X frequency (161 Hz) GVF = 0.98
TP-30-D-9651-2200 GVF = 1 and 0.97, X = 1 and 0.53

Droplet Injection, p1 = 30 bar, Flowrate = 2200 Am3/hr
Signal Amplitude
1X frequency (161 Hz)
GVF = 0.97
0
GVF = 1.00
0
100
200
300
400
500
600
700
800
900
1000
Frequency [Hz]
Signal Amplitude
Figure 18. Effects of Two-Phase Flow over the Vibration Response of the Machine, Measured in the Horizontal Direction at the Driven End (p1 = 30 bar, 9651 rpm, Uniform Droplet Injection Pattern).
SUMMARY AND CONCLUSIONS

This paper presented the results of tests performed on a centrifugal compressor operating under wet gas conditions, showing the effects of suction pressure, gas-volume fraction, machine speed, and liquid injection pattern over the thermodynamic and mechanical performance of the machine. The application of centrifugal compressor technology is a viable option for single-stage, two-phase compression at gas-volume fractions at or above 0.97, corresponding to gas qualities as low as 0.522 for suction pressures of 30 bar (435.1 psi) and 0.738 for suction pressures of 70 bar (1015.3 psi). The exact level of gasvolume fraction will of course depend on the values of suction pressure and pressure ratio, as well as the distribution of the liquid phase within the gas when it enters the machine. For future applications, the relative densities and phase properties of the gas and liquids that are handled will need to be considered. The thermodynamic evaluation of the machine showed that relative to a dry gas compressor, the compressor pressure ratio increased when the gas-volume fraction was decreased within the values that were tested (GVF between 1.0 and 0.97). The increase in pressure ratio was attributed to the larger density of the fluid that was being handled by the compressor when liquids were injected. In addition, the compressor temperature ratio showed a slight decrease when liquids were injected. This was probably caused by a transfer of energy from the gas to the liquid (heating of the liquid), and a limited condensate phase transition through the compressor. The specific compressor power consumption was also reduced as the liquid fraction was increased. Nevertheless, when compared to separating the liquid and vapor phases and boosting them as separate streams, the specific power consumption for wet gas compression was larger. For the data presented herein, the polytropic head for two-phase compression can be represented by a nondimensional head coefficient provided that the proper two-phase terms are included in the calculations. The two-phase polytropic efficiency of the machine decreased as the gas-volume fraction was reduced. This effect was more pronounced for the tests executed at the lower suction pressure. No evidence of liquid erosion was detected by visual inspection of the machine internals after the test. It was noticed that the internals of the machine were cleaned by the liquid that had been ingested. A repeatable reduction in the noise level of the machine was detected when the compressor was handling the wet gas mixture. The dynamic pressure transducer data showed that the pressure fluctuations within the flow were attenuated by the presence of liquid in the gas-stream.
GVF = 1.00
0
0 100 200 300 400 500 600 700 800 900 1000 Frequency [Hz]
Figure 17. Effects of Two-Phase Flow over the Vibration Response of the Machine, Measured in the Horizontal Direction at the Driven End (p1 = 70 bar, 9651 rpm, Film Injection Pattern). Figure 18 presents the frequency spectra of the horizontal vibration component, measured at the driven end of the machine, while it was operating at 9651 rpm, with a suction pressure of 30 bar (435.1 psi) and handling a volumetric flow of 2200 Am3/hr (77,692.27 ft3/hr). The bottom spectrum corresponds to dry gas operation while the top spectrum represents the machine behavior while handling a two-phase gas mixture, with liquids injected with a uniform droplet pattern. As may be seen in the figure, the vibration spectra for both the dry gas and the wet gas compression show similar trends at frequencies above the machine running speed. However, the spectrum corresponding to the wet gas compression shows some peaks in the subsynchronous range. The appearance of a peak at one half the running speed suggests the presence of some type of rotor instability. The compressor manufacturer is currently conducting an investigation to determine the source of this instability. Note, however, that this behavior was observed when the machine was operating with a gas-volume fraction of 0.97, which at the suction pressure of 30 bar (435.1 psi) represents a gas quality of 0.530. In addition, it is important to state that the vibration amplitude at the running speed and its harmonics did not increase when the subsynchronous component appeared. Furthermore, this subsynchronous component disappeared when the gas-volume fraction was increased above 0.98 (quality was increased above 0.62). The figures presented above provide a sample of the machine rotordynamic behavior that was observed during the tests. It is important to note that the behavior characteristics presented for each combination of suction pressure, machine speed, and liquid injection pattern were exhibited by the machine throughout the whole range of volume flows that were evaluated during each test period. This information is not included in this paper as it would be repetitive and would produce an excessively long document.
120
In general, the test results showed that the vibration of the machine was not significantly affected by liquid hydrocarbon ingestion for both the uniform droplet as well as the fluid film injection pattern. This may not be the case if the liquids are not uniformly distributed. Also, for the case in which the quality of the gas was below 0.62, the appearance of a subsynchronous vibration suggested that liquids could have been entrained in the seal areas at the impeller eye and balance piston, causing some type of rotor instability. The effects of liquid phase change that may occur inside the machine should be further investigated prior to embarking in the design of a multistage centrifugal compressor for wet gas applications. Recall that the phase change inside the compressor may cause a mismatch between the stages, leading to performance shortfalls. Furthermore, the effects of liquid ingestion over the machine internal loss mechanisms should be investigated.
g p Tot TP
= = = =
Gas Polytropic Total Two-phase
REFERENCES
ASME PTC-10, 1997, Performance Test Code on Compressors and Exhausters, American Society of Mechanical Engineers, New York, New York. Colby, G. M., 2004, Performance Test ProcedureRevision 3, D-R Internal Document # 003-085-001, Dresser-Rand Company, Olean, New York. Marshal, D. F. and Sorokes, J. M., 2000, A Review of Aerodynamically Induced Forces Acting on Centrifugal Compressors, and Resulting Vibration Characteristics of Rotors, Proceedings of the Twenty-Ninth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 263-280. Moore, J. J., Walker, S. T., and Kuzdzal, M. J., 2002, Rotordynamic Stability Measurement During Full-Load, Full-Pressure Testing of a 6000 PSI Reinjection Centrifugal Compressor, Proceedings of the Thirty-First Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 29-38. Schultz, J. M., 1962, The Polytropic Analysis of Centrifugal Compressors, ASME Journal of Engineering for Power, 84, pp. 69-82. Scott, S. L., Evolution of Subsea Multiphase Pumping and WetGas Compression, Presentation delivered during the Welcoming Address at the Thirty-Third Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas. Stahley, J. J., 2000, Field Performance Test ProcedureRevision 1, D-R Internal Document # 003-347-001, Dresser-Rand Company, Olean, New York.
DISCLAIMER
The information contained in this paper consists of factual data and technical interpretations and opinions, which, while believed to be accurate, are offered solely for informational purposes. No representation or warranty is made concerning the accuracy of such data, interpretations, and opinions.
NOMENCLATURE
Parameters D GVF h . m MW n N p P. Q Ro T U U2 Wp x Z p p 1 2 l = = = = = = = = = = = = = = = = = = = = = Impeller exit diameter Gas-volume fraction Enthalpy Mass flow Molecular weight Polytropic volume exponent Machine rotational speed Pressure (absolute) Power Actual volumetric flow Universal gas constant Temperature Tangential velocity Impeller tip speed Specific volume Polytropic head Gas quality Compressibility factor Polytropic head coefficient Polytropic efficiency Flow coefficient
ACKNOWLEDGEMENTS
The authors would like to thank Dr. D. Lee Hill and Dr. J. Jeffrey Moore, formerly of Dresser-Rand, for their assistance during the realization of the wet gas tests. Also thanks to Mr. Mark Kuzdzal (Turbomachinery Symposium Advisory Committee Monitor) for reviewing this manuscript and providing his suggestions to improve this paper. Considering the fact that the tested compressor is a constituent part of the K-lab test loop, the authors are grateful for the opportunity of using the K-lab compressor for wet gas testing, and would like to thank Steinar Jrgensen and Evert Wahlberg of Statoil for their willingness and support during the test. Finally, the authors would like to thank Statoil and Dresser-Rand for their funding of this work and permission to publish the results.
Subscripts = Machine inlet = Machine discharge = Liquid
HYDRAULIC SHOP PERFORMANCE TESTING OF CENTRIFUGAL COMPRESSORS

by Gary M. Colby
Test Engineering Supervisor Dresser-Rand Company Olean, New York
Gary M. Colby presently is a Test Engineering Supervisor with Dresser-Rand Company, in Olean, New York. He is responsible for developing test methods to meet objectives for production compressors and analytical aerodynamic testing of centrifugal compressors. Mr. Colby has held several engineering positions over his 32 year career at Dresser-Rand Company. The majority of his work experience has been in the thermodynamic performance field of centrifugal compressors. He has more than 14 years of experience in testing of centrifugal compressors both in the shop and the field. Mr. Colby studied Mechanical Technology for two years at the State University of New York at Alfred. He has authored several papers on hydrocarbon testing of compressors.
the discharge of the stage reducing the gas volume. The ratio of discharge volume to inlet volume is the volume reduction of the stage. To examine the effect of various parameters on the volume reduction of a stage the real gas polytropic head equation is presented.
n 1 n n P 1545 2 1 (1) HEAD = T1 Z1 n 1 P MW 1
k 1 n 1 k for an ideal gas = efficiency n
(2)
ABSTRACT
The purpose of this tutorial is to familiarize the rotating equipment engineer with in-shop hydraulic performance testing methods. The tutorial discusses the theory of similarity testing, the assumptions, and inherent errors. The presentation includes a selected case for demonstration. It is intended for engineers who define performance test requirements for assigned projects, review test agendas, and witness vendor compressor performance tests. Allowable test departures described in the ASME PTC-10 (1997) are reviewed and discussed.
INTRODUCTION
Low pressure inert gas Type 2 (ASME PTC-10, 1997) hydraulic performance testing of a centrifugal compressor is based upon the similitude between the specified condition volume reduction and the test condition volume reduction. Comprehension of volume reduction and the variables that effect volume reduction of the compressor are key to understanding the accuracy and limitations of the test. For a multistage compressor volume reduction is a critical parameter. The volume reduction of the first stage determines the impeller selection (capacity) of the second stage impeller that, in turn, determines the selection of the following stage and so forth through the discharge of the compressor. Once the final selection is determined the design volume reduction for the rotor is established. In conducting the in-shop performance test, the test volume reduction must match the design volume reduction to be representative of the compressor performance on the specified gas, the objective of the test.
Using Equation (1) assume, for the purpose of demonstration, that all variables remain constant with the exception of head and pressure ratio. If the head is increased, the pressure ratio must increase to maintain equality in the equation. The higher pressure ratio results in increased discharge pressure and lower discharge volume. Increasing head increases the volume reduction; decreasing head decreases volume reduction. Now for example assume that all the variables in Equation (1) are constant except for molecular weight and pressure ratio. If the molecular weight is reduced, the pressure ratio is also reduced to maintain equality. The reduced pressure ratio results in less volume reduction for the stage. Decreasing the molecular weight decreases the volume reduction; increasing the molecular weight increases the volume reduction of the stage. The same principle may be applied to the other variables in Equations (1) and (2) with reference to the pressure ratio and the following Table 1 developed. Table 1. Effect on Volume Reduction for Changes in Operating Parameters.
VOLUME REDUCTION
The volume reduction of a compressor stage is dependent upon the polytropic head (work output) and efficiency of the impeller, the gas density, and properties of the medium being compressed. Work applied to the gas through the impeller due to centrifugal force and diffusion results in increased pressure at
147
148
Note that the change in inlet pressure has no effect on the volume reduction of the stage. The discharge pressure changes proportionately to maintain equality; the volume reduction is the same. Pressure effect on the stage volume reduction is limited to the pressure effect on the gas isentropic exponent (k) and the compressibility factor (z). To visually represent volume reduction the author will use a three-stage compressor for an example. Figure 1 shows three stage curves of flow coefficient (inlet capacity) versus head for the compressor. The design capacity of the first stage impeller is identified by the solid line. The volume reduction of this stage under specified conditions yields the design capacity of the second stage represented by the solid line on the second stage curve. The third stage curve design capacity is established by the volume reduction of the second stage impeller.
Design VRD High Mol. Wt. VRD
for the flow coefficient as the impeller diameter is the same test to design. The head coefficient and efficiency of an impeller vary with the machine Mach number. Machine Mach number is defined as the tip speed of the first stage impeller divided by the inlet sonic velocity of the gas [Equation (5)].
Mach Numbermachine =
U = A0
U kgRTZ
(5)
Stage 1
Stage 2
Stage 3
Figure 2 shows a typical stage performance map of head coefficient and efficiency versus inlet flow coefficient at three machine Mach numbers. The variance between the test Mach number and the design Mach number will affect the head coefficient (and volume reduction) of the impeller. Therefore performance testing is required to be conducted at a Mach number relatively close to the design Mach number for each stage and is critical to the accuracy and relevance of the test. For this reason the ASME PTC-10 (1997) provides an allowable departure chart (Figure 3.3 of Code) for differences in the machine Mach number between test and design (Figure 3). The greater the machine Mach number the tighter the allowable tolerance. The Code only addresses the machine Mach number at the first stage, assuming that all following stages have the same percentage departure, test versus specified.
1.1
H ead
H ead
H ead
Capacity
Capacity
Capacity
normalized Effi ci e ncy
Figure 1. Stage Curve Example for a Three-Stage Compressor. If the compressor were operated on a molecular weight higher than the specified molecular weight the volume reduction would be greater. For purpose of demonstration assume that only the mole weight is changed. Starting at the design inlet capacity to the first stage, the greater volume reduction due to the increase in molecular weight reduces the operating capacity into the second stage relative to design. At the reduced capacity the second stage produces more head and volume reduction, in addition to the higher mole weight resulting in further deviation from the design capacity at stage three. The operating capacity at each stage at the higher volume reduction due to the higher mole weight is represented as a dashed line on the three stage compressor maps shown in Figure 1. If this were a test using a higher than design mole weight gas, the volume reduction would be much greater than design. The volume reduction of the compressor would need to be reduced back to design in order to produce a relative performance to the design condition. This is achieved by lowering the head of the compressor relative to design. Therefore, the shop test head will be lower than the design head. The polytropic head of the compressor is compared, test to design, using the polytropic head coefficient. The polytropic head coefficient () and efficiency of a stage are dependent upon the geometry of the impeller, the capacity being passed, the speed, and the inlet Mach number. The polytropic head of a stage is the product of the head coefficient and the impeller tip speed squared [Equation (3)].
1 0.9 0.8 0.7 1.4
Low Mach Number Med Mach Number High Mach Number
normalized Head Coefficient
1.2 1 0.8 0.6 0.7 0.8 0.9 1 1.1 normalized Flow Coefficient 1.2 1.3
Figure 2. Typical Stage Curve at Three Mach Numbers.

0.4 0.3 0.2 0.1 0 -0.1 -0.2 -0.3
Fig. 3.3 from the ASME PTC 10-1997
Head =
U2
g
(3)
The impeller geometry for the test is the same as design. The head coefficient test to design, by definition, should be equal. Therefore to reduce the head the speed of the compressor is lowered. For a multistage compressor an overall may be calculated using the overall head and the summation of the impeller tip speeds squared [Equation (4)].
Head g
(4)
Mach No. Test - Mach No. Specified
0.2
0.4
0.6 0.8 1 Mach No. Specified
1.2
1.4
1.6
The capacity of the compressor is compared during the test using a dimensionless flow coefficient, Q/ND3. Typically only Q/N is used
Figure 3. Allowable Test Mach Number Departure from Design. (Courtesy ASME)
149
DETERMINATION OF TEST CONDITIONS

Of the variables that effect volume reduction the compressor manufacturer has control over the inlet temperature, test gas medium, test speed, and inlet pressure. The test inlet temperature is dependent upon the cooling capacity of the manufacturers selected test stand. Typically the test inlet temperature is set at 100F. The difference in volume reduction caused by a difference in inlet temperature between design and test is corrected via the test speed. Table 2 lists the typical inert test gases used for an ASME PTC10 (1997) Type 2 in-shop performance test with the respective mole weights and approximate k values. Table 2. Typical Inert Gas Mediums for In-Shop Performance Testing.
Table 3. Specified Operating Conditions.

Inlet pressure Discharge pressure Inlet temperature Discharge temperature Molecular weight Inlet compressibility Inlet capacity Mass flowrate Discharge capacity Polytropic head Power Bearing loss Speed Machine Mach number Reynolds number 150 510 110 274 26.3 0.961 1999 1342.8 747.0 44,401 2362 52 13,100 0.658 3.63E+06 1.187 Aft3/min lbs/min Aft3/min ft lbs/lb gas hp hp rpm psia psia F F Mols
Inlet pressure has no effect on volume reduction except for the gas property variance as discussed previously. Inlet pressure does have a significant effect on the test. The pressure directly relates to the amount of mass compressed. The mass determines the power requirement as well as the level of effect the convection and radiation losses to the atmosphere have on the test results. The test Reynolds number is directly proportional to inlet pressure. The effects of heat loss and Reynolds number on the performance test are discussed later. In most cases the test suction pressure may be approximated at 10 percent of the specified value. A sample compressor application is presented as an example to discuss the selection of the test inlet conditions, gas medium, and speed. The process compressor selected is a five-stage straightthrough compressor design. The specified operating conditions are shown in Table 3. If helium was selected as the test medium for the sample compressor the difference in molecular weight, 4.0 versus 26.3, would result in the test volume reduction to be greatly reduced and therefore the required test speed would exceed the maximum continuous speed of the compressor. For this reason the test gas medium selected is almost always higher in mole weight than the specified gas. A second criterion for selection of the test gas is the k value of the test medium versus the k value of the specified gas. The k value not only affects the volume reduction but also the inlet Mach number. Generally the test gas with the closest k value to the specified gas is selected. Carbon dioxide was selected for the test gas in the sample compressor. The inlet temperature was set at 100 F. The inlet pressure was set at 44 psia. The test speed to achieve the same overall volume reduction may now be calculated based on thermodynamic law as follows. Thermodynamic Law:
Isentropic exponent (k)
Rearranging:
P2 P 1
1/ n
V1 V2
(8)
To have similitude between the specified gas conditions and the V1 test gas conditions, the volume reduction must be equal. V2 Therefore:
P V V1 = 2 must equal 1 V V 2 specified P 2 test 1 test
1/ n
(9)
The polytropic volume exponent (n) is a function of the gas properties and efficiency of the compressor. By definition, the efficiency of the test should equal the efficiency at design. Knowing the test gas properties the test polytropic volume exponent may be calculated and the required test pressure ratio to meet the specified volume reduction may be established. The test required head is then determined to meet the specified volume reduction using Equation (1). Head for a compressor is proportional to the speed squared [refer to Equation (3)]. Therefore:
Speedtest = Speed specified
Headtest Head specified
(10)
PV n = cons tan t
Therefore:
n n PV 1 1 = P 2V2
(6)
(7)
With the test gas selected and the test inlet conditions given the test speed was calculated at 11,070 rpm. The overall test conditions at the design point are shown in Table 4. These conditions are now reviewed and compared to the ASME PTC-10 (1997), Table 3.2, Allowable Departures from Specified Conditions for Type 1 and Type 2 tests. The comparison for the sample compressor is given in Table 5.
150

1.15 1.1
Table 4. Overall Test Conditions at Design Point.

Inlet pressure Discharge pressure Inlet temperature Discharge temperature Molecular weight Inlet compressibility Inlet capacity Mass flowrate Discharge capacity Polytropic head Power Bearing loss Speed Machine Mach number Reynolds number K value 44.0 170.2 100.0 348.9 44.01 0.987 1799 587.9 672.0 31,707 743 37 11,070 0.693 1.03E+06 1.261 Aft3/min lbs/min Aft3/min ft lbs/lb gas hp hp rpm psia
normalized Mu and Efficiency
Design Condition Curve Carbon Dioxide at 11070 rpm
1.05 1 0.95 0.9 0.85 0.8 0.7 0.8 0.9 1 normalized Flow Coefficient 1.1 1.2
psia F F mols
Figure 4. Head Coefficient and Efficiency Versus Inlet Flow Coefficient at 11,070 RPM. For the majority of applications the predicted stage by stage curve will fall on the specified curve with little variation. The author purposely selected an example where the curves do not match well, even though the test condition is within the allowable departures as given by the ASME PTC-10 (1997) for a Type 2 test. To visualize what caused the variation at the design inlet flow coefficient the test flow coefficients for each stage were normalized using their respective design values as the reference (Figure 5). If the test is perfect, the test condition line will fall exactly upon the design line in Figure 5. However at 11,070 rpm, the Code established test speed, the first stage volume reduction is greater under the test condition than under specified conditions. This results in the second stage operating at a flow coefficient less than design. As shown earlier if a stage operates at a lower capacity it will generate more head. The additional head and the test variances result in stage two producing a greater volume reduction. This effect continues through the stages until the fourth and fifth stages operate at almost 3 percent less relative capacity than under the specified condition. Overall the volume reduction at the test condition (11,070 rpm) is greater than the design and therefore a poor curve match.
1.02 1.01 1 0.99 0.98 0.97 0.96
Table 5. Allowable Departures for Sample Compressor at 11,070 RPM.
Volume Reduction Review for Example Compressor Design Condition Test on CO2 at 11070 rpm
There are inherent errors in Type 2 testing based on some of the assumptions taken when calculating the test conditions. The speed calculation is based on the inlet and discharge endpoints assuming a straight line polytropic path on the pressure-enthalpy diagram. The path is truly a slight curve. There are differences in Mach number between the test and specified condition. The allowable Mach number departure between design and test is compared only at the first stage of a section. The departure is not equal at all stages in a multistage compressor. The compressibility factor (z) change from inlet to discharge typically is greater under specified conditions than under the low-pressure test condition. These all cause variations in the individual stage volume reduction from specified. To evaluate the overall effect of the variances, compressor performance at the test condition is calculated on a stage by stage basis. The results are then converted to the dimensionless head coefficient (), efficiency, and flow coefficient. This test performance curve is plotted on the predicted curve at specified conditions for comparison. The results from the stage by stage calculation of the sample compressor are presented in Figure 4. The head coefficient versus capacity curve for the test condition is not in agreement with the specified gas curve.
normalized Flow Coefficient
Inlet Flange
Stage One
Stage Two
Stage Three
Stage Four
Stage Five
Discharge Flange
Figure 5. Normalized Flow Coefficient11,070 RPM. To correct for the higher volume reduction the test speed of 11,070 rpm was lowered reducing the volume reduction and improving the stage by stage match. A corrected test speed of 10,849 rpm was determined. A second stage by stage calculation was conducted at the revised speed and the predicted curve plotted on the specified gas curve (Figure 6). The predicted curve for the test is now found to agree closely with the specified gas predicted curve. At the design inlet flow coefficient the stage flow coefficients were normalized and plotted as before (Figure 7). At the lower test speed of 10,849 rpm no stage operates greater than 1 percent from its design flow coefficient.

1.15 1.1
151
Design Condition Curve Carbon Dioxide at 10849 rpm
1.05 1 0.95 0.9 0.85 0.8 0.7 0.8 0.9 1 normalized Flow Coefficient 1.1 1.2
Figure 6. Head Coefficient and Efficiency Versus Inlet Flow Coefficient at 10,849 RPM.
1.02 1.01 1 0.99
Volume Reduction Review for Example Compressor Design Condition Test on CO2 at 10849 rpm
normalized Flow Coefficient
through the compressor casing. Inlet pressure is directly related to inlet density and therefore the mass compressed and subsequently the power. The test must be run within the power limits of the available shop driver. Reynolds number relates to the boundary layer and frictional losses of the medium in the flowpath. Type 2 in-shop performance tests are almost always conducted at a lower Reynolds number than under specified conditions. The lower the Reynolds number the greater the frictional losses and lower the efficiency. This means that the head and efficiency observed during the test will be lower than what would be observed in the field at the same flow coefficient. The ASME PTC-10 (1997) (Figure 3.5) provides a chart showing the allowable departure in test Reynolds number relative to the specified value. The Code also provides a method to calculate a correction to be applied to the test data for the variance in Reynolds number. Utilizing the ASME PTC-10 (1997) method for Reynolds number correction, a graph is presented for the sample compressor (Figure 8). For this sample the allowable range for test Reynolds number is 10 percent to 138 percent of the specified value. If the test was conducted at the minimum 10 percent limit the anticipated correction would be 1.2 percent. Note in Figure 8 that the correction is not linear. The lower the absolute value of Reynolds number the greater the correction exponentially. The minimum allowable Reynolds number absolute value given by the Code is 90,000.
120
normalized Mu and Efficiency
Percent of Specified Reynolds Number Test Suction Pressure PSIA Head and Efficiency correction factor
0.98 0.97 0.96
80 40 0 150 100 50 0
Inlet Flange
Stage One
Stage Two
Stage Three
Stage Four
Stage Five
Discharge Flange
Figure 7. Normalized Flow Coefficient10,849 RPM. The stage by stage test condition at the lower speed is again compared to the allowable departure table to ensure that the Code requirements are met. Table 6 yields the departure values from the specified condition for the stage by stage calculations at the Code speed of 11,070 rpm and the adjusted speed of 10,849. Table 6. Allowable Departures for Sample Compressor at 10,849 RPM.
1.015 1.010 1.005 1.000 0.995 0.0E+000 1.0E+006 2.0E+006 Test Reynolds Number 3.0E+006 4.0E+006
Figure 8. Reynolds Number Correction. Most users do not allow the manufacturer to correct the test data for the differences in Reynolds number, test to design. The manufacturer should select a suction pressure to maintain the Reynolds number correction at its lowest level within the power and pressure restraints of the test stand. Even if correction is not allowed, it is recommended that the user be aware of the test level. There are two methods for determining the efficiency of the compressor. A torquemeter may be used to measure the work input into the compressor. From this power the friction of the bearings and seals may be subtracted to obtain the power input into the gas. The head, work output, is divided by the work input to achieve the efficiency. The second (most common) method, is the heat balance method where the work input is measured via the enthalpy rise from inlet to discharge. Using the heat balance method requires an understanding of the heat lost to atmosphere through the casing by way of conduction and radiation. The heat loss to atmosphere subtracts from the work input measured and if not minimized, eliminated, or accounted for will result in a falsely high efficiency. As the mass throughput increases, the percentage of heat loss to atmosphere relative to the
The stage by stage calculation of the test condition is a check of the test condition for the combined effect of all variances between test and specified conditions. Requesting a plot of the test curve on the specified curve from the manufacturer will validate that the test condition meets the objective of the test.
DETERMINATION OF SUCTION PRESSURE WITH RESPECT TO AVAILABLE POWER, REYNOLDS NUMBER, AND HEAT LOSS TO ATMOSPHERE
In determining the test suction pressure consideration must be given to the power available in the manufacturers shop, the Reynolds number variance from specified, and the heat loss
152
work input decreases. Here again the importance is seen of test suction pressure. The higher the pressure the greater the mass. For the sample compressor the convection and radiation losses were estimated at various test suction pressures and the results were plotted in Figure 9. The 44 psia suction pressure selected for the sample compressor yields an approximate 0.44 percent heat loss to atmosphere, essentially countering the Reynolds number correction. In the few cases where the heat losses cannot be minimized it is recommended that the casing be insulated for the performance test. The user is advised to ask what the approximate heat losses are for the in-shop performance test to ensure that the efficiency observed is representative.
1.600 1.400
Heat Loss to Atmoshere Percent of Total Heat Input
1.200 1.000 0.800 0.600 0.400 0.200 0.000 0 20 40 60 80 100 Test Suction Pressure - PSIA 120 140 160
Figure 10. Typical Test LoopStraight-Through Compressor Design. During the test the measured value of the balance leakage is known via the flowmeter installed into the return line. The discharge pressure of the compressor may be assumed to be the upstream condition of the seal and the flowmeter upstream pressure assumed as the downstream condition. From these data a pseudo clearance may be calculated and used to convert the results to specified conditions. This is considered a pseudo clearance as the calculated value corrects for any error in the constant. Back-to-back compressor designs have two major recirculation paths that need to be accounted for in the gas power calculation: the division wall leakage and the end seal leakage. The division wall seal is between the first section discharge and the second section discharge. Leakage from the second section discharge passes across the seal into the first section discharge through the interstage and recompressed by the second section impellers. The end seal leakage (also referred to as the seal balance leakage) passes across the end seal at the section two inlet back to the first section inlet. This leakage is compressed by the first section impellers and then passes through the interstage piping and vessels back into the second section suction. Essentially the division wall seal leakage is recycled around the second section and the end seal leakage is recycled around the first section. Arrangement of the test loop will allow for the measurement of these leakages so that they may be compared to predicted values and used in the conversion of test results. Figure 11 shows a loop arrangement for a back-to-back compressor that enables the leakages to be measured. Each section is piped with its own test loop similar to the straight-through example. An orifice is placed in the seal balance line to measure the seal leakage directly. It should be noted here that during the test each end seal is referenced separately for seal supply (oil or gas) purposes. The two loops are connected by a balance line between the first section discharge and the second section inlet, which is also orificed. Figure 10 shows an additional balance line, discharge to discharge, the purpose of which will be discussed later. During the second section test the seal balance line valve is open as is the discharge to inlet loop balance line valve. The discharge to discharge loop balance line valve is closed. This allows the suction pressure of the second section to be very close to the discharge pressure of the first section enabling leakage across both of the seals in question. To maintain a mass balance in the system,
Figure 9. Convection and Radiation Losses to Atmosphere.
MEASURING RECIRCULATION
At the specified operating point API 617, Seventh Edition (2002), states that the horsepower shall be guaranteed within 4 percent of the quoted value. The percentage varies dependent on the user. Note that efficiency is not guaranteed. The power required is the product of the total mass compressed and the work input into the gas [Equation (11)]. It is recommended that where possible, any recirculation or leakage losses should be measured during the shop performance test and the results applied to the specified condition in the conversion of results.
GHP =
Head Mass 33000 Efficiency
(11)
Shown in Figure 10 is a typical loop schematic for a straightthrough single section compressor. The straight-through unit loop consists of an inlet flowmeter, a discharge throttle valve, and a heat exchanger. The balance piston return line is piped through a flowmeter and into the test loop upstream of the main inlet flowmeter. This enables measurement of the balance piston seal leakage. The predicted balance leakage is computed similar to an orifice equation. Manufacturers do not all use the same equations for calculating predicted leakages but the physics of the computation is common. A simplified example is given in Equation (12).
mass = C D S
where: C D S P = = = = Constant Seal diameter Seal clearance Pressure
( Ph Pl )
densityh
(12)
Subscripts: h = High side of seal l = Low side of seal
153
was not the actual surge point of the compressor as observed in the field. Today with the advent of faster data acquisition equipment and vibration monitoring, surge may be accurately determined during the shop test. There are five criteria that may be used to determine the minimum stable flow point.
Section 1 Section 2
Audible surge Onset of subsynchronous vibration Peak head Flow instability Pressure instability
Audible flow reversal in the compressor is readily evident as is flow and pressure instabilities. Observing the vibration frequency spectrum is very important in identifying surge. Subsynchronous vibration will occur at the onset of surge providing there is adequate energy input into the unit. The frequency of the vibration generally points to the source of the flow instability. Frequencies in the 6 to 16 percent range may be indicative of a stall in the stationary components of the flow path where frequencies in the 70 to 90 percent range indicate surge is initiated in the impellers. Observing the head coefficient is another method of determining surge. In many cases the head drops off before any other indications are noted. An example is presented in Figure 12. A unit was throttled toward surge slowly with a data acquisition system set to calculate the performance every four seconds. As the unit approached lower flow the head coefficient stopped rising and started to decrease as the flow was reduced further. The head then started to rise again as the flow started to reduce further until an audible surge was found. At the bottom of the head droop, subsynchronous vibration was observed at the journal bearing. The minimum stable flow point was recorded at a capacity corresponding to the initial peak head point. This was at a 10 percent higher capacity than the audible surge capacity.
Peak Head
Figure 11. Typical Test LoopBack-to-Back Compressor Design. whatever leakage leaves the second section loop to the first section loop must return to the second section loop. It does so via the discharge to inlet loop balance line. The division wall leakage is the difference in mass flow between the loop balance line flowmeter and the end seal flowmeter. The evaluation of the leakage rate is conducted in the same manner as the balance drum leakage. The upstream condition for the division wall is the second section discharge; the downstream condition is the first section discharge. For the end seal, the upstream condition is the second section suction and the downstream condition the first section suction. The flow through the second section impellers is the second section main orifice minus the seal balance orifice. While testing the first section the valve arrangement must be adjusted to eliminate the recycles so that a proper enthalpy rise may be observed. If the valves were left open then the division wall leakage would mix with the first section discharge mass resulting in an error in the first section discharge temperature observed at the flange. During the section one performance test the discharge to inlet loop balance valve is closed and the discharge to discharge balance line valve opened. The end seal valve is also closed. This arrangement causes the two opposing discharges to be at the same pressure eliminating the division wall seal leakage.
Polytropic Head Coefficient
Audible Surge
Onset of Subsynchronous vibration
THE STANDARD TEST PROCEDURE

Typically a shop performance test is conducted at only the relative design point speed. The test typically consists of five test points at equally spaced capacities from overload to surge, with one point being at the design inlet flow coefficient. At each flow point the compressor is allowed to reach thermal equilibrium before the data point is recorded. Surge points are also observed at two alternate speeds to define the slope of the surge line. The offspeed surge points are not normally thermally stable before the point is recorded as determining the head and flow coefficient of the point is the objective, not the efficiency. The off-speed surge points also assist the manufacturer in determining where, in which stage, surge was initiated. If the three surge points are very close in flow coefficient, surge is initiated early in the compressor, the first or second stage. A greater separation between the three surge points would indicate surge to be initiated in one of the later stages. Alternate speed points are also observed for constant speed compressors. In this case the slope of the surge line is established for off-design mole weight conditions. In the past determining surge was done by throttling the compressor until audible surge was detected. Many times this capacity
Inlet Flow Coefficient
Figure 12. Head Versus Inlet Flow CoefficientTransition Curve near Surge.
ADDITIONAL SPEED LINE TEST

When should alternate test speed lines, in addition to the specified design point speed line, be requested? It depends upon the compressor, the process, and the user. Discussing the compressor operation with the process engineer will assist the rotating equipment engineer with what are the critical parameters required for the process to run efficiently. If there are alternate operating conditions that are critical to the process the engineer may request a second speed line corresponding to that condition to be tested. It is a matter of risk assessment. The author recommends reviewing the Mach number of the alternate condition with respect to the specified operating point. If the variance is greater than would be allowed by the Code then an additional line may be warranted.
154
Methods used to extrapolate the overall test performance data established at the specified operating point to alternate operating points varies by manufacturer. A review of the manufacturers method of extrapolation may also determine whether an additional test at an alternate operating point is warranted.
EVALUATION OF TEST RESULTS

Evaluating the pressure and temperature readings of the performance test to head and efficiency terms should always be conducted using real gas laws and an equation of state for gas properties that is well referenced for the test medium. This tutorial is intended to address the method for establishing proper test conditions for in-shop performance testing. The author recommends the evaluation procedures outlined in ASME PTC-10 (1997) be followed.
The Type 1 test point(s) should be established by agreement between the user and the manufacturer dependent upon the objective of the test. It has been the authors experience that this objective is not always well defined in the proposal stages of a project. It is recommended that discussions between the user and the manufacturer concerning the objectives of the test get started as early in the project as possible. This will ensure that the test stand loop design will accommodate all of the objectives.
SUMMARY
The information provided in this tutorial should provide the process and rotating equipment engineer with a better understanding of performance testing methods for centrifugal compressors. When reviewing the ASME PTC-10 (1997) the engineer will have a greater understanding of why the limitations outlined in the Code are given.
TYPE 1 PERFORMANCE TESTING

An ASME PTC-10 (1997) Type 1 test is conducted with the specified gas at or very near the specified operating conditions. Deviations from the specified gas conditions are subject to limitations given in Tables 3.1 and 3.2 of the Code. The same principles apply to the Type 1 test as to the Type 2 test described earlier. It the test was conducted using the actual specified gas and inlet conditions there would be no deviations from the specified condition to the test. However, this is rarely the case. In many instances the inlet temperature at the specified condition cannot be achieved due to the cooling capacity of the manufacturers test stand and the specified gas is not available. The test may be conducted on a gas different from the specified gas; however, the test gas must have a mixture k-value very close to the specified gas. This is required to ensure that the Mach number is the same as specified and that the thermodynamic conversion of the work input to the gas medium produces the same pressure ratio. Typically the manufacturer has a local gas supply close to pure methane. With this source carbon dioxide, propane, or other gases may be blended to achieve the specified molecular weight and kvalue. If the test inlet temperature is greater than the specified inlet temperature the test molecular weight will have to be higher than specified to offset the temperature change and any associated change in compressibility factor. The higher mole weight mixture must still have a mixture k value close to specified. Reviewing Equation 1 and Equation 5 it can be seen that the product of the gas constant, R (1545/mole weight), inlet temperature, and inlet compressibility are common to volume reduction and machine Mach number. If a blended gas has the same k value and the return-to-zero (RTZ) product is maintained at the inlet, the performance will be the same as on the specified gas. The author recommends that the inlet RTZ product and inlet pressure be maintained within 2 percent of the specified value. The ASME PTC-10 (1997) (Table 3.2) allows a much greater tolerance. Comparable plots of compressor performance maps showing polytropic head, efficiency, pressure ratio, and power may be produced for the specified gas and the planned test gas medium. These plots will demonstrate how the compressor performance operating on the test gas blend and conditions conforms to the compressor performance under the specified gas conditions. These curves should be provided in the test agenda.
NOMENCLATURE
MW T Z P n k U A0 g V N Q GHP Mass D S Density C = = = = = = = = = = = = = = = = = = = Molecular weight, mols Temperature, degrees Fahrenheit Compressibility factor, dimensionless Pressure, pounds force per square inch Polytropic volume exponent, dimensionless Isentropic exponent Tip speed, feet per second Polytropic head coefficient, dimensionless Sonic velocity, feet per second Gravitational constant, feet per second squared Specific volume, cubic feet per pound mass Speed, revolutions per minute Capacity, cubic feet per minute Gas horsepower Mass flow rate, pounds per minute Diameter, inches Radial clearance, inches Pounds mass per cubic foot A given constant
Subscripts: 1 2 h l = = = = Inlet Discharge High pressure side (upstream) Low pressure side (downstream)
REFERENCES
API Standard 617, 2002, Axial and Centrifugal Compressors and Expander-Compressors for Petroleum, Chemical and Gas Industry Services, Seventh Edition, American Petroleum Institute, Washington, D.C. ASME PTC-10, 1997, Performance Test Code on Compressors and Exhausters, American Society of Mechanical Engineers, New York, New York.
FUNDAMENTALS OF FLUID FILM JOURNAL BEARING OPERATION AND MODELING

by Minhui He
Machinery Specialist
C. Hunter Cloud
President BRG Machinery Consulting, LLC Charlottesville, Virginia
and James M. Byrne

President Rotating Machinery Technology, Inc. Wellsville, New York
Minhui He is currently working as a Machinery Specialist at BRG Machinery Consulting LLC, in Charlottesville, Virginia. His responsibilities include vibration troubleshooting, rotordynamic analysis, as well as bearing and seal analysis and design. He is a member of STLE, and is also conducting research on hydrostatic journal bearings and hydrodynamic thrust bearings for turbomachinery applications. Dr. He received his B.S. degree (Chemical Machinery Engineering, 1994) from Sichuan University in China. From 1996 to 2003, he conducted research on fluid film journal bearings in the ROMAC Laboratories at the University of Virginia, receiving his Ph.D. degree (Mechanical and Aerospace Engineering, 2003). C. Hunter Cloud is President of BRG Machinery Consulting, LLC, in Charlottesville, Virginia. He began his career with Mobil Research and Development Corporation in Princeton, New Jersey, as a Turbomachinery Specialist responsible for application engineering, commissioning, startup, and troubleshooting for production, refining, and chemical facilities worldwide. During his 11 years at Mobil, he worked on numerous projects, including several offshore gas injection platforms in Nigeria, as well as serving as reliability manager at a large U.S. refinery. Currently, Mr. Cloud also serves as Lab Engineer at the University of Virginias ROMAC Laboratories, where he is pursuing a doctorate. His research focuses on the measurement of turbomachinery stability characteristics. He is a member of ASME, the Vibration Institute, and the API 684 Rotordynamics Task Force.
155
James M. Byrne is President of Rotating Machinery Technology, Inc., in Wellsville, New York. He began his career designing internally geared centrifugal compressors for Carrier, in Syracuse, New York. Mr. Byrne continued his career at Pratt & Whitney aircraft engines and became a technical leader for rotordynamics. Later he became a program manager for Pratt & Whitney Power Systems, managing the development of new gas turbine products. Mr. Byrne holds a BSME degree from Syracuse University, an MSME degree from the University of Virginia, and an MBA from Carnegie Mellon University.
ABSTRACT
Widely used in turbomachinery, the fluid film journal bearing is critical to a machines overall reliability level. Their design complexity and application severity continue to increase making it challenging for the plant machinery engineer to evaluate their reliability. This tutorial provides practical knowledge on their basic operation and what physical effects should be included in modeling a bearing to help ensure its reliable operation in the field. All the important theoretical aspects of journal bearing modeling, such as film pressure, film and pad temperatures, thermal and mechanical deformations, and turbulent flow are reviewed. Through some examples, the tutorial explores how different effects influence key performance characteristics like minimum film thickness, Babbitt temperature as well as stiffness and damping coefficients. Due to their increasing popularity, the operation and analysis of advanced designs using directed lubrication principles, such as inlet grooves and starvation, are also examined with several examples including comparisons to manufacturers test data.
156
INTRODUCTION
The objectives of this tutorial are to provide the reader the following with respect to fluid film journal bearings:
What are some common operational limits?
Surface speeds: in the old days, less than 200 ft/s; today, up to 450 ft/s Unit or specific load, WU: in the old days, less than 250 psi; today, up to 900 psi Babbitt
lined bearings typically operate below 200F, while alternative materials and lubricants can run above 250F.
A basic understanding of their physics and operational considerations
A basic understanding of their modeling fundamentals The knowledge to better interpret more advanced papers and
topics
Film thickness values must typically be greater than 0.001 inch,

with more aggressive applications above 0.0005 inch. Figure 1 shows a number of severe applications in successful operation today. Twenty years ago, most of these designs would have been ruled out as too aggressive. Today, they perform reliably as a result of advanced design features and the tools necessary to model and predict their performance.
1000 800 600 400 200 150
Older designs limited to 250 psi and 200 ft/s
A good reference source for the future

This tutorial is not:
A design guideline. The authors do not intend to teach how to
design a bearing for any particular application. The literature is replete with fine design guidelines including those by Nicholas and Wygant (1995).
A bearing primer. The authors expect the reader to have a basic

understanding of fluid film bearings, their use, and basic operation. They do not describe all types of bearings, nor the evolution of their design.
Limit Curve Tmax < 200 oF
Limit Curve Tmax < 220 oF
A thrust bearing tutorial. The authors focus solely on journal
bearings, although many of the topics and much of the physics are also applicable to thrust bearings. The authors primary audience is plant machinery engineers evaluating new versus old designs to fix their problems as well as central engineering machinery specialists in charge of selecting and auditing bearing designs for new machinery. Their goal is to prepare these individuals to ask good questions of those performing a bearing design or analysis. Plant engineers must understand the limits of any analysis so that they can manage risk and assess alternatives. Bearing designers will also find the material useful in supplementing their expertise. These individuals must understand the underlying physics behind the computer program they are running. They too must understand the limitations and risks associated with their analysis. They must understand all of the options and inputs to their bearing code, plus understand what the output is telling them. Why are the fundamentals of journal bearing operation and modeling important?
200
250
300
350
400
450
Surface Velocity (f/s)
Figure 1. Severe Journal Bearing Applications. Figure 2 shows an example of one such severe application. In this case, an old technology bearing design was replaced with a modern bearing design utilizing a number of advanced features. The results: a 30 degree temperature reduction with better dynamic characteristics. This bearing change allowed a multimillion dollar compressor train to enter service.
Severe Application: N=10,580 rpm, 369 f/s, 298 psi Results: 30 Degree F Temperature reduction!
Designs are more and more aggressive with less margin for
error.
Loads and speeds continue to increase in new machinery. While the basic fluid dynamics of fluid film bearings are well
understood, secondary effects such as elastic deformations, heat transfer to the solids, and turbulence are less well established. old analysis methods to fall short.
120 110 100 90 80 45
Old Technology Bearing
230F
Innovation breeds new designs and technologies that cause the The desire for lower power loss and lower oil consumption The desire for improved reliability forces better understanding. The cost of redesign (trial and error) is enormous. The cost of a plant outage is greater. You cannot test everything!
How does a poor bearing design manifest itself?
200F
New Technology bearing
122F
131F
140F
149F
158F
50
55
60
65
70
Inlet Temperature (degC)

Figure 2. Comparison of Bearings with Old and New Technologies. The first section of this tutorial begins with a discussion on the operational aspects of fluid film bearings. Bearing geometrical aspects are discussed and the basic physics of fluid film bearing operation are developed. The second section uses what was learned in the operational section and describes the means by which one can model or predict fluid film bearing behavior.
High bearing metal temperatures, eventually leading to bearing

failure
High machinery vibrations Excessive power loss Excessive oil consumption
157
OPERATION
The operational characteristics of a journal bearing can be categorized into static and dynamic aspects. Static characteristics include a bearings load capacity, pad temperature, power loss, and the amount of oil it requires during operation. A bearings load capacity is often measured using either eccentricity ratio that relates directly to its minimum film thickness, or the maximum pad temperature. A bearings dynamic performance is characterized by its stiffness and damping properties. How these properties interact with the rotor system determines a machines overall vibrational behavior. The main objective of this section is to provide a general understanding of the basic physics that governs a bearings static and dynamic operation. By comparing several common bearing designs, the key performance issues of interest will be examined. At the end of this section, one should understand the following:
Figure 3(b) defines some other key geometric parameters with respect to a tilting pad bearing. The pad pivot offset is given by the ratio:
Pivot Offset
(2)
Centrally pivoted pads (50 percent offset or = 0.5) are the most commonly applied. However, 55 to 60 percent pivot offsets are often seen in high load applications because of their relatively low pad temperatures (Simmons and Lawrence, 1996). While the bearing assembly radius Rb determines the largest possible shaft size that can fit in the bearing, the individual pads may be machined to a different radius indicated by Rp. These radii along with RJ establish a very important bearing design parameter, preload or preset, which is defined by two clearances:
Development of hydrodynamic pressure or load capacity Relationships between viscous shearing, temperature
power loss, and load capacity
cb = Rb RJ
rise,
Assembled or Set Bore Clearance Pad Machined Clearance
c p = R p RJ
(3)
General
influence of dynamic coefficients on rotordynamics including stability
Preload or preset, m, is subsequently defined using their ratio:
Speed and load dependency of static and dynamic properties Different behaviors of fixed geometry and tilting pad bearings
Geometric Parameters Before discussing the operational aspects of journal bearings, some basic geometric parameters need to be defined. Figure 3(a) shows an arbitrary bearing pad of axial length L and arc length P. The pad supports the journal of radius RJ rotating at speed . The radial clearance, c = Rb RJ, allows the journal to operate at some eccentric position defined by distance e and attitude angle . The attitude angle is always measured with respect to the direction of the applied load W and the line of centers. For a fixed geometry bearing, the line of centers establishes the minimum film thickness location. However, this is generally not true for a tilting pad bearing. Instead, the trailing edge of a tilting pad often becomes the minimum film thickness point.
c b m = 1 c p
(4)
Figure 4 shows how preload affects the relative film shapes within the bearing. For m = 0.0, the pad radius and assembly radius are equal. Typically, preload values are positive, which, as shown in Figure 4, causes the shaft/bearing center (OJ = Ob) to sit lower relative to the pad center OP. Thus, one develops the connotations of preloading the bearing. While many associate preload with only tilting pad bearings, it can also be used in the design of fixed geometry bearings. A lemon bore or elliptical bearing is the most common example (Salamone, 1984).
LINE OF CENTERS
RB
RP
RJ
RB
RJ
Figure 4. Pad Preload. All of these geometric design parameters can significantly affect a bearings static and dynamic characteristics. For example, tighter clearance and higher preload usually lead to greater load capacity and higher stiffness. Smaller clearance usually means higher Babbitt temperature, etc. In this tutorial, however, the authors will predominately focus on the influences of the operating parameters, such as shaft speed and bearing load. Excellent discussions on the effects of various geometric parameters can be found in Jones and Martin (1979) and Nicholas (1994). Static Performance (1) To be classified as a bearing, a device must fundamentally carry a load between two components. A journal bearing must accomplish this task while the shaft rotates and with minimum wear or failure. Inadequate load capacity leads to either rubbing contact between the journal and bearing surfaces, or thermal failure of the lubricant or bearing materials. Therefore, the first step is to explain the load carrying mechanism in a fluid film bearing.
(a)
(b)
Figure 3. Bearing Geometry. Since typically the journal position relative to the bearing is of interest, an eccentricity ratio is defined using the radial clearance:
E= EX
e 2 = E2 X + EY c e e = X , EY = Y c c
At rest, normally the eccentricity ratio E would be expected to be 1.0 with the journal sitting on the bearing pad. E can be greater than 1.0 if the shaft sits between two tilting pads.
158
Load Capacity Section summary:
A convergent wedge, surface motion, and viscous lubricant are

line loci are shown through examples of three common designs.
necessary conditions to generate hydrodynamic pressure or load capacity.
Typical pressure profiles, journal eccentricity ratios, and center Hydrodynamic forces in fixed geometry bearings have strong
cross-coupled components. Such cross coupling is negligible in tilting pad bearings due to the pads ability to rotate. In the early 1880s, the underlying physics of how a fluid film journal bearing supports a loaded and rotating shaft was a mystery. Using some of the lessons learned from pioneers in this field, the authors will examine the most important physical phenomena governing a bearings load capacity: hydrodynamic pressure. The concept of hydrodynamic lubrication was born from the experimental work of Beauchamp Tower (1883, 1885). Commissioned to study the frictional losses in railroad bearings (Pinkus, 1987), Tower encountered a persistent oil leak when he decided to drill an oiler hole in his bearing (Figure 5). After a cork and wooden plug were blown out of the hole, Tower realized that the lubricating oil was becoming pressurized. Tower altered his design such that the oil was supplied through two axial grooves that allowed him to install pressure gauges on the bearing surface. Figure 6 shows an example of the resultant pressures that Tower measured. Integrating this pressure distribution, Tower discovered that it equaled the load he applied on the bearing. In one experiment, Towers pressure profile integration yielded a film force of 7988 lbf compared with the applied load of 8008 lbf, an amazingly accurate result (Dowson, 1998).
Figure 6. Towers Pressure Measurements. (Courtesy Tower, 1885)
Figure 7. Examples of Hydrodynamic Lubrication. Figures 7(b) and (c) show two other examples of fluid being dragged into a convergent clearance. The journal creates such convergent clearance because of its eccentric operation and/or the radii difference between it and the pad. For a perfectly centered journal with a zero preloaded pad, the inlet film thickness equals the minimum or outlet film thickness (hi = ho) and no pressure would be expected to be developed in the film. The phenomenon of hydroplaning is another good example. Here the tire deformation creates a converging clearance that generates enough pressure in the water film to support the weight of the car. Both situations can be treated like the plane slider with hi > ho. To understand the pressure distributions and the film forces developed in some typical journal bearing designs, here the authors examine a two axial groove bearing [often referred to as a plain journal bearing (Salamone, 1984)], a pressure dam bearing, and a tilting pad bearing with four pads [Figure 8(a)]. Running at the same diameter, axial length, bore clearance, preload, oil viscosity, and speed, Figure 8(b) displays each bearings circumferential pressure distribution. Each bearing has its journal position fixed downward halfway within the clearance at EX = 0.0 and EY = 0.5. The two axial grooves pressure distribution in Figure 8(b) has a peak pressure over 750 psig. It is important to notice that the pressure distribution is not symmetric about this peak. Furthermore, the peak pressure does not occur at the minimum film thickness position (270 degrees) where one would instinctively anticipate. These two pressure distribution characteristics are fundamental to all bearing types where hydrodynamic pressures are developed. No pressure is developed in the upper half of the bearing because of the diverging clearance and the relatively low oil supply pressure (20 psig). This condition, which exists in most fixed geometry bearings, causes the film to cavitate and restricts the pressure in the film to the vapor pressure of the lubricant. Physically on a cavitated pad, the fluid film is ruptured and the rotating shaft drags streamlets across this region (Heshmat, 1991). The films positive pressure area results in a 1374 lbf effective force on the shaft at an angle of 56 degrees. Recalling that the journal was displaced vertically downward only, one should notice that the fluid film has now generated a responding force with a
Figure 5. Towers Experimental Bearings. (Courtesy Tower, 1883) While Tower was conducting his experiments, Osborne Reynolds (1886) derived the theoretical justification for the load carrying capacity of such journal bearings. He found that a fluids pressure would increase when it is dragged by a moving surface into a decreasing clearance, like the plane slider situation shown in Figure 7(a). Such a situation demonstrates the governing principle of hydrodynamic lubrication. Without relative motion or a converging clearance, no pressure or load capacity will be developed. It is the pressure in the lubricant film that carries the external load and separates the solid surfaces, which further confirmed Towers observations.
159
Figure 8. Pressure Profiles for Three Bearing Designs. significant horizontal component. Such a reaction is the main reason plain journal bearings create such dynamic stability concerns. A cross coupling is experienced since movement in one direction causes a force component in the perpendicular direction. This behavior will be discussed further with respect to dynamic characteristics. Since its lower half film profile is the same as that of the two axial groove bearing, the pressure dam bearing has an identical pressure distribution in its lower half. However, because of the presence of the dam, the upper half now has a converging wedge that generates a positive pressure profile. The peak pressure occurs at the dam location. With the upper half pressure counteracting the pressure developed in the lower half, the force exerted on the shaft has slightly reduced in magnitude to 1162 lbf and rotated more horizontally in direction. Film pressure is developed on the bottom two pads of the tilting pad bearing. At steady-state, the moments on a pad must be balanced in order for the pad to reach an equilibrium tilt angle. When this occurs, the pad pressure force vector passes directly through the pivot. Unlike the two axial groove and pressure dam bearings, the tilting pad has produced a resultant film force that is almost purely vertical. In this case, a vertical displacement of the journal has resulted in an almost directly vertical force. This desirable characteristic totally relies on the pads ability to tilt even though the tilt angles are very small (on the order of 0.01 degree). Boyd and Raimondi (1953) were among the first to explain this behavior. Both they and Hagg (1946) realized the implications from the dynamics standpoint, which will be discussed later. As a final point on Figure 8(b), one should notice the additional reduction in the tilting pad bearings film force magnitude (1109 lbf versus 1374 lbf for the two axial groove bearing). This is expected since the pad area that carries load has been reduced. The two axial groove bearing has 150 degrees of lower half pad arc length, while the tilting pad bearing has only 2 72 degrees = 144 degrees with a supply groove in between. While good for demonstrating film forces and their nonlinear nature, setting the journal at a fixed eccentricity like in Figure 8(b) does not represent a realistic operating condition. In reality, the bearing will adjust the shafts position till the hydrodynamic force balances the applied load W. For the same three bearings, Figure 8(c) shows the pressure profiles when a constant load (W = 1100 lbf) is applied downward at 270 degrees. Also shown are the resultant shaft eccentricity ratios where the shaft has reached its steady-state equilibrium position. Comparing Figures 8(b) and (c), the journal inside the two axial groove bearing has now had to shift horizontally in order to create a film profile that only opposes the vertical load. This is evident in the more vertical orientation of the pressure distribution. Similar
behavior is observed for the pressure dam bearing. However, because of the pressures created by the dam and its angular orientation, the shaft reaches a position of higher eccentricity and greater attitude angle than the plain journal bearing. Both fixed geometry bearings are able to support the load at a lower eccentricity than the tilting pad bearing. This higher load capacity is expected when one recalls the resultant film forces created in Figure 8(b). One should now have a basic feel for the pressures developed by hydrodynamic lubrication. Through different bearing geometries, one has seen how different converging wedges create different pressure distributions, and, thus, various abilities to support load. What has not been emphasized is the importance of the lubricants viscosity that determines the pressure generation just as much as the bearings geometry. As the next section will describe in detail, the lubricants viscosity will decrease because of the internal heat generated during operation. Discussions and comparisons, so far, have kept the viscosity constant or isoviscous. With this restriction removed, Figure 9 demonstrates that a bearings load capacity is a strong function of its operating condition. In Figure 9(a), the load is fixed at 1100 lbf and the shaft speed varies from 1000 rpm to 19,000 rpm. When stationary, the shaft sits on the bottom with zero attitude angle and unity eccentricity ratio (for the tilt pad bearing, the eccentricity is slightly higher because the shaft rests between the pads). As the shaft accelerates, the journal is lifted higher and higher by increasing hydrodynamic pressure.
(a)
(b)
Figure 9. Load Capacity Trends Allowing for Viscosity Degradation Effects . Although all three bearings exhibit this general trend, different loci of the journal center are observed for each bearing. For the plain journal bearing, the journal center moves approximately along a circular arc. With increasing speed, the journal gradually approaches the bearing center because it requires less and less of a convergent wedge to produce a 1100 lbf hydrodynamic force. The pressure dam bearing behaves similar to the plain journal bearing at low speeds. At high speeds, the dam generates significant hydrodynamic pressure that pushes the journal away from the bearing center. For the tilting pad bearing, the journal center is directly lifted in the vertical direction maintaining very little attitude angle. Such small attitude angles are only possible because of the pads ability to tilt. Figure 10 shows an example of what occurs when this tilting ability is lost. Here, a noticeable attitude angle was observed when the loaded pad was locked. When this is encountered, it may be attributable to pivot design, operating conditions, or even thermocouple or resistance temperature detector (RTD) wiring problems. Figure 9(b) reverses the situation, keeping speed constant and varying load. Like a speed increase, load reduction allows the journal position to reach a lower eccentricity. At 100 lbf, the journal is almost perfectly centered for both the tilting pad and two axial groove bearings. Once again, because of the dam, the pressure dam bearing maintains a higher eccentricity ratio even at this light loading.
160
boundary lubrication. Compared to hydrodynamic lubrication, the mating surface roughnesses in boundary lubrication become important and the lubricant film shows increased friction coefficient (Elwell and Booser, 1972; Gardner, 1976). However, since the shaft typically does not vibrate at such low speed, boundary lubrication does not necessarily mean bearing failure. A generally accepted criterion is that the minimum film thickness must be at least twice the surface roughness to ensure successful operation. Viscous Shearing and Temperature Rise Section summary:
Viscous shearing causes temperature rise. Temperature rise affects bearing performance through lubricant
viscosity reduction and solid deformations.
Shaft speed is the primary operating factor compared to load.

Figure 10. Shaft Centerline Test Data for a Tilting Pad Bearing with a Locked and Unlocked Loaded Pad. (Courtesy Brechting, 2002) If the lubricants viscosity was not allowed to change, the centerline trends in Figure 9(a) and (b) would be identical. In other words, increasing speed and decreasing load would be equivalent. This is why the dimensionless Sommerfeld number S, which combines speed and load effects, was often used to define bearing similarity in early isoviscous studies. The Sommerfeld number is still used today to compare bearings and is typically defined as: While producing load carrying capacity, the lubricant film also generates heat that causes temperature rise in operation. It is well known that a lubricants viscosity is extremely sensitive to temperature. Table 1 provides some indication for several common turbine oils. Figure 9 has already shown some thermal effects: due to different temperature rises, increasing speed and decreasing load are not equivalent, and the centerline trend in Figure 9(a) differs from that in (b). To fully understand the thermal effects of journal bearings, one must grasp the principle of viscous shearing, which is their heat generation mechanism. Table 1. Lubricant Viscosity at Different Temperatures.
Temperature (F) 104 212 ISO VG 32 3.75e-6 5.98e-7 Absolute Viscosity (Reyns = lbf-s/in2) ISO VG ISO VG 46 68 5.42e-6 8.06e-6 7.68e-7 9.99e-7
S=
WU c
(5)
One should note that it does not include all geometry factors such as preload or pivot offset. Since thermal and deformation effects are also absent, caution must be used when comparing bearing performance using this rather simplified relationship. Since the external load is a vector that has direction, journal steady-state position is as much dependent on the loads direction as it is on its magnitude. Figure 11 shows the resultant pressure profiles and eccentricity ratios when the 1100 lbf load is now directed horizontally. Because the load is now pushing toward their axial groove, both fixed geometry bearings pressure areas are dramatically reduced. Likewise, their load capacity is reduced as evidenced by their higher shaft eccentricity ratios. Tower came to the same realization during his experiments. The tilting pad bearing, however, achieves the same eccentricity ratio as before. This can be attributed to its symmetry (four pads equally distributed) and each pads ability to tilt and generate a load carrying pressure. If the load was directly on pad, one would expect some reduction in load capacity versus the between-pad loading (Boyd and Raimondi, 1953; Jones and Martin, 1979).
Figure 12(a) shows the flow of lubricant being sheared by two parallel surfaces. Since the lubricant adheres to both surfaces, it remains stationary on the upper surface and moves at the same velocity as the lower plate. For laminar flow, layers of lubricant move smoothly and the velocity profile is a straight line. In case of the convergent film within a bearing, Figure 12(b) shows that the actual lubricant flow is a little more complex. Nevertheless, the shearing type flow is still dominant unless the journal eccentricity is very high. This shearing motion creates frictional stresses between the lubricant layers. Per Newton, the fundamental relation for fluid friction (as a stress) takes the form = (du/dy). Using the parallel plate model, it can be simplified to = U/h. Thus, increasing lubricant viscosity or shaft speed increases the viscous shearing and, consequently, heat generation. This heat generation due to viscous shearing impacts a bearings performance in several ways:
Reduction in lubricant viscosity due to increased temperature Thermal growth and distortion of surrounding surfaces affecting
the film shape
Heating
of the lubricant and bearing materials toward their thermal failure limits Figure 13(a) demonstrates the influence of shaft speed on bearing temperature rise. For the two axial groove bearing, as the shaft accelerates from 1000 rpm to 19,000 rpm, the peak pad temperature substantially increases from 125F to 230F, which is near the failure limit. The operating viscosity is consequently reduced according to Table 1. Because of this viscosity reduction, speed increases are less and less effective in producing hydrodynamic pressure to lift the journal. This is apparent in Figure 9(a). Meanwhile, since heavier load results in smaller h on the loaded pad, the external load also affects pad temperature. However, as shown in Figure 13(b), its thermal influence is substantially weaker compared to the shaft speed.
Figure 11. Pressure Distributions with a Horizontal Load (W = 1100 LBF, N = 7000 RPM). Load capacity is of great concern in slow roll, turning gear operation with rotational speeds around 10 to 15 rpm. At such low speeds, the lubricant is unable to generate much supporting pressure, resulting in a very thin film likely in the regime of
161
(a) Lubricant Shearing Between Surfaces

Peak Pressure
Figure 14. Pressure Distribution Comparison, Variable Viscosity (W =1100 LBF, N = 7000 RPM). The increased temperature in the fluid film is the result of mechanical work done by the shaft. In turn, friction caused by the shaft shearing the lubricant produces a resistive torque on the shaft and consumes mechanical power. This friction loss is closely related to a bearings size, clearance, shaft speed, and oil viscosity. A bearings size dictates the area of shearing. Therefore, the partial arc design, which eliminates the top pad of a plain journal bearing, is often used to minimize friction loss (Byrne and Allaire, 1999). As shown in Figure 15, power loss grows with increasing shaft speed. And the partial arc bearing saves noticeable amounts of horsepower, especially at high speeds. In recent years, an industrial trend is to use directly lubricated bearings with reduced supply oil to decrease power loss. Similar to the partial arc design, this practice effectively reduces a bearings shearing area through starvation, which will be discussed later in the modeling section.
Shear u(y) Pressure Shear Pressure u(y) u=U
u=U
(b) Lubricant flows within a Bearing Convergent Wedge

Figure 12. Shearing Flows in Lubricant Film.
250 240 230 220 210 200 190 180 170 160 150 140 130 120 0 2 4 6 8 10 12 14 16 18 20
Maximum Pad Temperature Two Axial Groove Bearing, W=1,100 lbf
10 9 8 7 6 5
Friction Power Loss, W=1,100 lbf
250 240 230 220 210 200 190 180 170 160 150 140 130 120
Maximum Pad Temperature Two Axial Groove Bearing, N=7,000 rpm
2 Axial Groove Partial Arc
Tmax(Deg F)
Tmax(Deg F)
Shaft Speed (E+3 rpm)
1000
2000
Load (lbf)
3000
4000
4 3 2 1 0 0 2 4 6 8 10 12 14 16 18 20
(a)
(b)
Figure 13. Maximum Pad Temperature Versus Speed and Load. Considering the thermal effects on lubricant viscosity, Figure 14 presents pressures and journal positions of the same three bearings under the same operating condition. Compared to the isoviscous results in Figure 8(c), the pressure profiles do not show significant changes because the sums of the pressures must still equal the 1100 lbf applied load. The thermal effects on load capacity are most evidently shown by the new journal equilibrium positions. The two axial groove bearings eccentricity ratio has increased from 0.32 to 0.53, and the attitude angle also decreased by about 10 degrees. For the tilting pad bearing, the journal has moved vertically downward from a position of 0.5 eccentricity ratio to a 0.7 position. The journal position drop is the result of reduced load capacity because of viscosity reduction due to shearing heat generation. It can also be explained as the following: to generate the same force with a less viscous oil, the bottom pad needs to have a smaller clearance and larger wedge ratio hi/ho, which is achieved by the increased journal eccentricity. Power Loss Section summary:
Figure 15. Friction Power Loss Versus Shaft Speed. Petrov (1883) conducted pioneering work on viscous friction and proposed Petrovs Law, which is still used as a quick estimate for bearing power loss. He estimated the frictional torque according to:
Tfriction = 2R 2 LU c U = (2RL) R = (Surface Area )[Shear Stress]( Moment Arm) c
(6)
Mechanical
shearing.
energy is converted into heat through viscous
Shaft speed is the primary operating factor compared to load.
Power loss predictions using Petrovs law turn out to be liberal because the shaft is assumed centered (unloaded) within the bearing clearance. Since a bearing is always loaded statically and dynamically, it has more friction loss according to Figure 13(b) (temperature rise is the result and indicator of mechanical energy loss). Turbulent flow also increases friction loss due to additional eddy stresses.
162

Convergent Wedge Moving Surface Viscous Lubricant
Supply Flowrate Section summary:
Oil supply is necessary to maintain steady-state operation. Many older bearings work in a flooded condition. Many newer bearing designs use directed lubrication.
Another important static parameter is the oil supply flowrate that the bearing requires. Lubricant flows into a pad at its leading edge, exiting at its trailing edge and axial ends. The majority of the oil leaving the trailing edge enters the next pad and continues to circulate inside the bearing. However, the oil flowing axially leaves the bearing through the end seals and drains, and must be replenished by fresh oil from the lube system. Therefore, the supply flowrate needs to be at least equal to this side leakage rate. Using the previous two axial groove bearing as an example, Figure 16 plots the minimum required flowrate as functions of shaft speed and applied load. Since either higher speed or heavier load leads to stronger hydrodynamic pressure, the side leakage, driven by the film pressures, increases as a result. In practice, the supply flowrate is often larger than this minimum requirement to maintain the temperature rise between the oil supply and drain within recommended limits (typically between 40 to 60F). However, this is only helpful in reducing the mixed sump temperature in a flooded bearing. Methods to determine the supply flowrate and inlet orifice size can be found in Nicholas (1994).
4
Hydrodynamic Force
Viscous Shearing
Temperature Rise
Power Loss
Deformations
Viscosity Reduction
Figure 17. Basic Working Principles Within a Fluid Film Bearing. Dynamic Performance Section summary:
A bearing is a component of an integrated dynamic system. A bearing can be dynamically represented as springs and
dampers in a linearized model.
Stiffness
load.
and damping coefficients have significant rotordynamic implications.
Dynamic coefficients are dependent on shaft speed and applied There are two types of instability related to bearings: oil whirl
and shaft whip. Desirable steady-state operation, where the bearing is running with sufficient load capacity and acceptable temperatures, helps to ensure the long-term reliability of the bearing itself. However, the bearings dynamic properties must also be acceptable for the overall machines reliability. This is because a bearings dynamic properties, in conjunction with dynamics of the rest of the rotor system, govern all aspects of a machines vibrational performance. The dynamic performance of journal bearings first came under scrutiny because of the vibration problems encountered by Newkirk and Taylor (1925). In this landmark paper, Newkirk and Taylor describe the first published account of a rotor going unstable due to oil whip. Initially, they thought the vibration was caused by improper shrink fits, which were the only known source of whipping instability (Newkirk, 1924). They eventually found that the bearings parameters such as clearance (Figure 18), loading, alignment, and oil supply (in some tests, the supply was cut off!!) controlled the instability.
Minimum Required Supply Flow (Side Leakage) Two Axial Groove Bearing, W=1,100 lbf
1.8
Minimum Required Supply Flow (gpm)
Minimum Required Supply Flow (gpm)
3.5
1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0
2.5
1.5
Minimum Required Supply Flow (Side Leakage) Two Axial Groove Bearing, N=7,000 rpm
0.5
5000
10000
15000
20000
1000
2000
3000
4000
Rotational Speed (rpm)
(a)
Total Load (lbf)
(b)
Figure 16. Minimum Required Supply Flow Versus Speed and Load. One should also bear in mind that lubricant is dragged into the bearing clearance by shaft rotation, not pumped into it by high supply pressure. Therefore, the function of the oil pump is simply to send enough oil into the bearing and keep it circulating. Furthermore, the bearing clearance will accept only a finite quantity of oil; supply less and the bearing will be starved, supply more and the extra oil will fall to the sides and be wasted. The amount of oil that the bearing clearance requires is simply a function of the shaft speed, clearance, and minimum film thickness. Again, supplying more oil than the bearing clearance can consume will not effectively reduce pad temperatures and is simply a waste of oil flow. The basic working principles governing the steady-state operation of fluid film journal bearings are summarized in Figure 17. A convergent wedge, a moving surface, and viscous lubricant are the three ingredients necessary to generate the film hydrodynamic forces to support the applied load. An accompanying phenomenon is viscous shearing that causes temperature rise and power loss. The temperature rise and power loss are related because the energy used to heat up the film is converted from the shaft mechanical energy. Increased temperatures lead to oil viscosity reduction and bearing deformation. In turn, the deformations also change the bearing geometry and, thus, the wedge shape.
Figure 18. Newkirk and Taylors Oil Whip Measurements. (Courtesy Newkirk and Taylor, 1925) With the considerable development of steam turbine technology, the 1920s continued to provide evidence that a machines vibration was heavily linked to the operation of the bearings. In two papers, Stodola (1925) and one of his pupils, Hummel (1926), introduced
163
the concept that a bearings oil film dynamically acts like a spring. They found that, when this oil springs stiffness was considered, their rotor critical speed calculations could be improved (Lund, 1987). They also realized that the oil film stiffness could be very nonlinear, i.e., the film force variation was not directly proportional to the journal position variation. While many continue to study this nonlinear complication, most of the machines in operation today were designed using linearized stiffness and damping properties for the oil film. As shown in Figure 19, the fluid film can be represented by springs and dampers. The static load W, such as gravity, establishes the journals steady-state equilibrium position. Then, some dynamic force, such as rotor unbalance forces, pushes the journal away from its equilibrium and causes it to whirl on an elliptical orbit (Figure 20). To have an acceptable vibration level, the orbits size must be relatively small compared to the bearing clearance. When this is the situation, the vibration is said to be in the linear range and the film dynamic forces are directly proportional to the displacements (x, . . y) and associated velocities (x, y). This relationship is given by:
where the Kij and Cij are called linearized stiffness and damping coefficients, respectively. In other words, at an instantaneous journal position, the horizontal and vertical forces due to the oil film can be obtained by expanding Equation (7):
( ) Fy = (K yy y + K yx x + Cyy y + Cyx x )
Fx = Kxx x + Kxy y + Cxx x + Cxy y
(8)
Kxx Fx F = K yx y
Kxy x Cxx C K yy y yx
Cxy x Cyy y
(7)
where the negative sign implies that the force is acting on the rotor. To justify the use of linearized dynamic properties, one should understand the other situation where the vibration levels are relatively large. Figure 21 shows the classic example of an unstable shaft where the orbit nearly fills up the entire bearing clearance. Here the rotor has almost reached the so-called limit cycle. Since this motion is large relative to the bearing clearance, linearized coefficients are inadequate to represent the film dynamic forces. Therefore, they cannot be used to predict the actual amplitudes for such large vibrations. However, the strength of the linearized coefficients is their ability to predict whether or not such unstable vibrations will occur. This ability, combined with their accuracy in predicting vibration amplitudes within the range of interest [up to 40 percent of the clearance according to Lund (1987)], enables modern rotordynamics to be firmly based on their use.
Figure 21. Unstable Rotor Exhibiting Large Vibrations. Now, let us define and explain those linearized dynamic coefficients in Equations (7) and (8). The following two stiffness coefficients are called principal or direct coefficients:
Figure 19. Dynamic Properties of the Fluid Film.
Kxx = Fx / x
Horizontal Pr incipal Stiffness
K yy = Fy / y Vertical Pr incipal Stiffness
(9)
where each relates the change in force in one direction due to a displacement in the same direction. In other words, these direct stiffnesses provide a restoring force that pushes the journal back toward its steady-state equilibrium position. As shown in Figure 22(a), a positive horizontal perturbation x generates a negative horizontal force Fxx = Kxx(x), a negative vertical perturbation y yields a positive vertical force Fyy = Kyy(y). The combination is a radial force that tries to push the journal back to Os.
OS y
x Kxx x FK Kyy y
OS
y
OS y
C yy y
FCCK Kyx x Kxy y
FC
C xx x
a) DIRECT STIFFNESS FORCE
b) DIRECT DAMPING FORCE
c) CROSS COUPLED STIFFNESS FORCE
Figure 20. Journal Steady-State Position and Orbit.
Figure 22. Dynamic Forces in the Fluid Film.
164
The principal stiffnesses are extremely important with respect to the machines vibration performance. Their magnitude, relative to the shaft stiffness, governs the location and amplification of the rotors critical speeds. They are equally important for stability purposes. Large asymmetry of Kxx and Kyy is the main cause for split critical speeds (API Standard 684, 2005) and noncircular (elliptical) orbit shapes. Although such asymmetry can be very beneficial with respect to stability (Nicholas, et al., 1978), symmetry of these coefficients is usually preferred for unbalance response considerations. Two principal or direct damping coefficients are also present:
Cxx = Fx / x
Horizontal Pr incipal Damping
Cyy = Fy / y Vertical Pr incipal Damping
(10)
Here the damping coefficients relate the change in force due to a small change in velocity. Because of the rotors whirling motion, the combination of these two principal damping coefficients produces a force that is tangential to the vibration orbit. Furthermore, as shown in Figure 22(b), this direct damping force acts against the whirling motion, helping to retard or slow it. Like their direct stiffness counterparts, the principal damping terms dictate much about the machines unbalance response and stability. They are often the predominant source of damping in the entire machine. However, their effectiveness in reducing critical speed amplification factors and preventing subsynchronous instabilities is determined also by the bearings direct stiffness coefficients as well as the shaft stiffness. For the direct damping to be effective, the bearing cannot be overly stiff because the damping force relies on journal motion. Also, contrary to ones initial instincts, large amounts of damping can actually be detrimental. Barrett, et al. (1978), in an important fundamental paper, highlighted this fact and verified that the optimum amount of bearing damping is a function of the bearing (direct) and shaft stiffnesses. The off-diagonal stiffness coefficients in Equation (7), Kxy and Kyx, are the infamous cross-coupled stiffness coefficients. The meaning of cross coupled becomes apparent when these stiffnesses are defined as:
Unlike their fixed geometry counterparts, tilting pad bearings produce very little cross-coupled stiffness, which explains their popularity. This fact was actually touched on earlier in Figures 8 and 9 where the tilting pad bearings journal position moved only vertically under a vertical load. It consistently maintains a small attitude angle, indicating a small amount of cross-coupled stiffness present. As shown in Figure 9, the attitude angle of the two fixed geometry bearings approaches 90 degrees at light loads, implying the cross-coupled stiffnesses are very large relative to their direct counterparts. Thus, instability is often encountered when running fixed geometry bearings at light loads. Like the static performance, the dynamic coefficients vary with shaft speed and external load. Figure 23 shows the two axial groove bearings coefficients as functions of the shaft speed. The vertical stiffness Kyy and damping Cyy decrease significantly as the speed increases from 1000 to 10,000 rpm. Meanwhile, the bearing becomes less stable because it loses considerable damping while retaining a high cross-coupled stiffness. Therefore, an unbalance response analysis must include the coefficients variations for accurate prediction of critical speeds and vibration levels. Such speed dependency is also the reason why the amplification factor of the first critical speed should not be used as a measure of stability for higher speeds like maximum continuous speed.
2.5E+06
Two Axial Groove Bearing Constant Load: 1100 lbf at 270 deg
X
30000
Two Axial Groove Bearing Constant Load: 1100 lbf at 270 deg
2E+06
Damping (lbf-sec/in)
Kxx Kxy -Kyx Kyy
25000
X
Stiffness (lbf/in)
20000
Cxx -Cxy = -Cyx Cyy
1.5E+06
15000
1E+06
10000
500000 5000
X X
5000
10000
15000
20000
5000
10000
15000
20000
Shaft Speed (rpm)
Shaft Speed (rpm)
Figure 23. Two Axial Groove Bearing Dynamic Coefficients Versus Speed. The stabilities of the plain journal and pressure dam bearings are compared in Figure 24. The tilting pad bearing is not presented because its stability is not an issue. Here, stability is measured by the rigid rotor threshold speed, which excludes the rotor effects and is solely dependent on the bearing properties and loading (Lund and Saibel, 1967). As shown in Figure 24, the plain bearing is predicted to be unstable at around 10,000 rpm using the bearing coefficients at 700 rpm; using the coefficients at 10,000 rpm, instability is predicted at 7500 rpm. Therefore, a rigid rotor would go unstable at 7600 rpm where the curve intersects the 1 line. The pressure dam bearing shows improved stability since the intersection is beyond 10,000 rpm. Experimental results and more discussions can be found in Lanes, et al. (1981), and Zuck and Flack (1986).
20 18
Kxy = Fx / y K yx = Fy / x
(11)
As an example, the coefficient Kyx relates a vertical force due to a horizontal displacement. Thus, the horizontal and vertical directions have become coupled. This exactly corresponds to the behavior highlighted in Figure 8 for the two fixed geometry bearing, where a displacement in one direction resulted in force component perpendicular to this displacement. Almost all structures have such cross-coupled stiffness terms but most are symmetric in nature where Kxy = Kyx. Rotor systems are unique in that this symmetry usually does not exist (Kxy Kyx and usually Kxy > 0, Kyx < 0). Fundamentally, their presence and their asymmetry result from the various fluids rotating within a turbomachine, such as oil in bearings and gas in labyrinth seals. Figure 22(c) illustrates why asymmetric cross-coupled stiffnesses are detrimental. Instead of opposing the rotors whirling motion like the direct damping, the cross-coupled stiffnesses combine to create a force pointing in the whirl direction, promoting the shaft vibration. When the direct damping force is unable to dissipate the energy injected by the cross-coupled stiffness force, the natural frequency (typically the lowest one with forward whirling direction) will become unstable, causing the shaft to whirl at this frequency (Ehrich and Childs, 1984). This frequency will appear in the vibration spectrum, typically as a subsynchronous component. Such self-excited vibration is the reason why these cross-coupling coefficients are of such concern for stability purposes.
Threshold Speed(E+3 rpm)
16 14 12 10 8 6 4 2 0 0 2 4 6 8 10
Two Axial Groove 7600 rpm Pressure Dam
1x
Figure 24. Rigid Rotor Stability Threshold Speeds Versus Shaft Speed.
165
Examining all the ways a bearings dynamic properties can influence a machines rotordynamics is beyond the scope of this tutorial. The literature on the subject is extensive and the reader is encouraged to examine API Standard 684 (2005) for further explanation and references. Fundamentally, a machines rotordynamic performance becomes an interplay of how the dynamics of various components (bearings, shaft, seals, supports, etc.) interact when combined together as a system. In other words, the overall dynamic performance is governed by the system, not one particular component in general. Oil whirl is one exception where the bearings dynamic properties dominate the rotordynamic behavior of the system. First observed by Newkirk and Taylor (1925) who called it journal whirl, this instability phenomenon has received considerable interest even though its occurrence is rare in most machinery applications. Some exceptions are gearboxes and internally geared compressors operating at low power and resulting in small gear forces. As shown in Figure 25, the frequency of the systems first forward mode follows the 0.5 line at low shaft speeds. When the system becomes unstable at such low speed, the frequency of the subsynchronous vibration equals half of the running speed and tracks it as the rotor accelerates. Since the shaft does not experience much bending, it can be regarded as a rigid body or just a mass inside the bearing. Therefore, the oil whirl instability is dominated by the dynamic properties of the bearing. With increasing shaft speed, the unstable mode steps into the territory of shaft whip where the subsynchronous frequency is locked at a constant value. Unlike oil whirl, the rotors mode shape in whip undergoes noticeable bending and its flexibility plays a significant role in the systems overall dynamics.
6000 5000 4000 3000 2000
F F F
1x 0.5x
vertical rotor (1925), the use of some centering device (Muszynska and Bently, 1995), misalignment, overhung mass effects, or the presence of external forces from gearing or partial arc steam admission forces that can negate the gravity loading. Again, oil whirl is driven by large bearing cross coupling, and thus occurs only with fixed geometry bearings.
MODELING
Accurate evaluation of a bearings performance has become a vital factor in the design, operation, and troubleshooting of rotating machinery. A number of computer programs have been developed to accomplish this task. This section presents the major aspects of bearing modeling, the mainstream techniques used in those computer codes as well as their potential limitations. First, the general areas that are required in most modern bearing analysis will be covered. Such areas include:
Hydrodynamic pressure Temperature Deformations Turbulence Dynamic coefficients

The next step is to assemble those components into a functional computer algorithm. Then, discussion will switch to special situations such as direct lubrication and starvation. Some modeling difficulties and challenges will be addressed at the end. The objective is to shed some light on those computer tools, and, thus, help engineers to use them properly. Hydrodynamic Pressure Section summary:
Hydrodynamic pressure is the primary physical phenomenon to

model.
The
B B
Reynolds equation is the governing equation for thin lubricant film. Hydrodynamic pressure modeling is the foundation of an accurate bearing analysis. In general fluid dynamics, the films pressure, and velocity distributions are governed by the coupled continuity equation and the momentum equations. The continuity equation comes from the basic law of mass conservation. Each of the three momentum equations, known as Navier-Stokes equations for incompressible flow, is essentially Newtons second law in each direction of the three-dimensional space. Thus, a simple theoretical analysis requires simultaneous solution of four equations, which is not trivial because iterations must be employed and the momentum equations are nonlinear. If other parameters such as temperature and turbulence are considered, more equations must be added to the formulation and the solution procedure quickly becomes very complex. Fortunately, such a procedure can be avoided in a bearing analysis and the hydrodynamic pressure can be directly calculated from the following linear equation.
F F
F B
1000
F
Oil Whirl
Shaft Whip
2000
4000
Shaft Speed (rpm)
6000
8000
10000 12000 14000
h3 p h3 p U h + = x x 12 x z 12 z 2

Figure 25. Campbell Diagram Showing Oil Whirl and Shaft Whip. In real life, most unstable machines exhibit shaft whip directly without exhibiting oil whirl behavior. To produce the 0.5 oil whirl, the bearing must be unloaded, allowing it to operate at very low eccentricity. Hamrock (1994) theoretically deduced that oil whirl would occur if the bearing had a constant pressure (zero) throughout the film. Obviously, a constant (zero) pressure can only be achieved with a centered or unloaded shaft. The unloading may be caused by the lack of gravity load like Newkirk and Taylors

x Pr essure
z Pr essure
(12)
Shear
Equation (12) is the classic Reynolds equation (Reynolds, 1886). During his derivation, Reynolds had to make several assumptions. Most of all, he utilized the fact that the film thickness is much smaller than the bearings diameter (the typical c/D ratio is on the order of 103). Consequently, the momentum equations can be significantly simplified by neglecting the small terms. From these simplified equations, the pressure across the film is shown to be constant and the velocity components can be directly solved.
166
Then, the Reynolds equation is obtained by substituting the velocity components into the continuity equation and integrating across the film (Szeri, 1979). This classic Reynolds equation also assumes that viscosity is invariant across the film and the flow is laminar. However, Equation (12) can be made more general by relaxing these two conditions (Constantinescu, 1959; Dowson, 1962). The left hand side of Equation (12) includes the pressure flow terms that represent the net flowrates due to pressure gradients within the lubricant area; the right hand side is the shear flow term that describes the net entraining flowrate due to the surface velocity. If the shaft is stationary (U = 0), the right hand side equals zero and no lubricant can enter the bearing clearance. The pressure gradients on the left hand side must be zero to satisfy the flow continuity. Similarly, if the pairing surfaces are parallel (h/x = 0), the right hand side also becomes zero and no hydrodynamic pressure can be developed. Therefore, it is shown from the Reynolds equation that the rotating shaft and convergent wedge are the necessary conditions to generate hydrodynamic pressure. If the clearance is divergent (h/x > 0), the Reynolds equation will give artificially negative pressure. However, since the lubricant cannot expand to fill the increasing space, cavitation occurs in this region and the divergent clearance is occupied by streamlets and vaporliquid mixture (Heshmat, 1991). Figure 26 shows a two axial groove bearing and the bottom pad pressure distribution solved from the Reynolds equation. Hydrodynamic pressure is smoothly developed in the area of convergent clearance. Axially, the pressure distribution is symmetric about the midplane and goes down to the ambient pressure at the edges. No hydrodynamic pressure is generated in the cavitated region that exists near the trailing edge of the bottom pad and on the entire top pad.
to the momentum equations, the energy equation for bearing analysis has been substantially simplified because of small film thickness. The three-dimensional energy equation for laminar flow is usually written in the form of:
T T T C p u +v +w = y z x

Convection
u 2 w 2 T T T + k k + k + + x x y y z z y y

Conduction Dissipation
(13)
As shown in Equation (13), the steady-state temperature is determined by three terms. The dissipation term describes the internal heat generation due to viscous shearing. As would be expected, the heating intensity is shown to be proportional to the lubricant viscosity. The heat convection term describes the rate of heat transfer due to the lubricants motion. And the conduction term determines the heat transfer between the lubricant and surrounding surfaces. It can be shown by dimensional analysis that the heat convection term is usually much larger than the conduction term. Thus, the film physically constitutes a heat source; while some of that heat is conducted away through the solid surfaces, the majority of it is carried away by the flowing lubricant. To achieve better computational efficiency, two simplified forms of Equation (13) are often used in practice. The first one is the adiabatic equation that is obtained by neglecting the conduction term in Equation (13). The adiabatic energy equation was derived by Cope in 1949 and has been widely used for a long time. It implies that no heat is transferred to the solids and the film temperature is constant radially. Figure 27 shows the typical adiabatic temperature solution for a smooth pad that has convergentdivergent clearance. Most of the temperature rise takes place in the convergent clearance section where significant viscous shearing occurs. In the divergent region, the temperature rise is significantly reduced due to weak heat generation in the vapor-liquid mixture. Axially, the temperature is almost invariant, showing only slight increase at the edges.
Figure 26. Pressure Distribution on the Bottom Pad of a Two-Axial Groove Bearing. Temperature Section summary:
ent D iverg
Including temperature effects is critical for accurate bearing performance predictions.
z (Axia
l)
l) rentia cumfe x (Cir
The modeling involves shaft, fluid film, and bearing pads. Energy equation is the governing equation.
One early idea to model the thermal effects is the approach of effective viscosity (Raimondi and Boyd, 1958). This method employs an empirical equation to calculate an effective temperature. From the effective temperature, an effective viscosity is determined and used in the Reynolds equation. While this simple idea recognizes the viscosity reduction due to temperature rise, its effectiveness is very limited and it fails to give the maximum pad temperature, which is an important operation parameter. To accurately model the thermal effects, the temperature distribution must be solved from the governing energy equation. Similar
Figure 27. Adiabatic Temperature Solution for a ConvergentDivergent Film. For many years, bearing designs had used adiabatic theory and isoviscous theory to bracket a bearings actual performance. However, this notion was later invalidated by a number of studies. It became clear in the 1960s that the radial temperature variation must be taken into account for accurate bearing modeling (McCallion, et al., 1970; Seireg and Ezzat, 1973; Dowson and Hudson, 1963). Moreover, in such a situation as a steam turbine where the hot shaft conducts heat into the film, the adiabatic assumption is clearly inappropriate. Therefore, another form of the
167
simplified energy equation, which includes the radial heat conduction, has become more popular in modern bearing analysis. Since the temperature varies little in the axial direction, as shown in Figure 27, the axial heat transfer can be eliminated (T/z = 0) and the original three-dimensional energy equation is reduced to twodimensional. As shown in Figure 28, this two-dimensional equation solves the temperature on the x-y plane, which is perpendicular to that of the adiabatic energy equation and the Reynolds equation.
Mesh for the 2D Energy Equation with Conduction
Mesh for the Reynolds Equation & the Adiabatic Energy Equation
Fluid
Solid
Figure 28. Numerical Meshes for the Governing Equations. Figure 29 presents the temperature contour obtained from this two-dimensional energy equation. The lower rectangular section is the bearing pad and the upper section is the fluid film. For better visualization, the film thickness is enlarged 103 times and the padfilm interface is highlighted by a bold line. The convergentdivergent clearance is clearly shown on the upper boundary. In the circumferential direction, the film and pad temperatures increase from the leading edge, arrive at the maximum value around the minimum film thickness location, and decrease near the trailing edge as the result of heat conduction. The radial temperature variation is shown to be significant with a hot spot close to the pad surface. This trend is generally true for most bearings regardless of the specific contour values in this example.
100 90 80 70 60 50 40 30 20 10 0 -10
92
76
Most theoretical algorithms solve one form of the energy equation or another. Usually, the Reynolds equation is solved first using assumed lubricant viscosity. After the pressure distribution is obtained, the velocity components can be derived and the energy equation is solved to give the temperature distribution. Then, the viscosity is recalculated and the Reynolds equation is solved again using the updated viscosity. This procedure continues till the difference between two consecutive iterations is sufficiently small. One may have noticed that the energy equation does not directly govern the temperature in the solid pad. One way to obtain the pad temperature is to solve a separate heat conduction equation. Since the temperature and heat flux must be continuous at the film-pad interface, the heat conduction equation is coupled with the energy equation and they can be solved through iterations. The second approach is to extend the energy equation into the solid pad. In fluid film, the energy equation has the convection, conduction, and dissipation terms. In solid pad, convection and dissipation terms are set to zero, leaving the heat conduction equation. In order to satisfy the heat flux continuity, harmonic averaging is employed to modify the heat conductivity on the film-pad interface (Paranjpe and Han, 1994). The thermal effects on the predicted bearing performance are demonstrated through the example of a two axial groove bearing reported in Fitzgerald and Neal (1992). All thermohydrodynamic (THD) results are calculated using the two-dimensional energy equation including heat conduction. Figure 30 compares the pad surface temperatures along the axial centerline. The theoretical results have close agreement with the experimental data. Figure 31 shows the journal eccentricity ratio under various loads and speeds. The THD analysis consistently gives more accurate results compared to the isoviscous hydrodynamic (HD) analysis, especially in the case of 8000 rpm that has relatively high temperature rise. The predicted vertical stiffness coefficients Kyy are plotted in Figure 32. The difference due to the inclusion of the thermal effects can be as much as 30 percent at high speeds.
100
Load
90
Temperature(C)
70
79
80
Figure 29. Temperature Contour from the 2-D Energy Equation with Conduction. In some situations, neither radial nor axial temperature profile can be assumed constant and the full three-dimensional energy equation must be solved. For example, in a pressure dam bearing, the temperature inside the pocket is much lower than that in the land regions (He, et al., 2004). Or if the bearing is misaligned with respect to the shaft, the temperature is higher at one axial edge where the film thickness is minimum.
76
7066
73
70
63
60
73
66
79
82
85
89
94
92
Film
85
50
90
Angular Location (deg)
180
270
360
Pad
76
89
60
Circumferential Direction
50
100
150
Figure 30. Pad Surface Temperature Comparison, L/D=0.5, N=8000 RPM, W=5.43 kN. Deformations Section summary:
63
Elasticity modeling has become more and more important as a result of increasing high speed, heavy load applications. It involves deformations of bearing pad, pivot, shaft, and shell. Theoretical models require caution in use.
Deformations change a bearing operating geometry, and, consequently, affect all aspects of a bearing performance. Because of its flexible assembly, a tilting pad under high speed and/or heavy load
168
1 0.9 0.8
because the temperature rise is not uniform and the pivot constrains the deformations near the pad center. In this particular example, the mechanical load had little effect on the preload while the thermal deformation increases it.
Eccentricity Ratio
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0 2 4 6 Exp., 8000 rpm Exp., 2000 rpm THD, 8000 rpm THD, 2000 rpm HD, 8000 rpm HD, 2000 rpm 8 10
Cb (in) Cp (in) m Original 0.00310 0.00585 0.47 Deformed 0.00312 0.00589 0.47
Pivot (a) Mechanical Deformation
Load (KN)
Cb (in) Cp (in) m
Original 0.00310 0.00585 0.47
Deformed 0.00274 0.00571 0.52
Figure 31. Thermal Effects on the Predicted Eccentricity Ratio.

14 12 10 8 6 4 2 0
W=9.43 kN
THD HD
Pivot (b) Mechanical and Thermal Deformations
Figure 33. Pad Deformations Obtained by 2-D Finite Element Method. In addition to the pad deformations, the journal and bearing shell also experience thermal growth in operation. Due to the rotation, the journal temperature is usually assumed constant and its deformation is modeled as free thermal growth at uniform temperature (Kim, et al., 1994). The bearing outer shell can also be modeled in a similar fashion. In addition, the pivot deformation under heavy mechanical load can be calculated from the Hertian contact theory (Kirk and Reedy, 1988; Nicholas and Wygant, 1995). Figure 34 presents the temperature predictions of a four-pad tilting pad bearing experimentally investigated by Fillon, et al. (1992). The thermoelastohydrodynamic (TEHD) analysis takes into account both the thermal and elastic effects. As shown in this figure, the inclusion of deformations brings the theoretical results closer to the experimental data. While this example shows that the inclusion of elasticity can improve the predictions, the deformation models, especially the journal and shell models, must be used very carefully. In fact, it is often inadequate to model the journal and shell thermal growth as free expansions. Since the journal is part of the entire shaft, its growth is not free and its proper modeling requires the knowledge of the entire shaft temperature distribution. Meanwhile, the bearing shell also cannot expand freely because it is constrained by the bearing housing. Its deformation is significantly affected by the housing conditions, including its stiffness, temperature, and shrink fit interference. A poor evaluation of the journal and shell deformations can introduce very large errors in the modeling predictions. Although the physics seem straightforward, accurate modeling of elasticity is one of the most difficult tasks in a bearing analysis. Turbulence Section summary:
Kyy(E+8 N/m)
Speed(RPM)
5000
10000
Figure 32. Thermal Effects on the Predicted Direct Stiffness. is subject to deformations that are composed of two parts: mechanical deformation due to pressure and thermal deformation due to temperature rise. The simplest elastic model treats a tilting pad as a one-dimensional curved beam (Ettles, 1980; Lund and Pedersen, 1987). If the deformed pad is assumed circular, the clearance variation C can be calculated and the bearing is modeled with a modified clearance c = co + c. Alternatively, the beam equation can be numerically integrated and the nodal displacements are used to correct the film thickness. A more advanced approach is to formulate the problem based on the principle of virtual work and solve it using the finite differences or finite element method (Brugier and Pascal, 1989; Desbordes, et al., 1994). Although the actual pad is three-dimensional, two-dimensional plain strain approximation is often used since the deformations are primarily on the x-y plane. Figure 33(a) shows the shape of a tilting pad under mechanical deformation. The finite element grid before deformation is plotted with the dashed lines and the deformed pad is plotted with the solid lines. The mechanical deformation is shown to be mainly in the radial direction. Since the displacements around the pivot are smaller than those near the ends, mechanical deformation effectively increases more cp than cb. Consequently, the pad preload, m, may increase or decrease depending on the relative cp and cb variations. Figure 33(b) shows the pad deformations under both mechanical and thermal loads. Since the thermal deformation is dominant in this example, the total deformation is shown as largely thermal growth with decreased cb and cp. The pad preload also varies
Turbulent
bearings have different behaviors compared to laminar bearings.
The turbulence effects must be included in modeling.

Two different types of flow may exist in fluid film bearings: laminar and turbulent. In laminar flow, the fluid particles are moving in layers with one layer gliding smoothly over the adjacent layers. In turbulent flow, the fluid particles have irregular motion and the flow properties, such as pressure and velocity, show erratic fluctuations with time and with position. Since it is impossible to track the instantaneous flow properties, their statistical mean values

Experiment THD Analysis TEHD Analysis
Load
169
600
90 85 80 75 70 65 60 55 50 45
Tmax
Turbulent Theory Laminar Theory
90 85 80
500
Temperature(C)
75 70 65 60 55 50 45 40 0 90 180 270 360
400
300
200
40 35 30 0
HP Loss
Laminar Theory
100
Figure 34. Elastic Effects on the Pad Temperature Calculations. are sought in a turbulent flow calculation. The flow regime is usually indicated by the Reynolds number defined as Re = Uh/. The flow is laminar at low Reynolds numbers. As Re increases, the flow becomes unstable and partially turbulent, and eventually evolves into full turbulence at high Re. From the definition of Re, one can deduce that turbulence is likely to occur in large bearings due to high surface velocity and relatively large clearance. A turbulent bearing exhibits increased power consumption along with a sharp change of bearing eccentricity (Wilcock, 1950). Compared to laminar flow, turbulent flow has increased stress due to the fluctuating motion. Since stress is proportional to viscosity, turbulent flow can be treated as laminar flow with increased effective viscosity, which is defined as the superposition of turbulent (eddy) viscosity and the actual viscosity of the lubricant. Thus, the Reynolds equation can be extended into the turbulent regime using effective viscosity values. Several models have been developed to evaluate the eddy viscosity. The models in Constantinescu (1959), Ng and Pan (1965), Elrod and Ng (1967), Safar and Szeri (1974) are similar in that they all utilize the law of wall in which the eddy viscosity is assumed as a function of wall shear stress and distance away from the wall. On the wall surface, the flow is laminar and there is no eddy viscosity contribution. The eddy viscosity increases as the position moves from the wall to the core of the film. The specific formula used to quantify the eddy viscosity is different in those references. A distinct alternative is the bulk flow theory developed by Hirs (1973). Ignoring the detailed turbulence structure, his theory directly correlates the wall shear stress with the mean flow parameters using an empirical drag law. In addition, the fluctuating motion also enhances the heat transfer across the film. Analogous to the effective viscosity, an effective heat conductivity can be defined and employed to generalize the energy equation into turbulent flow regime. Figure 35 shows maximum temperature and power loss as functions of shaft rotational speed. The experimental data represented by the discrete symbols are taken from Taniguchi, et al. (1990). As shown in this figure, the Tmax curve has a shift that corresponds to the flow regime transition: when the flow is laminar, Tmax increases smoothly with the increasing shaft speed; Tmax stays flat or even shows slight decrease during the flow regime transition; Tmax resumes smooth increase after the transition is completed. Due to the increased effective heat conductivity, the analysis including turbulent effects yields lower and more accurate Tmax predictions. The inclusion of turbulence also significantly improves the friction loss prediction. Bouard, et al. (1996), compared three popular turbulent models: the Ng and Pan model, the Elrod and Ng model, and the Constantinescu model. They concluded that, if a bearing is turbulent, the turbulent effects must be taken into account and these three models gave similar results.
1000
Rotational Speed (rpm)
2000
3000
4000
0 5000
Figure 35 Comparisons of the Results from Turbulent and Laminar Theories. Dynamic Coefficients Section summary:
The reduced coefficients of a tilting pad bearing are dependent

on the shafts precession or whirl frequency. The bearing dynamic coefficients can be calculated by numerically perturbing the journal position or by solving the perturbed Reynolds equations. The first approach is straightforward. After establishing the steady-state journal position, the hydrodynamic force is calculated at a slightly different position. Since the force is somewhat different at this new journal position, the force variation due to the small displacement is obtained and a stiffness coefficient is easily calculated from the definition of F/x. A damping coefficient is similarly calculated with a velocity perturbation. The second approach involves more mathematics because the perturbed Reynolds equations must be theoretically derived. Then, the dynamic coefficients are obtained by directly integrating the pressure solutions from those perturbed equations. A fixed geometry bearing has eight dynamic coefficients because such journal-bearing system has only two degrees of freedom (the journal translation in X and Y). However, the dynamic system of a tilting pad bearing has more degrees of freedom because the pads can rotate. These extra degrees of freedom lead to additional dynamic coefficients that are related to the pads tilting motion. For example, a five-pad tilting pad bearing has 58 dynamic coefficients. In practice, it is convenient to reduce these coefficients to eight equivalent ones that are related to journals X and Y motions [Equation (7)]. This procedure is called dynamic reduction or dynamic condensation. As shown in Figure 36, the reduced coefficients are not constants, but dependent on the frequency of the shaft whirl, which means the shaft perceives different bearing stiffness and damping at each vibration frequency. A widely debated topic, more discussions on this frequency dependency can be found in Lund (1964), Parsell, et al. (1983), and API Standard 684 (2005). The Coupled Algorithm Figure 37 shows the structure of a comprehensive thermoelastohydrodynamic algorithm that assembles the various models discussed above. The basic block is the classic hydrodynamic analysis. Since the film thickness h is required in the Reynolds equation, the journal operating position must be known in order to calculate the hydrodynamic pressure. However, we only know that the journal is operating at equilibrium where the resulting
170

5 Pad, Load Between Pad Center Offset, L/D = 0.75 0.3 Preload, S = 2.2
Rotor Speed = 10,000 rpm

220
2.0E+05
1.8E+05
Kyy
1.6E+05
Cyy
200
180
Kxx
1.4E+05
Cxx
160
1.2E+05
140
1.0E+05
120
film. Thus, it is reasonable to assume that the journal acquires the average film temperature and is constant. According to this model, heat flows into the journal in the hot sections, dumped back into the fluid film in the cool sections, and the journal is adiabatic in a bulk sense. The pad inlet temperature is determined in the preceding oil groove (Heshmat and Pinkus, 1986). As shown in Figure 38, two streams of lubricant are mixed in the groove: cool lubricant from the supply line and hot lubricant carried over from the previous pad. Therefore, at the pad inlet, the lubricant temperature is at some mixing value, which can be calculated by applying energy conservation to the groove control volume. Including various deformation models, the THD block can be further extended to a complex thermoelastohydrodynamic analysis. It should be pointed out that the structure shown in Figure 37 is not unique. People have used a variety of structures to achieve the same objective. However, regardless of the specific structure, a computation always begins with a group of assumed initial values, and ends after convergence has been reached for every iteration loop.
8.0E+04
2000
4000
6000
8000
10000
100 12000
Whirl Frequency (cpm)
Figure 36. Frequency Dependent Stiffness and Damping, Tilting Pad Bearing. hydrodynamic force balances the external load. Therefore, the Reynolds equation is initially solved with assumed journal position and the actual position is searched through iterations. If a pad can tilt, its tilt angle also needs to be iteratively determined using the fact that, at equilibrium, the moment about the pivot must be zero.
Initial Values
Figure 38. Mixing in an Oil Groove.

Pressure
Special Situations In some special applications, the TEHD models presented above are not sufficient to predict a bearings properties. Those special situations require additional modeling efforts in order to achieve satisfactory theoretical predictions. However, these special situations are often not modeled although they should. Direct Lubrication
Pad Tilt Angle
HD
Journal Position
Film & Pad Temperatures
Journal Temperature
THD
Inlet Temperature
TEHD
Deformations
Output
Figure 37. Structure of a Sample TEHD Algorithm. From the HD block, the algorithm can be expanded to a higher level that includes the thermal effects on the lubricant viscosity. As mentioned earlier, the energy equation must be added to the formulation and iteratively solved with the HD block. In addition, the journal and pad inlet temperatures also need to be calculated as important boundary conditions. During a revolution, a point on the journal surface travels across the hot and cool sections of the fluid
In recent years, exceedingly high pad temperature has become an increasing problem in rotating machinery operations. One solution to this problem is the use of direct lubrication designs, such as the inlet pocket and spray bar. As suggested by the name, the idea is to directly supply cool oil into the pad clearance and block hot oil carryover from the previous pad. According to Figure 38, more Qsupply and less Qout will lead to lower mixing temperature Tin, and consequently, lower temperature on the ensuing pad. Such direct lubrication designs have been successfully used and are gaining popularity in industry (Edney, et al., 1998). Following this idea, Brockwell, et al. (1994), developed a new groove mixing model assuming all cool oil in the inlet pocket enters the film. Since such model yields significantly reduced inlet temperature, lower pad temperature is predicted in their THD analysis. The predicted peak temperatures also have good agreement with their experimental data. Later, He, et al. (2002), noticed several interesting trends in the same group of test data. First, compared to a conventional pad, an inlet pocket pad does not always have lower temperature near its leading edge; instead, it consistently shows a smaller temperature gradient in the circumferential direction, which leads to the reduced peak temperature. This trend is clearly displayed in Figure 39. Second, on the curves of maximum temperature versus shaft speed, some flat sections are observed, and before those flat sections, the pocketed and
171
Y
conventional pads often have similar peak temperatures. As shown in Figure 35, the flat sections are likely the indicator of flow regime transition. To simulate these trends, He, et al. (2002), proposed a different theory that attributes the cooling effects to turbulent flow that elevates heat transfer. According to their model, the inlet pocket destabilizes the flow and causes early turbulence onset. Figure 39 shows the theoretical results that employed their triggered turbulence model. The predicted pad temperatures have close agreement with Brockwells experimental data. However, He, et al. (2002), did not identify the trigger that prompts turbulence on an inlet pocket pad.
Continous Film
Cavitated Film
(a)
(b)
Figure 40. Flooded and Starved Lubrication Conditions. For a multipad bearing, the level of starvation is different on each pad because a loaded pad has smaller operating clearance compared to an unloaded one. Therefore, starvation tends to occur on the unloaded pads first, and gradually spread onto the loaded pads. It also means that required pad flow is a function of eccentricity (or load) and speed, and different for each pad. Since pad flow is usually controlled by inlet orifices preceding each pad, the flow to each pad can be tailored, but only for a single load/speed condition. Also, note that the unloaded pads require more oil flow than the loaded pads. As shown in Figure 41, when the bearing has 100 percent supply flow (flooded), hydrodynamic pressure is developed on all pads because the unloaded top pads have 0.6 offset pivots. The hydrodynamic forces on those pads are labeled as F1 to F5, respectively. When the total supply flowrate to the entire bearing is cut by half, pad #3 and #4 are totally starved and pad #5 exhibits a 6.7 percent starvation region, the two bottom pads are still flooded. When the flowrate is reduced to 40 percent, the starvation region on pad #5 is expanded to 10 percent and pad #2 has a 6.7 percent starvation region. If the flowrate is further reduced to 30 percent, pad #5 becomes 100 percent starved and a 13.3 percent starvation area shows up on pad #2.
Y Pad #4
Y Pad #4
Figure 39. Pad Temperature Comparisons Between Conventional and Inlet Pocket Bearings, Experimental and TEHD results. Since the direct lubrication designs are relatively new, their cooling mechanisms are still not clear and subject to debate. More work needs to be done to understand their underlying physics and improve their theoretical modeling. Starvation During operation, lubricant must be continuously supplied into a bearing to replenish the side leakage. Many bearings are operating in flooded condition in which the amount of supply oil is more than what the bearing actually needs. With enough oil, continuous fluid film is always established at the leading edge of a pad that has convergent clearance. The flooded lubrication condition is schematically shown in Figure 40(a). A bearing can also be working in a starved condition in which the amount of oil is not enough to fill the pad leading edge clearance and the inlet region is cavitated. As shown in Figure 40(b), the continuous film is formed a certain distance away from the inlet where the clearance is sufficiently reduced. Figure 40 also shows conventional cavitation that is caused by divergent clearance near the pad trailing edge. Examples of starved applications include ring-lubricated bearings and direct lubrication designs that may be starved to minimize power consumption (Heshmat and Pinkus, 1985; Brockwell, et al., 1994). To model starvation, the main task is to determine the continuous film onset angle f. If f is known, the bearing can be analyzed using the standard models and the effective arc length from f to the trailing edge. f can be iteratively determined by comparing the available and required flowrates: at a certain location, if more lubricant is available to fill the clearance, the predicted f should be upstream where the larger clearance can accommodate the extra fluid; otherwise, the available lubricant can only fill a smaller space and f should be predicted further downstream (He, et al., 2003). Clearly, this search is coupled with the search of journal position.
Pad #5 F4 F3 F5 Shaft X W=1200 lb F2 F1 Pad #3
Pad #5 Pad #3 F5 Shaft X W=1200 lb F2 F1
Pad #1
Pad #2
Pad #1
Pad #2
100% Flow
Y Pad #4
50% Flow
Y Pad #4
Pad #5 Pad #3 F5 Shaft X W=1200 lb F2 F1
Pad #5 Pad #3 Shaft X
W=1200 lb F2 F1
Pad #1
Pad #2
Pad #1
Pad #2
40% Flow
30% Flow
Figure 41. Development of Starvation in a Five-Pad Tilting Pad Bearing. (Courtesy He, et al., 2003)
172
A starved bearing exhibits increased temperature and decreased friction loss. As shown in Figures 42 and 43, the pad temperature is increasing as the bearing becomes more and more starved. Meanwhile, since starvation leads to reduced continuous film area, the power loss due to viscous shearing is decreased. Besides higher temperature, starvation also reduces a bearings load capacity and stiffness. An example in He, at al. (2003), shows that a bearings horizontal stiffness Kxx can quickly diminish as the result of overly reduced flowrate. In addition, starvation may result in dry friction rubs (which may excite shaft vibration) and pad flutter (which may damage the fluttering pads).
Experiment, 100% Flow Experiment, 40% Flow Experiment, 35% Flow THD Results, 100% Flow THD Results, 40% Flow THD Results, 35% Flow
130 120
Load
Temperature(C)
110 100 90 80 70 60 50 0 72 144
temperature that is governed by groove mixing shown in Figure 38. A detailed modeling of the three-dimensional flow is impractical, and would involve turbulence, heat exchange with the solids, and possible two-phase flow of liquid and air. Therefore, as mentioned earlier, a simple equation based on energy balance is used as a practical approximation. In this model, a hot oil carryover factor is required to address the fact that not all exit flow Qout enters the next pad. The hot oil carryover factor, which is a function of the bearing design and operation condition, cannot be accurately obtained. Instead, it is usually estimated between 75 to 100 percent based on experience. Therefore, significant error can be introduced as the result of a poor estimate. Errors are also introduced on the back of a pad where heat convection boundary condition is usually applied. Similar to the hot oil carryover factor, the convection coefficient is unknown and often specified at an engineers best estimate. In some situations, such as misaligned shaft and bearing, the axial temperature cannot be assumed constant. The full three-dimensional energy equation must be employed, which leads to the difficulty of determining the boundary conditions at the axial ends.
Deformation boundary conditionsAs discussed above, to accurately model the journal and outer shell deformations, the shaft and bearing housing need to be taken into account. However, the shaft and housing conditions are difficult to obtain and they are dependent on a machines specific design and operation. Their modeling essentially goes beyond the scope of a bearing analysis.
216 288 360
Flow regime transitionTo analyze a possibly turbulent bearing,
Figure 42. Pad Temperature Versus Flowrate. (Experimental Results from Brockwell, et al., 1994)
50 45 40 35 30 25 20 15 10 5 0 20 40 Experiment, LOP THD Results, LOP Experiment, LBP THD Results, LBP
the difficult question is when to apply the turbulence model. In most analyses, two critical Reynolds numbers are employed to determine flow regime transition. If the actual Re in the bearing is smaller than the lower critical Reynolds number, the flow is considered laminar; if Re is larger than the upper critical Reynolds number, the flow is modeled as full turbulence; if Re is between those two threshold numbers, the flow is transitional and the eddy viscosity is scaled by a percentage factor (Suganami and Szeri, 1979). However, there is no reliable way to determine those critical Res. Although they are usually prescribed as constants, studies have indicated that they are functions of bearing geometry and operating condition (Xu and Zhu, 1993). Again, large errors can be introduced if a modeling is based on incorrect flow regime type.
Complex
geometriesThese include the inlet pockets, spray bars, bypass cooling grooves, and hydrostatic lift pockets for startup. Future research is needed to investigate these unconventional designs.
CONCLUSIONS
In this tutorial, major areas of journal bearings operation and modeling are discussed. With respect to the operational aspects, we have learned that:
60 80 100
A bearings load carrying capacity comes from the hydrodynamic pressure developed in the fluid film.
% Flow
A convergent wedge, a moving surface, and a viscous lubricant

are necessary to generate hydrodynamic force.
Figure 43. Power Savings Versus Flowrate. (Experimental Results from Brockwell, et al., 1994) Additional Comments on Modeling The modern TEHD theories generally give fairly accurate predictions for a bearings performance parameters. A variety of computer programs have been developed and successfully used in bearing design and analysis. Although significant progress has been made since Osborne Reynolds, bearing modeling still faces a number of challenges. To name a few:
Hydrodynamic forces have cross-coupled components that lead Due
to large attitude angle and stability issues for fixed geometry bearings. to their pads ability to tilt, tilting pad bearings have minimum cross-coupled forces and stiffnesses, which lead to their superior performance.
Viscous shearing generates heat in film, which leads to temperature rise and viscosity reduction.
Temperature
boundary conditionsIn thermal analysis, the difficult task is not to write down the equations, but to assign appropriate boundary conditions. The most important one is the film inlet
Hydrodynamic
pressure and temperature rise causes elastic deformations that change the film shape.
Viscous shearing also results in mechanical power loss.
173
There is a minimum required flowrate for oil supply. However,

a bearings dynamic properties can be represented by linear springs and dampers. coefficients all have significant rotordynamic implications. of a global system involving rotor, seals, supports, etc.
many bearings work in a flooded condition with more oil than the minimum.
For relatively small vibrations like those normally encountered, The direct stiffness, direct damping, and cross-coupled stiffness Dynamically, it is important to remember that a bearing is part There are two predominant types of vibration instability: oil
whirl and shaft whip. Associated only with fixed geometry bearings, the former is largely determined by the bearing properties and load. Applicable to either type of bearing, shaft whip is governed by the combined system.
Both
static and dynamic characteristics are speed and load dependent. To predict a bearings performance, theoretical models have been developed and successfully used in industry. The major models and mainstream techniques can be summarized as follows:
A theoretical model that includes pressure, temperature and elas Pressure calculations are the foundation of a TEHD algorithm.
computer code usually solves some form of the energy equation.
ticity effects is often called thermoelastohydrodynamic algorithm. When properly used, a TEHD analysis can yield good prediction of a bearings performance. The Reynolds equation is usually employed in computer programs.
For most analysis, thermal effects must be taken into account. A The elastic deformations should be included in the analysis of
high speed, heavily loaded bearings. However, the models need to be used with caution. properties of a turbulent bearing. Turbulent flow is usually associated with large bearing size, high shaft speed, and low lubricant viscosity.
A turbulence model must be available to accurately predict the For a tilting pad bearing, the dynamic coefficients can be highly frequency dependent. Most TEHD algorithms cannot be applied to predict direct lubricated or starved bearings. These special cases require additional enhancements. difficulties and challenges.
Kij L m Ob Op OJ p Q R Rb Rp RJ Re S T Tmax U u v W WU w X Y x y z p
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
Stiffness coefficients, i, j = X or Y Bearing axial length Pad preload Bearing center Pad arc center Journal center Pressure Flowrate Journal radius Bearing set bore radius Pad set bore radius Journal radius Reynolds number Sommerfeld number Temperature Maximum pad temperature Journal surface velocity Fluid velocity in circumferential (x) direction Fluid velocity in radial (y) direction Applied load Unit load [WU = W/(L.D)] Fluid velocity in axial (z) direction Horizontal direction Vertical direction Circumferential direction along a pad Radial direction across film Axial direction along a pad Pad offset factor Pad arc measured from leading edge to the pivot location Journal attitude angle Heat conductivity Lubricant viscosity Pad arc length Lubricant density Shear stress Shaft rotational speed
REFERENCES
API Standard 684, 2005, Tutorial on Rotordynamics: Lateral Critical, Unbalance Response, Stability, Train Torsional and Rotor Balancing, Second Edition, American Petroleum Institute, Washington, D.C. Barrett, L. E., Gunter, E. J., and Allaire, P. E., 1978, Optimum Bearing and Support Damping for Unbalance Response and Stability of Rotating Machinery, ASME Journal of Engineering for Power, 100, (1), pp. 89-94. Bouard, L., Fillon, M., and Frene, J., 1996, Comparison Between Three Turbulent ModelsApplication to Thermohydrodynamic Performances of Tilting-Pad Journal Bearings, Tribology International, 29, pp. 11-18. Boyd, J. and Raimondi, A. A., 1953, An Analysis of the PivotedPad Journal Bearing, Mechanical Engineering, 75, pp. 380-386. Brechting, R., 2002, Static and Dynamic Testing of Tilting Pad Journal Bearings as a Function of Load Angle and Journal Speed, Master Thesis, University of Virginia. Brockwell, K., Dmochowski, W., and DeCamillo, S. M., 1994, Analysis and Testing of the LEG Tilting Pad Journal BearingA New Design for Increasing Load Capacity, Reducing Operating Temperatures and Conserving Energy, Proceedings of the Twenty-Third Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 46-56. Brugier, D. and Pascal, M., 1989, Influence of Elastic Deformations of Turbo-Generator Tilting Pad Bearing on the Static Behavior and on the Dynamic Coefficients in Different Designs, Journal of Tribology, 111, pp. 364-371.
Current state-of-the-art bearing modeling still faces a variety of

NOMENCLATURE
c cb cp co c Cij Cp D e eX eY E EX EY Fx Fy h hi ho = = = = = = = = = = = = = = = = = = = Bearing clearance Assembled clearance Pad machined clearance Nominal bearing clearance Clearance variation due to deformation Damping coefficients, i, j = X or Y Lubricant specific heat Journal diameter Journal eccentricity Journal eccentricity projected on the horizontal (X) axis Journal eccentricity projected on the vertical (Y) axis Journal eccentricity ratio Journal eccentricity ratio projected on the horizontal (X) axis Journal eccentricity ratio projected on the vertical (Y) axis Film force in horizontal (X) direction Film force in vertical (Y) direction Film thickness Film thickness at wedge inlet Film thickness at wedge outlet
174
Byrne, J. M. and Allaire, P. E., 1999, Optimal Design of Fixed Pad Fluid Film Bearings for Load Capacity, Power Loss, and Rigid Rotor Stability, Report No.UVA/643092/MAE98/530, ROMAC Laboratories, University of Virginia. Constantinescu, V., 1959, On Turbulent Lubrication, Proceedings of the Institution of Mechanical Engineers, 173, pp. 881-900. Cope, W., 1949, The Hydrodynamic Theory of Film Lubrication, Proceedings of the Royal Society, London, United Kingdom, Series A, 197, pp. 201-217. Desbordes, H., Fillon, M., Wai, C., and Frene, J., 1994, Dynamic Analysis of Tilting-Pad Journal BearingInfluence of Pad Deformations, Journal of Tribology, 116, pp. 621-628. Dowson, D., 1962, A Generalized Reynolds Equation for Fluid Film Lubrication, International Journal of Mechanical Science, 4, pp. 159-170. Dowson, D., 1998, History of Tribology, Second Edition, Suffolk, United Kingdom: Profession Engineering Publishing Ltd. Dowson, D. and Hudson, J., 1963, Thermo-Hydrodynamic Analysis of the Infinite Slider Bearing: Part IIThe Parallel Surface Bearing, Lubrication and Ware Convention, Institution of Mechanical Engineering, Paper 4 and 5. Edney, S., Heitland, G. B., and Decamillo, S., 1998, Testing, Analysis, and CFD Modeling of a Profiled Leading Edge Groove Tilting Pad Journal Bearing, Presented at the International Gas Turbine & Aeroengine Congress & Exhibition, Stockholm, Sweden. Ehrich, F. F. and Childs, D. W., 1984, Self-Excited Vibration in High-Performance Turbomachinery, Mechanical Engineering, May, pp. 66-79. Elrod, H. and Ng, C., 1967, A Theory for Turbulent Fluid Films and Its Application to Bearings, Journal of Lubrication Technology, 89, pp. 346-362. Elwell, R. C. and Booser, E. R., 1972, Low-Speed Limit of LubricationPart One: What is a Too Slow Bearing? Machine Design, June 15, pp. 129-133. Ettles, C., 1980, The Analysis and Performance of Pivoted Pad Journal Bearings Considering Thermal and Elastic Effects, Journal of Lubrication Technology, 102, pp. 182-192. Fillon, M., Bligoud, J., and Frene, J., 1992, Experimental Study of Tilting-Pad Journal BearingsComparison with Theoretical Thermoelastohydrodynamic Results, Journal of Tribology, 114, pp. 579-587. Fitzgerald, M. and Neal, P., 1992, Temperature Distributions and Heat Transfer in Journal Bearings, Journal of Tribology, 114, pp. 122-130. Gardner, W. W., 1976, Journal Bearing Operation at Low Sommerfeld Numbers, ASLE Transactions, 19, (3), pp. 187-194. Hagg, A. C., 1946, The Influence of Oil-Film Journal Bearings on the Stability of Rotating Machinery, ASME Transactions, Journal of Applied Mechanics, 68, pp. A211-A220. Hamrock, B. J., 1994, Fundamentals of Fluid Film Lubrication, New York, New York: McGraw-Hill. He, M., Allaire, P., Barrett, L., and Nicholas, J., 2002, TEHD Modeling of Leading Edge Groove Journal Bearings, IFToMM, Proceedings of 6th International Conference on Rotor Dynamics, 2, pp. 674-681. He, M., Allaire, P., Barrett, L., and Nicholas, J., 2003, THD Modeling of Leading Edge Groove Bearings under Starvation Condition, The 58th STLE Annual Meeting, New York, New York, 2003.
He, M., Allaire, P., Cloud, C. H., and Nicholas, J., 2004, A Pressure Dam Bearing Analysis with Adiabatic Thermal Effects, Tribology Transactions, 47, pp. 70-76. Heshmat, H., 1991, The Mechanism of Cavitation in Hydrodynamic Lubrication, Tribology Transactions, 34, pp. 177-186. Heshmat, H., and Pinkus, O., 1985, Performance of Starved Journal Bearings with Oil Ring Lubrication, Journal of Tribology, 107, pp. 23-32. Heshmat, H. and Pinkus, O., 1986, Mixing Inlet Temperatures in Hydrodynamic Bearings, Journal of Tribology, 108, pp. 231248. Hirs, G., 1973, A Bulk-Flow Theory for Turbulence in Lubricant Film, Journal of Lubrication Technology, 95, pp. 137-146. Hummel, C., 1926, Kristische Drehzahlen als Folge der Nachgiebigkeit des Schmiermittels im Lager, VDIForschungsheft 287. Jones, G. J. and Martin, F. A., 1979, Geometry Effects in TiltingPad Journal Bearings, STLE Tribology Transactions, 22, pp. 227-244. Kim, J., Palazzolo, A., and Gadangi, R., 1994, TEHD Analysis for Tilting-Pad Journal Bearings Using Upwind Finite Element Method, Tribology Transactions, 37, pp. 771-783. Kirk, R. and Reedy, S., 1988, Evaluation of Pivot Stiffness for Typical Tilting-Pad Journal Bearing Design, Journal of Vibration, Acoustics, Stress, and Reliability in Design, 110, pp. 165-171. Lanes, R. F., Flack, R. D., and Lewis, D. W., 1981, Experiments on the Stability and Response of a Flexible Rotor in Three Types of Journal Bearings, ASLE Transactions, 25, (3), pp. 289-298. Lund, J., 1964, Spring and Damping Coefficients for the Tilting Pad Journal Bearing, ASLE Transactions, 7, pp. 342-352. Lund, J. W., 1987, Review of the Concept of Dynamic Coefficients for Fluid Film Journal Bearings, ASME Journal of Tribology, 109, pp. 37-41. Lund, J. and Pedersen, L., 1987, The Influence of Pad Flexibility on the Dynamic Coefficients of a Tilting-Pad Journal Bearing, Journal of Tribology, 109, pp. 65-70. Lund J. and Saibel, E., 1967, Oil Whip Whirl Orbits of a Rotor in Sleeve Bearings, Journal of Engineering for Industry, 89, pp. 813-823. McCallion, H., Yousif, F., and Lloyd, T., 1970, The Analysis of Thermal Effects in a Full Journal Bearing, Journal of Lubrication Technology, 92, pp. 578-587. Muszynska, A. and Bently, D. E., 1995, Fluid-Induced Instability of Rotors: Whirl and WhipSummary of Results, Noise and Vibration Conference, Pretoria, South Africa. Newkirk, B. L., 1924, Shaft Whipping, General Electric Review, 27, (3), pp. 169-178. Newkirk, B. L. and Taylor, H. D., 1925, Shaft Whipping Due to Oil Action in Journal Bearings, General Electric Review, 28, (8), pp. 559-568. Ng, C. and Pan, C., 1965, A Linearized Turbulent Lubrication Theory, Journal of Basic Engineering Series D, 87, pp. 675-688. Nicholas, J. C. and Wygant, K. D., 1995, Tilting Pad Journal Bearing Pivot Design for High Load Applications, Proceedings of the Twenty-Fourth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 33-48.
175
Nicholas, J. C., 1994, Tilting-Pad Bearing Design, Proceedings of the Twenty-Third Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 179-194. Nicholas, J. C., Gunter, E. J., and Barrett, L. E., 1978, The Influence of Tilting Pad Bearing Characteristics on the Stability of High-Speed Rotor Bearing Systems, Report No.UVA/643092/MAE81/141, ROMAC Laboratories, University of Virginia. Paranjpe, R. S. and Han, T., 1994, A Study of the Thermohydrodynamic Performance of Steadily Loaded Journal Bearings, Tribology Transactions, 37, pp. 679-690. Parsell, J. K., Allaire, P. E., and Barrett, L. E., 1983, Frequency Effects in Tilting-Pad Journal Bearing Dynamic Coefficients, ASLE Transactions, 26, pp. 222-227. Petrov, N., 1883, Friction in Machines and the Effect of the Lubricant, Inzhenernii Zhurnal, St. Petersburg, Russia, 1, pp. 71-104, 2, pp. 227-279, 3, pp. 377-436, 4, pp. 435-464 (In Russian). Pinkus, O., 1987, The Reynolds Centennial: A Brief History of the Theory of Hydrodynamic Lubrication, ASME Journal of Tribology, 109, pp. 2-20. Raimondi, A. and Boyd, J., 1958, A Solution of the Finite Journal Bearing and Its Application to Analysis and DesignPart III, Transactions of ASLE, 1, pp. 159-209. Reynolds, O., 1886, On the Theory of Lubrication and Its Applications to Mr. Beauchamp Towers Experiments Including an Experimental Determination of the Viscosity of Olive Oil, Philosophical Transactions, 177, pp. 157-234. Safar, Z. and Szeri, A. Z., 1974, Thermohydrodynamic Lubrication in Laminar and Turbulent Regimes, Journal of Lubrication Technology, 96, pp. 48-56. Salamone, D. J., 1984, Journal Bearing Design Types and Their Applications to Turbomachinery, Proceedings of the Thirteenth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 179-192. Seireg, A. and Ezzat, H., 1973, Thermohydrodynamic Phenomena in Fluid Film Lubrication, Journal of Lubrication Technology, pp. 187-198. Simmons, J. and Lawrence, C., 1996, Performance Experiments with 200 mm, Offset Pivot Journal Pad Bearing, Tribology Transactions, 39, pp. 969-973.
Stodola, A., 1925, Kristische Wellenstorung infolge der Nachgiebigkeit des Oelpolsters im Lager, Schweizerische Bauzeiting, 85, pp. 265-266. Suganami, T. and Szeri, A. Z., 1979, A Thermohydrodynamic Analysis of Journal Bearings, Journal of Lubrication Technology, 101, pp. 21-27. Szeri, A., 1979, TribologyFriction, Lubrication, and Wear, New York, New York: McGraw-Hill Book Company. Taniguchi, S., Makino, T., Takeshita, K., and Ichimura, T., 1990, A Thermohydrodynamic Analysis of Large Tilting-Pad Journal Bearing in Laminar and Turbulent Flow Regimes with Mixing, Journal of Tribology, 112, pp. 542-550. Tower, B., 1883, First Report on Friction Experiments, Proceedings Institute of Mechanical Engineering, pp. 632-666. Tower, B., 1885, Second Report on Friction Experiments, Proceedings Institute of Mechanical Engineering, pp. 58-70. Wilcock, D., 1950, Turbulence in High Speed Journal Bearings, Transactions of the ASME, 72, pp. 825-834. Xu, H. and Zhu, J., 1993, Research of Fluid Flow and Flow Transition Criteria from Laminar to Turbulent in a Journal Bearing, Journal of XiAn Jiaotong University, 27, pp. 7-14. Zuck, C. and Flack, R., 1986, Experiments on the Stability of an Overhung Rotor in Pressure-Dam and Multilobe Bearings, ASLE Transactions, 30, pp. 225-232.
BIBLIOGRAPHY
Allaire, P. E., 1997, Course Notes: Lubrication Theory and Design, Department of Mechanical and Aerospace Engineering, University of Virginia, Charlottesville, Virginia. Hagg, A. C. and Sankey, G. O., 1956, Some Dynamic Properties of Oil-Film Journal Bearings with Reference to the Unbalance Vibration of Rotors, ASME Transactions, Journal of Applied Mechanics, 23, pp. 302-306. Lund, J. W., 1974, Stability and Damped Critical Speeds of a Flexible Rotor in Fluid-Film Bearings, ASME Journal of Engineering for Industry, Transactions ASME, 96, (2), pp. 509-517. Wills, J. G., 1980, Lubrication Fundamentals, Mobil Oil Corporation, New York, New York: Marcel Dekker, Inc.
176
APPLIED RISK AND RELIABILITY FOR TURBOMACHINERY PART IRELIABILITY PROCESSES

by Charles R. Rutan
Senior Engineering Advisor, Specialty Engineering Lyondell Chemical Company Alvin, Texas
Charles R. (Charlie) Rutan is Senior Engineering Advisor, Specialty Engineering, with Lyondell Chemical Company, in Alvin, Texas. His expertise is in the field of rotating equipment, hot tapping/plugging, and special problem resolution. He has three patents and has consulted on turbomachinery, hot tapping, and plugging problems all over the world in chemical, petrochemical, power generation, and polymer facilities. Mr. Rutan received his B.S. degree (Mechanical Engineering, 1973) from Texas Tech University. He is a member of the Advisory Committee of the Turbomachinery Symposium, and has published and/or presented many articles.
(MOSAR), management oversight risk tree (MORT), Weibull Analysis, and Six Sigmahave been developed to aid in quantifying the risk of extending the operational time of the critical turbomachinery.
KEPNER-TREGOE
Kepner-Tregoe Problem Solving and Decision Making (PSDM) is a step-by-step process that helps people resolve business. Used in organizations worldwide, PSDM helps individuals, groups, and/or teams efficiently organize and analyze vast amounts of information and take the appropriate action. This process provides a framework for problem solving and decision making that can be integrated into standard operating procedures. It is used to enhance other operational improvement tools such as Six Sigma, Lean Manufacturing, and others. PSDM comprises four distinct processes:
Situation Appraisal is used to separate, clarify, and prioritize

ABSTRACT
There are several methods of analysis to define the reliability of the critical rotating equipment at various facilities within a company. Part I of this tutorial is intended to present several of these methods. The following Part II presents the calculations.
concerns. When confusion is mounting, the correct approach is unclear, or priorities overwhelm plans, Situation Appraisal can be used.
Problem Analysis
INTRODUCTION
In the mid 1980s, the authors supervisor gave him an opportunity to better define the critical machinery reliability as turnarounds were extended. The task was to develop an equation that would produce a number to be used as a guide such that the plant management would have a feeling as to the risk for postponing the next olefin unit turnaround or minimizing the amount of inspection performed on the major rotating equipment, e.g., minor or major overhaul(s). Based on the critical machinery on the history, design, vibration, and process conditions he developed a weighted number that he tried to use as an indicator to justify the overhaul requirements. At the time, this was not accepted well by the plant management because other facilities in the company had longer/shorter shutdown intervals as well as published reports of other companies in the same commodity chemical industry and the Solomon report that was published every two years did not agree with his conclusions derived from this number. During this period of time there were two methods: Problem Solving and Decision Making Managerial Analytics, a Monsanto Chemical Company event Kepner-Tregoes analysis system In later years several other methods of potential problem and/or root cause analysishazard and operability analysis (HAZOP), Federal Emergency Management Agency (FEMA), failure mode effect and criticality analysis (FMECA), event-tree analysis (ETA), Delphi, method organization for a systematic analysis of risks
193
is used to find the cause of a positive or negative deviation. When people, machinery, systems, or processes are not performing as expected, Problem Analysis points to the relevant information and leads the way to the root cause.
Decision Analysis is used for making a choice; it is intended to

clarify the purpose and balances risks and benefits to arrive at a supported choice.
Potential Problem/Opportunity Analysis is used to protect and
leverage actions or plans. Potential Problem Analysis should define the driving factors and identifies ways to lower risk. When one action is taken, new opportunities, good or bad, may arise. These opportunities must be recognized and acted on to maximize the benefits and minimize the risks.
MANAGERIAL ANALYTICS
Managerial Analytics (MA) is an analytical identification process developed by the Monsanto Company. Managing change and using change to manage requires the use of some combination of the five basic analytical processes.
Event Analysis is a systematic process to identify events and
events of change that impact the reliability of the turbomachinery. Events and events of change from the past and present have an impact on the present and future reliability. Events of change could be the stopping of wash oil injection or the rising of the sodium concentration in the steam or not repairing the spare rotor. Step 1. First person responsibility Step 2. Recognize events and their relationships Step 3. Establish priority Step 4. Separate and sequence components
194
Step 5. Identify intended nominal first person results Step 6. Identify resolution resources, threats, and opportunities Step 7. Identify intended nominal first person action Step 8. Select the analytical processes Step 9. Statement for the first analysis
ROOT CAUSE ANALYSIS (RCA)

Define the Problem Cause and Effect There are five elements of a cause and effect chart. 1. Primary effect: a. A singular effect of consequence that we wish to eliminate or mitigate b. The what in the problem definition c. It is always the most present cause in the analysis and frequently the most significant. d. It is the point at which we begin to ask why. 2. Actions and condition causes: a. They are both causes, actions are momentary and conditions exist over time. b. Actions can become conditions and conditions can become actions. c. Actions and conditions interact to create effect. 3. Casual connection caused by: a. Forces that cause going from present to past. b. Elicits a more specific response c. Minimizes storytelling d. If a cause connects it adds to the visual dialogue and therefore has value. 4. Evidence is the data that supports a conclusion and is presented as: a. SensedIt is processed through our senses of sight, sound, smell, taste, and touch. Sensed is the highest quality evidence. b. InferredThe ability to infer is derived from our understanding of known and repeated casual relationships. Inference is the next highest quality of evidence. c. Evidence is important because: i. It supports the reality of any single cause. ii. Solutions should only be applied to evidence-based causes. iii. It minimizes the influence of politics and power plays. 5. Stop or a ? Problem Definition There are four elements of problem definition. 1. WhatWhat is the problem? 2. WhenWhen did it happen? 3. WhereWhere did it happen? 4. SignificanceWhy are we working on this? Identify Effective Solutions The root cause is not what we seek, it is effective solutions. 1. Challenge each cause. 2. Offer possible solutions for each cause. Implement the Best Solutions 1. Prevents recurrences a. Prevents or mitigates this problem b. Prevents similar problems c. Does not create additional problems or unacceptable situations 2. Within your control a. Your control may be you, your department, your company, your suppliers, or your customers. b. Nature is not within your control. c. The facilitator is rarely the problem owner. 3. Meets your goals and objectives a. The goals of the overall organization b. The goals of your department or group
Deviation Analysis is a process to help determine the unknown
cause of an observed effect deviating from a standard effect in order to decide on action. This is used for both positive and negative deviations and in other words events of change that have happened. Step 1. Statement of the deviation effect Step 2. Deviation effect specifications Step 3. Unique characteristics of the involved dimensions Step 4. Change events Step 5. Possible causes Step 6. Test possible causes Step 7. Set priority on possible causes Step 8. Verification of cause
Action Analysis is used to select a single course of action from

Step 1. Statement of the intended nominal action Step 2. Set the desired effects Step 3. Classify the desired effects Step 4. Weight the want effects Step 5. Generate action alternatives Step 6. Filter the action alternatives through must effects Step 7. Score action alternatives to the wanted effects Step 8. Impact evaluation Step a. Identify an action alternative Step b. Identify impacts Step c. Identify potential deviations Step d. Assess impacts on the environment Step e. Summarize potential deviations and their likely causes Step f. Plan actions to manage likely causes Step g. Plan actions to manage potential deviations Step 9. Make best balanced selection
several courses of action. This is based on the projected performance of the action to attain a set of desired effects and assessing the impacts of and to that action if it were taken.
Action Planning helps to decide What will be done?, Who
will do it?, and When it will be done to reach one or more desired future effects that are not expected to occur unless something is done. It involves a set of interrelated and interdependent actions. Step 1. Define the action required Step 2. Define person(s) who have the prime responsibility to complete the required action Step 3. Define support resources Step 4. Define the date and time to initiate the action Step 5. Define the projected time of completion of the action
Potential Deviation Analysis is a process used to examine a

Section 1. Statement of the intended nominal action Section 2. Identification of potential deviations Step a. Identify a planned action Step b. Identify impacts Step c. Identify potential deviations Step d. Assess impacts to the environment Step e. Summarize potential deviations and their likely causes Section 3. Action planning for potential deviations Step a. Plan actions to manage likely cause Step b. Plan action to manage potential deviation Step c. Revise the action plan
planned action or a future event of change for significant impacts and deviations, and to plan additional actions to manage these results.
195
c. Your individual goals and objectives d. Must provide reasonable value
RELIABILITY ASSESSMENT OF ROTATING EQUIPMENT

Reliability assessment of rotating equipment (RARE) is one tool of an overall maintenance and turnaround strategy of critical rotating equipment. Origin of Initial Program and Concept The Hartford Steam Boiler Inspection & Insurance Company (HSB) expected the companies they insured to follow the steam turbine overhaul intervals recommended by the original equipment manufacturer (OEM). Due to world market pressure and technological advancements companies began extending the intervals between major unit turnarounds. These companies depended on the expertise of their rotating equipment engineers to help define how long they could reliably operate between turnarounds. Their goal was to lengthen the interval between major overhaul outages to coincide with the need for other inspections and testing mandated by government agencies or process needs. During the 70s and 80s the chemical, petrochemical, and refining industries would perform a major overhaul of their critical machinery every four to six years as a standard practice. As operating companies extended their turnaround intervals HSB became concerned about the increased risk their customers were taking. At this point in time, data showing the effects or risks of stretching run times between overhauls were not available. HSB decided to develop a program that would quantify these risks for their customers. M&M Engineering, a part of the HSB company at that time, was given the assignment to develop a tool to assess the reliability of steam turbines. The tool is called Steam Turbine Reliability Assessment Program (STRAP). HSB did not plan to use this program to structure premiums. The vision was to require a STRAP analysis whenever a company chose to run longer than the OEM recommended practice. If STRAP showed that they were at low risk then HSB would authorize or accept the companys plans, but if STRAP showed the company to have a high risk they were told that they needed to implement some risk reduction procedures or improvements to mitigate some of these risks. Consortium of Companies With this vision, M&M Engineering pulled together a group of recognized industry experts in the turbomachinery field. The experts brought with them their individual companys reliability assessment techniques, industry standards (API, ASME, etc.), recognized industry best practices, new technologies, and most important their personal experience. Starting with a blank sheet of paper the group spent countless hours defining, developing, categorizing, and weighing questions and responses. The caveat for this group was when completed they would take this program back to their respective companies and use it as another tool to justify turbine improvements that would improve reliability and extend run lengths, which then resulted in a significant savings of money. The turbine data accumulated, which consisted of original design specifications, history, uprates, upgrades, failures, and site specific information, were then entered into the program developed by M&M Engineering. The algorithms in the program took the weighted turbine data and generated a risk index number (RIN). The RIN is the number of days that the turbine will be down, during the run extension. It is based on statistical information that is calculated from the data contained in the database for the same general type of turbine. The program gives 25 specific items to be addressed and calculates a return on the investment (ROI) using site pricing data. Armed with this information the rotating equipment engineer and managers can make informed decisions to accept the projected risk or to take corrective actions
to minimize or mitigate the risks that were defined. Like home and car insurance, risk can never be eliminated but most risks can be minimized, then it becomes a decision on what level of risk the plant/company is willing to accept. With this information longterm shutdown strategy can be developed that is based on the operation, maintenance, and reliability practices of the unit.
STEAM TURBINE RISK ASSESSMENT PROGRAM

1. General Information Data Sheet a. Plant specifics? b. Size or class of the turbine (five categories)? c. Age of the turbine? d. Manufacturer of the turbine? e. What is the turbine driving? f. When the turbine was last dismantled? g. Etc.? 2. Turbine Performance (Design and Actual) a. Horsepower? b. Speed? c. Inlet flow? d. Temperature? e. Pressure? f. Etc.? 3. Site and Utility Data a. Location? b. Steam generation? c. Etc.? 4. Construction Features a. Type of turbine? b. Critical speed? c. Control system? d. Etc.? 5. Spares a. Complete turbines? b. Bearing sets? c. Complete rotor sets? d. Stationary diaphragms? e. Case? f. Nozzles blocks? g. Where are they stored? h. Couplings? i. Labyrinth seal sets? j. Control valve parts/governor valve assemblies? k. How many days are required to prepare the rotor for installation? l. Is there a periodic inspection for signs of corrosion on the spare parts? 6. Maintenance and Repairs a. How many hours to the repair shop? b. How often do you drain water from the oil reservoir? c. Do you drain it at startup? d. Turbine overhauls? e. Are there documented detailed overhaul procedures for this/similar turbines? f. How is the lube oil system cleaned during overhauls? g. Qualifications? h. Foundation inspection? i. Alignment? j. Inspections? k. Oil systems? 7. Turbine Operation a. Procedures? i. Do the plants written steam turbine operating procedures include: (1) Prestartup checklist? (2) Overspeed tests?
196
(3) Putting turbine on slow roll? (4) Normal operations? ii. Do you have a procedure for lining up sealing steam? iii. Is there a formal management of change? b. Tests? i. When is the governor or valve rack exercised online? ii. When do you test the trip and throttle valve? iii. When do you test the nonreturn valve? c. Qualifications? i. Do operators have the authority to shut down the turbine? 8. Monitoring and Protection a. Which of the following parameters are monitored and trended? i. Turbine steam inlet temperature? ii. Bearing metal temperatures? iii. Thrust? iv. Vibration? v. Lube oil pressure? b. Do the following parameters trigger an operatingspecific alarm? i. Turbine steam inlet temperatures? ii. Bearing metal temperatures? iii. Thrust? iv. Vibrationradial? c. Do the following parameters trigger an alarm followed by a trip? i. Vibrationradial? ii. Thrust? 9. UpgradesHave the following components been upgraded? a. Rotor assembly? b. Bearings? c. Casing? d. Coupling? e. Seals? f. Trip and throttle valve? g. Miscellaneous? 10. Steam System a. What supplies steam to the turbine? b. What type of make up water does it use? c. What type of condensate polishing does this unit use? d. Is steam purity monitored? e. Etc.? 11. Past Failures/Problems a. Turbine internal components? b. Bearings? c. Casing? d. Coupling? e. Seals? f. Trip and throttle valve and its components? g. Governor? h. Fouling? 12. Consequence Data a. Plant production in dollars per day (minimum/maximum)? b. What is the cost if the turbine goes down? c. What other costs are there if the turbine shuts down? d. How does the plant typically handle a rub during the startup of this turbine? e. How would the plant typically handle fouling of this turbine during normal operation? The total number of possible questions is about 3000, but for a large steam turbine 350 to 400 questions are normally answered. Question Weighting The team of experts then weighted each of the 3000+ questions and their respective answers based on their experience and
knowledge. As part of the development of failure probabilities, the team needed to set a baseline interval for overhaul outages. They decided to use a six year dismantle overhaul schedule frequency for Class 1 to 4 turbines and a five year overhaul schedule frequency for Class 5 turbines. Thus, the baseline probabilities developed were based on the turbine operating for six years or five years without opening the case. Since risk is calculated by multiplying probability times the consequence, the calculated risk would be for a six year/five year interval. Thus, the risk calculated would be in days of lost production over a five or six year interval. Because some STRAP users do not divulge financial data about production revenue, converting the days in lost production to dollars cannot be performed for all turbines. It was decided to create the term risk index number, which would allow all turbines to be compared to each other. The RIN is the risk of failure in days of lost production over a six year interval for Class 1 to 4 turbines and over a five year interval for Class 5 turbines. On the basis of the input data and the likelihood-consequence information, a risk for operation of the turbine may be calculated as a function of time between the dismantle inspections. In each case, a quantified list of recommendations to mitigate the risk will also be reported based on the greatest contribution to the risk. Inspection outage plans then may be tailored to optimize the time between overhauls on the basis of acceptable level of risk. The risk index number is a number generated by the program based on:
The questions and their corresponding answers. Industry standards (ASME, API, etc.). Accepted industry practices. Latest technology. Relative to other turbines in the company and/or the number of
turbines of the same design in the database. Probability The probability of failure of a component is the risk of failure in days of lost production (RIN) divided by the consequence of that component. Since risk is the product of probability times consequence there are questions that will significantly affect the RIN if answered with a poor option or not answered at all. An example of this is that the question of testing of overspeed trip and an answer of never tested would have a significant impact on the RIN and the recommendations by increasing the probability of failure. Program Aids The program has several aids for the user in an effort to make the output meet the user needs while providing the best output. Some of the aids are:
Check for missing answers. Check for inconsistent answers. Etc.

Results and Comparisons STRAP will compare the turbines with:
Other turbines in the company. All the turbines in the database. Turbines in the same class. Turbines by the same manufacturer. Turbines in the same industry. Etc.
Recommendations STRAP makes recommendations to improve the reliability of the turbine(s) based on return of investment. It is then up to the engineers to decide which recommendation can be executed, in what order, and during an outage and/or online.
197
Impact Based upon the consequence data of the operating unit that has been input to the program, the ROI can be calculated. Some of the basic questions are:
Unit production rate? Effect on the plant if a unit goes down per day? Impact of a trip of the turbine? Etc.?
RELIABILITY ASSESSMENT OF COMPRESSORS
The reliability assessment of compressors (RAC) program has been developed in a very similar manner to the STRAP program. Compressors are divided into the types of service to which they are being applied. In addition the program requires all the constituents on a percent molecular weight basis. In reality the RAC program is far more complicated than the STRAP program. There are a number of: A. General Information 1. What group this compressor belongs to? a. Charge gas/cracked gas? b. Air? c. Ammonia? d. Chlorine? e. Oxygen? f. Refrigeration (clean gas)? g. Other? 2. What industry environment this compressor operates in? a. Chemical/petrochemical? b. Gas? c. Refining? d. Other? 3. Compressor General Details? a. Compressor manufacturer (there are a multitude of manufacturers)? b. Model number? c. Serial number? d. Date manufactured? e. Compressor duty? f. Driver? g. Number of years in service? B. Construction 1. Type of end seals? 2. Type of interstage seals? 3. Internal coatings? 4. Has the rotor been high-speed balanced? 5. Do you have a lube and oil system designed API 614? 6. Type of bearings? 7. Materials of construction? 8. Type of coupling(s)? 9. What is the number of impellers? C. Past Failures/Problems 1. Has the compressor ever had past failures or problems? 2. Has the compressor ever operated in reverse?
3. How often does the compressor have vibration problems? 4. Has the lube oil system been contaminated? 5. Has the seal oil system been contaminated? 6. Has the buffer gas consumption increased? 7. Have you experienced compressor trips due to instrumentation problems? 8. Has the compressor ever been oversped? 9. Have you ever had problems with kinking in the past? D. Design Versus Actual 1. Inlet a. Pressure? b. Temperature? c. Molecular weight? 2. Discharge a. Pressure? b. Temperature? c. Molecular weight? 3. Brake horsepower required? 4. Speed? 5. Estimated surge ICFM? E. Process Gas Data 1. Is the process dry or wet? 2. Is the process gas corrosive? 3. Does the process foul? 4. Do you monitor gas composition online? 5. What molecular weight was the compressor designed for? 6. What is the current molecular weight of the process? 7. Process stream a. Air (MW 28.966) b. Carbon monoxide (MW 28.010) c. Ethylene (MW 28.052) d. Propane (MW 44.094) F. Site Data G. Control Systems H. Lube Oil Systems I. Spares J. Maintenance and Repairs K. Operation L. Monitoring and Protection M. Rerates and Upgrades N. Seal Fluid System O. Environment and Business Consequence Data
CONCLUSION
The Reliability Assessment of Rotating Equipment (RARE) is the combination of both the STRAP and RAC programs. Depending on the maintenance strategy of the company and/or the facility, either/or STRAP and RARE programs can be a significant benefit in assessing the critical machinery performance and identifying the areas that could or should be addressed to improve the reliability and define the obstacles to extending the run time between minor and major overhauls with an acceptable risk.
198
APPLIED RISK AND RELIABILITY FOR TURBOMACHINERY PART IIRELIABILITY CALCULATIONS

by Shiraz A. Pradhan
Consulting Engineer ExxonMobil Chemical Company Baytown, Texas
Shiraz A. Pradhan is a Consulting Engineer, with ExxonMobil Chemical Company, in Baytown, Texas. His recent experiences include the design, commissioning, and startups of oxoalcohol, halobutyle, polyethylene, polypropylene, and fluids projects in the Far East, South America, and the U.S. He previously worked with Esso/Imperial Oil in Canada as Machinery Engineer and with British Gas Corporation in the United Kingdom as Senior Project Leader. He has been involved in reliability audits and service factor improvement projects in ammonia, fertilizer, PVC, olefins and polyolfins plants, and oil and gas pipeline projects internationally. Mr. Pradhan has a degree (Mechanical Engineering) from the University of Nairobi, and an M.S. degree from Lehigh University in Pennsylvania. He holds the title of European Engineer (FEENI, Paris), is a fellow of the Institution of Mechanical Engineers, U.K., and is a registered Professional Engineer in the Province of Ontario, Canada.
Have more failure modes than electronic components

Fundamental to reliability assessment of mechanical components is the need for failure distribution and supporting data that describe the behavior of the components in the real world. This is more easily said than done. Cumulative Distribution Function For reliability prediction one would like to know the probability of a failure occurring before a time t. This can be derived by the equation: F (t), Probability of a failure before time t =
f (t )dt
(1)
As t approaches infinity, F(t) approaches 1. Reliability Function Reliability function is complementary to the cumulative distribution function and gives the probability of survival of a component or system to specified time t.
INTRODUCTION
Reliability is critical for all industries. For the petrochemical industry it assumes added significance because much equipment is unspared or has minimal redundancy. Table 1 shows a comparison between the commercial airlines, nuclear, and petrochemical industries. Table 1. System Characteristics for Different Industries.
System
Mission Length, Hr Access During Mission Access Between Mission
R(t ) = 1 F (t ) =
Failure Rate or Hazard Function
f (t )dt
t
(2)
This function allows the determination of the failure probability of a system or component in a small increment of time t, having survived to time t.
Commercial Airlines
<50 NIL
Nuclear
<5000* NIL
Petrochemical
8760> T > 70000 NIL
h(t ) =
f (t ) R(t )
(3)
Failure Distributions for Mechanical Systems

Full Limited FULL if a planned IRD**
Exponential Distribution This distribution is used extensively in the electronic industry and for some mechanical systems reliability assessment as well: where = failure rate per unit time
* Depending on mandated maintenance ** Inspection, Repair Downtime
COMPARISON BETWEEN ELECTRONICS AND MECHANICAL SYSTEMS

In electronic component reliability assessment the concept of constant failure rate is used (Ireson, et al., 1996). This is not the case for mechanical and machinery components. There are many reasons for this. Machinery components follow:
f (t ) = e(t ) for t > 0

and = 1 / MTBF
(4)
(5)
where MTBF = mean time between failure. The reliability function for exponential distribution is:
Have increasing failure rate pattern Are not standardized like electrical components
R(t ) = e (t )
(6)
199
For equipment that follows exponential distribution, the probability of having exactly k failures by time T is given by the Microsoft Excel function:
POISSON (k,t , False)

Some characteristics of the exponential distribution are:
(7)
Applies to situations where failure events are random and not due to wear, age, or deterioration Is a memoryless distribution. This means that the probability of failure is the same in all intervals of time. Has constant hazard rate This distribution is often applied to systems that are repairable.
Normal Distribution This distribution is applied to situations where the failures are due to wear. However mathematics of failure rate or hazard rate are complex. Log Normal Distribution This distribution has wide applicability to mechanical systems where failures are due to crack propagation, corrosion, and stresstemperature phenomenon. Weibull Distribution Weibull distribution (Dodson, 1994) is one of the most versatile of the failure distributions and has wide applicability for mechanical systems. It is defined by two parameters: called the characteristic life or scale factor and a constant called the shape parameter. The Weibull probability density function is:
Figure 1. The Weibull Probability Density Function.
f (t ) =
1 ( t /) t e
(8)
The Weibull reliability function is:
The characteristic life, , is the age at which 63.2 percent of the population will have failed. The shape parameter has several cases of interest in mechanical reliability assessment.
R(t ) = e ( t /)
(9)
Figure 2. Bathtub Curve for Mechanical and Electronic Components.
APPLICATIONS OF FAILURE DISTRIBUTIONS

Example 1 In this example there are two identical pumps in parallel as shown in Figure 3. They have negative exponential failure distribution and therefore:
When
When = 1This is a special case of Weibull distribution when
<1This indicates a decreasing hazard rate. In mechanical systems this is the initial run-in phase where faulty components with defects fail. With time these early failures diminish. This phase is often called the infant mortality or burn-in phase.
A = B =
(10)
When = 2the hazard rate is increasing linearly with time. When = 2.5the hazard rate is increasing and the distribution approximates the log normal distribution.
the distribution approximates a normal distribution. Figure 1 shows these various cases. Bathtub Curve This case is known as Rayleigh distribution.
it becomes an exponential distribution. As previously noted, for the exponential distribution the hazard rate is constant and failures are random. This phase designates the useful life of the component. The failure rate is reciprocal of the MTBF.
A B
Figure 3. Two Pumps in Parallel. Survival of one pump is sufficient to assure the success of the system. For this case the reliability of the system is given by:
When = 3.5For this case the hazard rate is increasing and
Rt = R A + RB R A RB
(11)
A bathtub curve is a plot of hazard rate against time. Figure 2 shows the curves for electronic and mechanical systems.
= 2e t e 2 t
(12)
200
Failure rate, = 1/ MTBF = 1/3 = 0.333 failures/year Mission time = 1 year a single pump, the reliability is: For 0.333 1
R = e = 71 percent and failure probability is 29 percent
Assume pump MTBF = 36 months (3 years)
For a parallel pump, system reliability is: R = 2e0.3331 e20.3331 = 1.4335 0.5137 = 92 percent and failure probability is 8 percent We will now evaluate the situation when one pump fails and the sister pump is operating without backup (Figure 4). Assume that pump mean time to repair (MTTR) = six days. Evaluating the probability of success for six days when the spare pump is operating without backup:
time = 6 days = 6/365 = 0.0164 years Repair 0.333 0.0164
R = e = 99.45 percent and failure probability is 0.55 percent
Installing a spare pump reduces the probability of system failure from 29 percent to less than 1 percent.
Figure 5. Reliability Versus Mean Time Between Failure and Repair Time. (Courtesy Bloch and Geitner, 1990) are two pumps in service and seven seal failures. Inputting the times to failure in the program yields the Weibull plot shown in Figure 6.
Figure 4. Only One Pump in Service. Example 2 A screw compressor has a failure rate of 0.0666 failure/year. What is the probability of two failures in exactly five years? This problem is solved by Equation (7).
Each Weibull distribution is applicable to a single failure mode of the equipment. Weibull plots as a straight line more data points give greater accuracy to the plot. Statistically The r2 in the plot gives an indication of the good fit. r2 = 1 is best. In engineering sometimes we are forced to work with fewer data
points. This increases the uncertainty of the plot.
POISSON(k,T,False)
T = 0.0666 5 = 0.333 k=2 The POISSON expression answer for the above values is 3.9 percent. Risk Based Maintenance With a spared system one is often faced with deciding if the repair of failed equipment should be carried out in an expedited or emergency basis. A typical situation is the repair strategy for, say, boiler feedwater pumps when the main pump has failed. Should it be repaired on an emergency basis? Plant operators have no confidence in the operating pump. Figure 5 can be used to aid in this decision. It relates the MTBF of the spare pump with MTTR or days unavailable and reliability. How long can the spared pump be out for repair so as not to compromise the target reliability of 99 percent. The following facts will aid in the decision:
Input Data: Reactor Seal Failures Failure # 1 2 3 4 5 6 7 Time to Failure 17 18 23 24 31 33 34
The spare pump is running satisfactorily. Its MTBF is 15 months.

In Figure 5 the x-axis is entered at 15 months and intersects the reliability line of 99 percent at a horizontal line that corresponds to an allowable outage of nine days. In this case there is no need to do emergency repair. Example 3 In this example we will use a commercially available Weibull program to plot the Weibull for a set of reactor pump seals. There
Figure 6. Weibull Plot for Reactor Pump Seals. The Beta () or the shape factor = 3.95 and the characteristic life () = 28.3 months. This suggests that for these pump seals hazard function is increasing and the seals have a wear-out mode. Table 2 is an example of a Microsoft Excel function reliability calculator that can be programmed. It takes Beta () and the characteristic life () from the Weibull plot and calculates the probability of failure and its converse, the reliability of the pump seals, based on the Weibull reliability function [Equation (9)] for a range of desired operating months.
201
Table 2. Reliability Calculator for Reactor Pump Seals, Based on Weibull.

Beta 3.948 Months Failure Probability 0.62 0.57 0.51 0.46 0.41 0.36 0.31 0.27 0.22 0.19 0.15 0.13 0.10 0.08 0.06 0.05 0.03 0.02 0.02 0.01 0.01 Reliability
Inputting a random number for the F(t) in Equation (14) yields an estimate of time to failure, t, for a given set of the Beta and Eta. With modern computers it is possible to execute several thousand simulations. Example 4
Eta (Months)
28.3
28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8
0.38 0.43 0.49 0.54 0.59 0.64 0.69 0.73 0.78 0.81 0.85 0.87 0.90 0.92 0.94 0.95 0.97 0.98 0.98 0.99 0.99
Reliability/Risk and Maintenance Planning A machinery engineer is faced with a decision to recommend to management if a turbine, which has operated successfully for four years, should be overhauled in the fifth or the sixth year of operation? In this example only three components of the turbine for which Weibull data are available from similar machines are considered. A commercially available reliability program was used for the simulation. Figure 7 shows the reliability block diagram (RBD) of the turbine and Table 4 shows the input data for the program.
Figure 7. Reliability Block Diagram of the Turbine. Table 4. Inputs for Turbine Components.
Mon Apr 25 14:06:48 2005 Block Name Failure Distro Param1 Param2 Param3 _________________________________ _(Beta)_____(Eta)___________ BladeErosion Weibull 4.20000 8.50000 0.00000 Controller-HydroCU Weibull 1.50000 11.0000 0.00000 TTV Weibull 4.70000 8.50000 0.00000 Valve-Control Weibull 1.50000 10.0000 0.00000
In Table 3, the Weibull equation is solved for critical values. From this table it is possible to get the expected time to failure for any desired reliability. Table 3. Inverse Weibull: Time to Failure for Desired Reliability.
Beta = 3.98 Eta = 28.28 Months Reliability 0.01 0.1 0.5 0.75 0.8 0.9 0.99 Time to Failure (Months) 41.637 34.932 25.773 20.627 19.341 15.993 8.820
The methodology for making a risk-based decision is as follows:
Step 1Assemble failure distribution for each component. The data for the turbine components are from the maintenance/failure histories from the operating plant, original equipment manufacturers (OEM) data, and public domain databases (Table 4). Step 2Make program simulations:
Number of program simulations: 1000 Mission times: five years and six years. The simulation program is run from time t = 0 to t = 6.
Step 3From the event log feature of the reliability program,
Reliability Simulations In multicomponent systems the mathematics gets too complex for closed form solutions. In such cases simulation is the next best approach. With the availability of computers, simulations have become relatively easy. The objective of the simulations is to predict the central tendency of a given variable. In this case it is to predict the probability of failure of a complex system. The algorithm for Monte Carlo simulations for mechanical systems is developed from the basic Weibull reliability function [Equation (8)], and noting that the failure function is:
assemble histograms of the number of failures for each component of the turbine for each successive year of mission time until the sixth year. The histograms essentially confirm the trend in the wear-out modes of the turbine components. Figure 8 shows a sample output from the event log and Figure 9 shows a histogram for the turbine components.
Starting Run 15 Time= Years 2.619166 Valve-Control Failed , TimeOperated=2.619166 System=Red 2.634584 Valve-Control Repaired, RepairTime=0.015419 System=Green 4.239981 Controller-HydroCU Failed , TimeOperated=4.224562 System=Red 4.243281 Controller-HydroCU Repaired, RepairTime=0.003300 System=Green 4.535766 TTV Failed , TimeOperated=4.517048 System=Red 4.548038 TTV Repaired, RepairTime=0.012272 System=Green 5.423212 BladeErosion Failed , TimeOperated=5.392222 System=Red 5.510067 BladeErosion Repaired, RepairTime=0.086855 System=Green 6.000000 Simulation Terminated End of Run #15
F (t ) = 1 R(t )
1
(13)
Taking logarithm of both sides, it yields the equation for the time to failure, t.
Figure 8. Sample Output from Event Log for Run #15.
1 Time to Failure = In 1 F (t )
(14)
Step
4Calculate the reliability of each component for a mission time of five and six years having survived four years. Use the Weibull reliability function [Equation (9)]:
202
Failure Trend of TTV Accumulated Over 20 Simulations
14 12 10 8 6 4 2 0 1 2 3 4 5 6 7 8

Failure Trend of Blade Erosion Accumulated Over 20 Simulations
6 5
# Failures
# Failures
4 3 2 1 0 1 2 3 4 5 Years 6 7 8 9
The analysis shows that there is an 11 percent additional risk in extending the operation from the fifth to the sixth year. The total risk is 19 percent for the control valve. Example 5
10
Years
Failure Trend of Hydraulic Control Unit Accumulated Over 20 Simulations

7 6 5 4 3 2 1 0 1 2 3 4 5 6 7 8 9 10 Years
5
Failure Trend of Valve Control Accumulated Over 20 Simulations
# Failures
# Failures
4 3 2 1 0 1 2 3 4 5 Years 6 7 8 9
A gas turbine generator (GTG) system is arranged in parallel as shown in Figure 11. Survival of one GTG line is sufficient for system success. All components have negative exponential distribution. The desire is to:
Forecast the system reliability and availability for a mission time of two years. Assess system vulnerability when one GTG is out for planned
maintenance for 10 days.
Figure 9. Histograms of Component Failures with Time.
R(t ) = e (t / n )
(15)
Sample calculations for trip and throttle valve (TTV):

Beta = 3.8 Eta = 9 years R(4 yr) = e at time = 4 years Reliability (4/9)3.8 = 95.5 percent
at time = 5 years Reliability R(5 yr) = e(5/9)3.8 = 89.8 percent Thus, R (5yr/having survived 4 yr) = R(5 yr)/R(4 yr) = 89.8/95.5 = 94.0 percent Failure probability = 1 94.0 = 6 percent at time = 6 years Reliability (6/9)3.8 R(6 yr) = e = 80.7 percent Figure 11. Two GTGs in Parallel. The reliability program model comprises all the components of the GTGs including the gearboxs, turbine and compressor sections, the generator, and the auxiliaries such as cooling water, lube system, control system, and fire suppression system. The results (Table 6) indicate that for two GTGs in parallel and for a mission time of two years, the system reliability will be 99 percent with a mean system failure rate of 0.01 failure in two years. The mean availability is >99 percent. Table 6. Results of Simulation: GTGs in Parallel, Mission Time = Two Years.
Mon Apr 25 15:20:35 2005 Results from 100 run(s): Parameter Total Costs Ao MTBDE MDT (1 runs) MTBM MRT (79 runs) %Green Time %Yellow Time % Red Time System Failures Minimum 66.00 0.991366829 1.982734 0.017266 0.283248 0.003004 95.527907 0.000000 0.000000 0 Mean 67.62 0.999913668 >1.999827 0.017266 >1.439166 0.013159 98.950059 1.041308 0.008633 0.010000 Maximum 72.66 1.0000000000 >2.000000 0.017266 >2.000000 0.030500 100.000000 4.065847 0.863317 1 Standard Deviation 1.37 0.000858989 n/a n/a n/a 0.006948 1.002106 0.975965 0.085899 0.099499
Thus, R (6 yr/ having survived 4 yr) = R(6 yr)/R(4 yr) = 80.7/95.5 = 84.5 percent Failure probability = 1 84.5 = 15.5 percent Table 5 and Figure 10 show the relative reliabilities and failure probabilities for the turbine components. Table 5. Relative Reliabilities and Probabilities of Failure for Turbine Components.
Component Reliability (5 yr /having survived 4 yr) 90.43 91.65 94.05 98.7 Failure Prob. (5 yr /having survived 4 yr) 9.57 8.35 5.95 1.3 Reliability (6 yr /having survived 4 yr) 80.9 83.2 84.5 96.4 Failure Prob. (5 yr /having survived 4 yr) 19.1 16.8 15.5 3.6
Control Valve Controller Hydraulic CU Trip and Throttle Valve Blade Erosion
Risk Assessment Probability of Failure (Having Survived 4 Yrs.)

25 20 15 10 5 0
R(t=2.000000) =0.990000
5 Yr. Run 6 Yr. Run
When one GTG is down for planned maintenance of 10 days, the results (Table 7) show that the system is vulnerable to failure within the 10 days. For the duration of the10 days the system reliability is only 97 percent and there is a potential of a mean failure of 0.03.
REFERENCES
Ireson, W. G., Coombs, C. F., Jr., and Moss, R. Y., 1996, Handbook of Reliability Engineering and Management, Second Edition, New York, New York: McGraw Hill. Dodson, B., 1994, Weibull Analysis, Milwaukee, Wisconsin: ASQ Press.
Figure 10. Relative Probability of Failure for Turbine Components (Example 3).
Bloch, H. and Geitner, F., 1990, Machinery Reliability Assessment, New York, New York: Van Nostrand Reinhold.
203
Table 7. Results of Simulation: One GTG Out for Overhaul, Mission Time = 10 Days.
Mon Apr 25 15:27:36 2005 Results from 100 run(s): Parameter Total Costs Ao MTBDE MDT (3 runs) MTBM MRT (3 runs) %Green Time %Yellow Time % Red Time System Failures Minimum 11.30 0.729199029 0.019973 0.004199 0.019973 0.000000 72.919903 0.000000 0.000000 0 Mean 11.33 0.993634398 >0.027216 0.005812 >0.027216 0.003339 99.363440 0.000000 0.636560 0.030000 Maximum 12.50 1.0000000000 >0.027390 0.007417 >0.027390 0.005819 100.000000 0.000000 27.080097 1 Standard Deviation 0.17 0.037137561 n/a 0.001314 n/a 0.002452 3.713756 0.000000 3.713756 0.170587
R(t=0.027390) =0.970000
204
TILTING PAD JOURNAL BEARING STARVATION EFFECTS

by John C. Nicholas
Owner and President Rotating Machinery Technology, Inc. Wellsville, New York
Greg Elliott
Senior Project Engineer Lufkin Industries Lufkin Texas
Thomas P. Shoup
Chief Engineer and Vice President of Operations Rotating Machinery Technology, Inc. Wellsville, New York
and Ed Martin
Project Engineer Lufkin Industries Lufkin Texas
John C. Nicholas is owner and President of Rotating Machinery Technology, Incorporated, in Wellsville, New York, a company that repairs and services turbomachinery, and manufactures bearings and seals. He has worked in the turbomachinery industry for 31 years in the rotor and bearing dynamics areas, including five years at Ingersoll-Rand and five years as Supervisor of the Rotordynamics Group at the Steam Turbine Division of Dresser-Rand. Dr. Nicholas received a B.S. degree (Mechanical Engineering, 1968) from the University of Pittsburgh and a Ph.D. degree in rotor and bearing dynamics (1977) from the University of Virginia. He holds several patents including one for a spray-bar blocker design for tilting pad journal bearings and another concerning by-pass cooling technology for tilting pad journal and thrust bearings. Dr. Nicholas, a member of ASME, STLE, and the Vibration Institute, has authored over 40 technical papers concerning rotordynamics and tilting pad journal bearing design and application. Greg Elliott is a Senior Project Engineer in the Power Transmission Division of Lufkin Industries in Lufkin, Texas. He works primarily in development, analysis, and design of high speed gear drives. He also provides support when finite element analysis, fatigue analysis, vibration analysis, or other assistance is needed in machinery design or problem solving. Previous activities have included development of Lufkins current N-D high speed gear product line. Mr. Elliott received a B.S. degree and an M.S. degree (Agricultural Engineering, 1982, 1990) from Texas A&M University.
1
Thomas P . Shoup is the Chief Engineer and Vice President of Operations at Rotating Machinery Technology, Incorporated, in Wellsville, New York. He has worked in the turbomachinery industry for 20 years in rotor and bearing system dynamics, including two years at the Steam Turbine Division of Dresser-Rand, five years at Jacobs/Sverdrup Technology, Inc., and 12 years at Siemens Demag Delaval Turbomachinery, Inc. Mr. Shoup is a member of ASME.
ABSTRACT
Improved turbomachinery aerodynamic performance requirements have increased journal bearing operating speeds and loads well above traditionally acceptable values. For example, for high performance gearboxes, pinion bearing surface speed requirements are often over 325 f/s with bearing unit loadings in the 500 psi range. In order to meet the design challenges for these severe applications, evacuated bearing housings have been utilized as an effective means of reducing journal bearing operating temperatures. Unfortunately, the use of evacuated housing designs has introduced a new and troubling phenomenajournal bearing starvation. This was never a problem with flooded designs with pressurized housings since any additional oil that may be required is simply drawn from the captured oil inside of the bearing housing. With the new evacuated housing designs, all required oil must be supplied by the oil inlet orifices. Often times, the amount of supply oil required to keep all pads from starving is well beyond reasonable. Thus, due to practicality, starvation in some form is allowed in almost all evacuated designs. This paper discusses evacuated journal bearing starvation and its possible detrimental effects on rotordynamics. Specifically, the
PROCEEDINGS OF THE THIRTY-SEVENTH TURBOMACHINERY SYMPOSIUM 2008
effect of starvation on journal bearing stiffness and damping is investigated. A case history is presented showing the effect of increasing oil flow on the location and amplification of a gearbox pinion critical speed during near zero load mechanical testing. As flow increased and the bearing became less starved, the location of the critical increased while the amplification decreased indicating a strong dependency of bearing stiffness and damping on oil flow. Concurrently, a similar but smaller bearing was tested under zero load starvation conditions. Essentially no effect on stiffness and damping was evident. From these results, the authors conclude that although increasing the oil flow solved the problem, starvation in itself was not the cause.
This paper discusses evacuated tilting pad journal bearing starvation and its possible detrimental effects on rotordynamics. Specifically, the effect of starvation on journal bearing stiffness and damping is investigated. A case history is presented showing the effect of increasing oil flow on the location and amplification of a gearbox pinion critical speed during no-load mechanical testing. Tilting pad journal bearing stiffness and damping test results will be presented for a zero load case with varying degrees of starvation (Harris and Childs, 2008). From these test results and the gearbox case history, the authors conclude that although increasing the oil flow solved the gear box problem, starvation in itself was not the cause.
INTRODUCTION
Improved turbomachinery aerodynamic performance requirements have increased journal bearing operating speeds and loads well above traditionally acceptable values. For example, for high performance gearboxes, pinion bearing surface speed requirements are often over 325 f/s with bearing unit loadings in the 500 psi range. In recent years, many gearbox applications have been above 350 f/s. Within the lead authors experience, the fastest journal bearing surface velocity for an American Petroleum Institute (API) gearbox is 389 f/s. Again, within the lead authors experience, faster surface velocities have been successfully achieved for high speed balancing applications with speeds up to 575 f/s. Achieving these extremely high surface velocities would not be possible with a 1970s vintage tilting pad journal bearing (TPJB). Early journal bearing designs were almost exclusively flooded. That is, the exit area for the oil was less than the oil inlet area. This created a positive pressure inside the bearing housing, thereby flooding the bearing with oil (Nicholas, 1994). In order to meet the design challenges for these severe applications with excessive surface velocities, evacuated bearing housings have been utilized as an effective means of reducing journal bearing operating temperatures. Tanaka (1991) presented experimental operating temperature data for a tilting pad journal bearing with bearing end seals (flooded and pressurized) and without bearing end seals (evacuated, nonpressurized). The bearing operated at lower temperatures without the end seals. Since then, many designs have been developed adopting the evacuated housing concept including Gardner (1994), Brockwell, et al. (1994), Ball and Byrne (1998), and Nicholas (2003). Unfortunately, the use of evacuated housing designs has introduced a new and troubling phenomenatilting pad journal bearing starvation. This was never a problem with flooded designs with pressurized housings since any additional oil that may be required is simply drawn from the captured oil inside of the bearing housing. With the new evacuated housing designs, all required oil must be supplied by the oil inlet orifices. Depending on the efficiency of the inlet oil supply mechanism, some oil escapes the bearing directly without participating in lubricating the pads. This certainly exacerbates the problem. Another issue with evacuated housing tilting pad journal bearings is the unloaded pads. For heavy loads, the loaded pads, with a much smaller journal-to-pad leading edge entrance area, require much less oil compared to the unloaded pads that have a much larger entrance area. In most cases, the amount of supply oil required to keep all pads, including the unloaded pads, from starving is well beyond reasonable. Finally, for high performance gearboxes, journal bearings are sized for peak performance at full load. The bearing is oversized for operation at near zero load during mechanical acceptance testing. Oil flow requirements for a full film on the loaded pads at full load results in starvation for all pads at near zero load. Again, the amount of supply oil required to keep all pads from starving at near zero load when the film thickness is larger is beyond reasonable. Thus, due to practicality, starvation in some form is allowed in almost all evacuated designs.
OIL FLOW REQUIREMENTS

This authors tilting pad journal bearing severe application experience plot is shown in Figure 1. The blue dots inside of the red box are API gearbox applications. Outside of the red box are test stand and high speed balance applications. Almost all applications shown on the plot are evacuated housing designs.
Figure 1. Tilting Pad Journal Bearing Severe Application Experience Plot. The steps used to determine the oil flow requirements for all of these applications are summarized below: 1. Assume that the hot oil carryover from pad-to-pad is equal to the amount of oil that passes through the pads minimum film thickness (refer to ACKNOWLEDGEMENT section). 2. Using this hot oil carryover amount, determine the minimum lubricating flow requirement for a full film on the loaded pad at full load and at full speed (i.e., no loaded pad starvation) and then multiply by the number of pads. This is the minimum per bearing lubricating oil flow requirement. 3. Increase the flow as necessary from the calculated minimum to meet the bearing operating temperature requirements. This results in a full film on the loaded pads at full load. It also results in starvation for the unloaded pads at any load condition and starvation for all pads during the no-load mechanical test. This design methodology worked successfully for all of the indicated Figure 1 applications except the one shown with the red triangle. Notice that it is safely within a batch of other successful applications as opposed to standing alone near the edge of the red box. Indeed, it is not the fastest application nor the most heavily loaded.
THE PROBLEM GEARBOX

The problem gearbox indicated by the red triangle in Figure 1 is a 24 MW, double helical, speed increaser driving three centrifugal compressors in offshore gas reinjection service. A photo of the box during mechanical testing is shown in Figure 2. The maximum
continuous pinion speed is 12,700 rpm. The 6.0 inch diameter tilting pad pinion bearings surface velocity is 333 f/s with a full load bearing unit load of 489 psi. The bearings geometric properties are summarized in Table 1. The actual bearing is shown in Figure 3. Note that this design does not use end seals. Also of note is the huge discharge opening between pads to enable the oil to easily exit the bearing. This bearing design is described in detail in Nicholas (2003).
Using the design steps outlined in the previous section, the minimum lubricating oil flow requirement for a full film on the loaded pads at full load and full speed is 26 gpm. The oil flow was increased to 34 gpm of lubricating oil plus 5 gpm of by-pass cooling oil (Nicholas, 2003) to properly cool the bearing for a total of 39 gpm. However, due to pressure from the customer to reduce oil flow, the bearings were shipped to the gear manufacturer with a total oil flow rate of 34 gpm (includes lubricating oil plus by-pass cooling oil). For reference, at zero load (i.e., gravity load only), the calculated full film lubricating oil flow requirement is 44 gpm. Adding in the 5 gpm of by-pass cooling flow, the total flow requirement for a full film at zero load is 49 gpm. Thus, it was anticipated that the bearing would be partially starved during the no-load mechanical test. Table 2 summarizes these results. Table 2. Pinion Bearing Oil Flow.
Figure 2. Problem Gearbox During Mechanical Testing. Running a pinion bearing partially starved during mechanical testing should not be a problem from the standpoint of load capacity. Since the bearing is sized for full load, it is obviously oversized at no-load. Partial starvation would, in effect, reduce the size of the bearing. At the leading edge of a partially starved bearing pad, there is not enough oil to fill the pad-to-journal gap. Thus, air is drawn into the pad and the initial section of pad is lubricated with an air-oil mixture. As this mixture moves farther into the pad, the film thickness decreases to the point where there is enough oil to fill the gap and a full film results. Thus, part of the pads leading edge is ineffective in providing load capacity during partial starvation. Since the bearing is operating in a zero load condition, this reduction in load capacity will not be a problem. However, the starved part of the pads leading edge will also not participate fully in providing the bearings stiffness and damping properties. Thus, some degradation in the bearings stiffness and damping is expected for partial starvation. Although the problem described in this paper would not mechanically harm the pinion or bull gear, it was significant logistically and commercially as it did cause this gearbox to fail the API mechanical acceptance test. Everyone involved may believe that this problem is only an artifact of the test conditions and that it would not manifest itself under real operating conditions. Nevertheless, to meet the API test specifications (API 613, 2003), and ship the gearbox, it was necessary to make the changes described in this paper. It is important to note that this gearbox ran flawlessly during the loaded string test. There was no indication of the problem described herein.
Figure 3. Tilting Pad Journal Pinion BearingEvacuated Housing Design. Table 1. Pinion and Test Tilting Pad Bearing Geometric Properties.
MECHANICAL TESTING
A speed-amplitude plot for the pinion from the initial no-load mechanical test run is shown in Figure 4 with a pinion bearing total oil flow of 34 gpm per bearing. For this plot and all other test results presented herein, an unbalance weight was placed on the coupling hub in order to excite and locate the pinions first critical speed. From Figure 4, note that the pinion critical speed is evident at about N1 = 13,500 rpm with an amplification factor of A1 = 13.5. This does not meet the API 613 (2003) acceptance criteria. Note that the terms
no-load and zero load are used herein to describe the mechanical test load, which, in actuality, was close to 5 percent of full load.
and, at the same time, decreasing the bearing clearance by 1.5 mils diametral (from 9.0 to 7.5 mils diametral, nominal), results in the speed-amplitude plot shown in Figure 7. The pinion critical appears to be at 15,250 rpm minimum with A1 = 10.9. This condition finally meets the API 613 (2003) acceptance criteria. These results are summarized in Tables 2 and 3.
Figure 4. Initial Pinion Response, Q = 34 gpm, Cb = 9.0 mils, Fabricated Baseplate. Increasing the inlet oil pressure appeared to push the critical up somewhat so the total bearing oil flow was increased to 45 gpm per bearing and the gearbox was retested. At the same time, since gearbox support stiffness was believed to be a significant factor, a stiffer solid baseplate replaced the original hollow, fabricated baseplate. The resulting speed-amplitude plot is shown in Figure 5. Now the pinion critical speed is at 14,200 rpm with a corresponding amplification factor of 14.2.
Figure 7. Final Pinion Response, Q = 55 gpm, Cb = 7.5 mils, Solid Baseplate. Table 3. Summary of Mechanical Test Results.
Figure 5. Interim Pinion Response, Pin = 25 psi, Q = 45 gpm, Cb = 9.0 mils, Solid Baseplate. Since this still does not meet the API 613 (2003) acceptance criteria, the inlet pressure was increased from the design value of 25 psi to 45 psi in an attempt to easily check the effects of increasing oil flow. The corresponding per bearing total oil inlet flow increase was from 45 to 60 gpm. The resulting response run is shown in Figure 6 with N1 greater than 15,200 rpm.
It is important to note that a similar problem also occurred on the low speed shaft with a maximum continuous speed of 6865 rpm. The low speed shaft problem was solved in a similar manner as described above. Due to space constraints, only the high speed pinion problem will be discussed herein. Additionally, the problem also occurred on the spare gearbox. Finally, subsynchronous vibration for both the pinion and the bull gear shafts was not an issue. Spectrum plots show very low subsynchronous vibration levels throughout the no-load mechanical and full load string tests.
ANALYTICAL CORRELATION
In an attempt to match the no-load test results of Figure 4 (N1 = 13,600 rpm, A1 = 13.5), a rotordynamics analysis was performed on the pinion. Initially, the tilting pad bearing analysis assumed a full film as the bearing code used in the analysis did not have the capability to calculate any starvation effects explicitly (Nicholas, et al., 1979). A reasonable gearbox case support stiffness value of Ks = 7.0106 was assumed. The nominal as-shipped bearing clearance of Cb = 9.0 mils diametral was also used. The resulting speed-amplitude plot is presented in Figure 8. The pinion critical is predicted at 13,700 rpm with an amplification factor of 7.6. The frequency is a good match but the amplification is predicted to be about 50 percent of the actual value indicating far less damping in the system than anticipated.
Figure 6. Interim Pinion Response, Pin = 45 psi, Q = 60 gpm, Cb = 9.0 mils, Solid Baseplate. Based on the favorable Figure 6 results, another increase in oil inlet flow seemed appropriate. Further increasing the total oil flow to 55 gpm per bearing (50 gpm lubricating plus 5 gpm by-pass)
Figure 8. Predicted Pinion Response, Full Film, Cb = 9.0 mils, Ks = 7.0106 lbf/in.
Figure 5 test results (N1 = 14,200 rpm, A1 = 14.2) were obtained with 45 gpm of total per bearing oil flow plus the stiffer, solid baseplate. The original fabricated, hollow baseplate is shown in Figure 9. In an attempt to increase the pinions critical speed by increasing the gearbox support stiffness, a new solid baseplate, Figure 10, replaced the original prior to the Figure 5 test. A rap test was performed on the gearbox with the new, solid baseplate. Results indicated a dynamic support stiffness of 20.0106 lbf/in at a frequency of 14,000 cpm. Using Ks = 20.0106 lbs/in and the nominal as-shipped bearing clearance of Cb = 9.0 mils diametral, results in the speed-amplitude plot shown in Figure 11. The pinion critical is now predicted at 17,200 rpm with an amplification factor of 5.5. The frequency is over predicted by 3,000 rpm and the amplification factor is under predicted by a factor of 2.6. These results are summarized in Table 4.
Table 4. Summary of Analytical Results.
STARVATION MODELING
In an attempt to better match the no-load test results with analytical predictions, starvation is included in the tilting pad journal bearing analysis. A tilting pad journal bearing computer code developed by He (2003) was used to predict the angle from the pads leading edge where a full film would occur. This predicted angle is 20 degrees. A simplistic approach would be to assume that the pad arc length is effectively reduced by 20 degrees, from 70 degrees to 52 degrees. This also reduces the pad pivot offset from the as-machined 65 percent to an effective value of 50 percent. Using these effective values in the original bearing code, Nicholas, et al. (1979), along with Cb = 9.0 mils diametral and Ks = 20.0106 lbs/in results in the speed-amplitude curves shown in Figure 12. Now the pinion critical is critically damped. Clearly, this model predicts too much damping.
Figure 9. Original Fabricated Hollow Baseplate.
Figure 12. Predicted Pinion Response, Starvation Model with 52 Degree Pad Arc Length, 50 Percent Pad Pivot Offset, Cb = 9.0 mils, Ks = 20.0106 lbf/in. Artificially decreasing bearing stiffness and damping independent of each other until the results match Figure 5 (N1 = 14,200 rpm, A1 = 14.2) with Ks = 20.0106 lbs/in produces the speed-amplitude curve shown in Figure 13. Now N1 = 14,700 rpm and A1 = 11.3, a reasonable match to test results. To obtain this match, the bearing stiffness was decreased to 70 percent of the full film value, which is reasonable, but the bearing damping was decreased to 15 percent of the full film value, an 85 percent decrease, which is quite unreasonable.
Figure 10. New Solid Baseplate.
Figure 11. Predicted Pinion Response, Full Film, Cb = 9.0 mils, Ks = 20.0106 lbf/in.
Figure 13. Predicted Pinion Response, Starvation Model with K = 70 Percent and C = 15 Percent of Full Film Values, Cb = 9.0 mils, Ks = 20.0106 lbf/in.
Returning to the code by He (2003), plotting the normalized bearing stiffness, K, and damping, C, as a function of lubricating oil flow results in the plot shown in Figure 14. The code does predict a drastic decline in K and C, but they decrease at about the same rate. The dots shown on the plot indicate a stiffness value that is 70 percent of full film and a damping value that is 15 percent of full film. The K value that is 70 percent of full film occurs at a predicted lubricating flow rate of 65 gpm while the C value that is 15 percent of full film extrapolates out to a lubricating flow rate of 36 gpm. Since both flow rates cannot occur at the same time, 70 percent of full film K and 15 percent of full film C also cannot occur at the same time. But, they would have to occur together in order to match the test results. Thus, it is difficult to envision a starvation model that matches the test results. These results are summarized in Table 4.
Figure 16. 44 Inch TPJB Zero Load Test DataPrincipal Damping Versus Oil Flow. From Figure 15, the test curves show a barely perceivable decline in bearing stiffness as the total oil flow decreases from a full film value of 16 gpm to a starved value of 10 gpm. From Figure 16, the test results indicate that the direct damping increases slightly and then declines as flow decreases. Certainly, no decrease in damping is evident from Figure 16 that approaches the 15 percent of full film value discussed previously. Also notice that full film analytical predictions from Nicholas, et al. (1979), are included on both plots. The nominal as-built bearing clearance was used for the analysis. Pivot stiffness was included and calculated by the Hertzian method from Kirk and Reedy (1988) and Nicholas and Wygant (1995).
ULTRA HIGH SPEED APPLICATION

Figure 14. Starvation Model, Normalized K and C Versus Oil Flow Showing K = 70 Percent and C = 15 Percent of Full Film Values. Soon after the problem gearboxes were shipped, a similar evacuated tilting pad bearing design with a 6.65 inch journal diameter was used for a high speed balance of a magnetic bearing rotor. This application is shown on the plot of Figure 1 in the extreme lower right-hand corner, 575 f/s surface velocity, 26 psi unit loading. The tilting pad bearings, used for the high speed balance only, supported the rotor on the magnetic bearing laminations. Since the laminations had a relatively large outside diameter, the tilting pad bearing surface velocity was extremely high with the evacuated bearing running up to 435 f/s. The actual bearing is shown in Figure 17. The journal surface velocity versus metal temperature plot from one of the initial test runs using the evacuated design is shown in Figure 18, blue line with solid blue circles (refer to ACKNOWLEDGEMENT section).
STARVATION TESTING
Coincidently, as the gearbox was experiencing problems with a 6.0 inch tilting pad journal bearing with an evacuated housing, a 4.0 inch diameter evacuated housing bearing was undergoing laboratory testing (Harris and Childs, 2008). Except for the test bearing being geometrically smaller, the designs were identical. Some of the geometric parameters are four pads, load between pivots, 65 percent pad pivot offset and L/D = 1.0 (Table 1). A special test was requested at 12,000 rpm and at zero load with the bearing flow rates varying from a full film to a starved condition (Table 5). These results are shown in Figures 15, direct stiffness, and 16, direct damping. Table 5. Test Bearing Oil Flow.
Figure 17. Ultra High Speed ApplicationEvacuated TPJ Bearing.
Figure 15. 44 Inch TPJB Zero Load Test DataPrincipal Stiffness Versus Oil Flow.
Figure 18. Evacuated Versus Flooded TPJB, High Speed Balance on Laminations.
The minimum oil lubricating flow requirement for operation at 435 f/s with the evacuated design is 36 gpm calculated as described in the OIL FLOW REQUIREMENTS section. Because of flow restrictions in the high speed balance facility, the evacuated bearings were designed for 28 gpm of lubricating flow. After the test resulting in the evacuated bearing data shown in Figure 18, it was determined that one of the oil pumps was not operational. Thus, the resulting lubricating flow was 20 gpm and the bearing ran 56 percent starved. Furthermore, because of a misunderstanding, it was believed during the run that the oil drains were pressurized. This was not the case and the evacuated bearing operated in a vacuum, further starving the bearing due to oil atomization. Even in this extreme starvation condition, the rotor critical was located as anticipated with a reasonable amplification factor.
For an evacuated housing journal bearing, this type of entrainment may occur in the starvation region at the leading edge of a tilting pad. However, if there is inadequate dwell time for the oil in the reservoir, air bubbles will be present in the oil when it is reintroduced to the bearing. Several weeks after the gearbox was shipped, the amount of air entrainment present in the oil cooler was measured at less than 5 percent. From Figures 4 and 6 (from Tao, et al., 2000), this is not nearly enough to account for an 85 percent reduction in full film damping. Mesh Oil and Air Impingement It may be possible that the oil that exits the gear mesh may jet into the bearing and interfere with the inlet oil or the lubricating oil film. Another factor that may affect the bearing is the windage from the gear mesh. Since the housing is evacuated, there are no end seals on the bearing. Thus, the journal-to-pad interface and the lubricating film are exposed to the oil that is forced out of the gear mesh and from mesh windage. On this gearbox, the oil exits the gear mesh at a velocity of around 1000 f/s. This may be sufficient to interfere with the lubricating film thereby affecting the bearings stiffness and damping properties. Whether this effect is sufficient to cause an 85 percent damping decrease from full film levels is unknown. To prevent these phenomena from occurring, some gearboxes have a shield at the end of the gear mesh. A mesh shield was not present on the problem gearbox nor has it been used on most of the applications shown in Figure 1. Another way to eliminate mesh oil or air impingement is to place an end shield on the mesh side of the journal bearings. No shields were present on the bearings for this gearbox. However, the inlet oil is protected from mesh impingement by side shields on the spray-bar blocker as shown in Figure 19.
POSSIBLE CAUSES
As stated previously, this gearbox pinion rotor experienced a vibration problem during no-load mechanical testing. The location of the pinions critical speed was well below predicted and the amplification factor was well above predicted. Increasing oil flow increased the critical speed location but had a minor effect on reducing the amplification factor. A similar problem occurred on the low speed shaft. It was solved in a similar manner. In the authors experience with dozens of gearboxes built in a similar manner with essentially the same evacuated tilting pad bearing, this was the first gearbox to exhibit this problem. However, it must be noted that unbalance testing was not conducted on all of the gearboxes and similar problems of this type may have been missed. Clearly, increasing oil flow had a large influence on the critical speed location. This seems to indicate that starvation was the cause of the problem. However, there are contrary issues that seem to negate this conclusion:
With all of the experience shown in Figure 1, it would seem
likely that this problem would have manifested itself previously since all of the bearings were similar in design, have evacuated housings, and were designed to operate partially starved at no-load.
A similar problem did not occur during the magnetic bearing It
rotor high speed balance with a similar evacuated tilting pad bearing design, at very light bearing loads, and in an extreme starvation condition. is a highly unlikely coincidence that the problem finally showed itself on both rotors at the same time on the same gearbox.
Increasing the total per bearing oil flow from 34 to 55 gpm, from
partial starvation to a full film, increased the critical speed frequency but the amplification factor remained unreasonably high.
An
85 percent decrease in full film bearing damping is necessary to match the test data. Laboratory test results at no-load for a similar but smaller bearing did not show a dramatic decrease in bearing damping or stiffness as starvation increased. Analytical starvation modeling does not predict this dramatic decrease in bearing damping necessary to match the test stand results. The conclusion is that increasing oil flow helped to solve the problem but was not the cause. It may have been a contributor, but not the sole cause of the problem. Other possible causes or contributors are outlined below. Air Entrainment Air entrainment is a condition where air bubbles are trapped or entrained in the lubricating oil. It is a well-known phenomenon for squeeze film dampers. It has been shown that air entrainment can drastically decrease the damping provided by a squeeze film damper (refer to Figures 4 and 6, Tao, et al., 2000). Figure 19. Spray-bar blocker Side Shields Protecting the Inlet Oil. Support Stiffness From Figure 8, with a reasonable gear case support stiffness of 7.0106 lbf/in, the pinion critical is predicted at 13,700 rpm. However, the predicted amplification factor remains well below actual. Regardless, with the solid baseplate, the support stiffness was measured at 20106 lbf/in. This is a relatively high value and, therefore, the problem cause is not likely to be a soft gear case support.
Pivot Stiffness To obtain all of the analytical results, the spherical pivot stiffness was included in the calculations. Using the method outlined in Kirk and Reedy (1988) and Nicholas and Wygant (1995), the calculated Hertzian spherical pivot stiffness is Kp = 24.7106 lbf/in. The bearings spherical pivot was reconstructed and its stiffness measured using a hydraulic press to simulate the pivot load. The measured pivot stiffness was 17.4106 lbf/in. Using the measured value for Kp, all of the analytical results show virtually no change. Thus, the problem cause is not likely to be a soft pivot. Pad Flutter Pad flutter is a tilting pad bearing phenomenon that may occur on the pads that are located opposite of the load vector. If these pads become unloaded (Nicholas, 1994), they may not be able to find an equilibrium position. Pad flutter may cause babbitt damage but rarely leads to high synchronous vibration. Furthermore, pad flutter occurs under heavy loads. Since this problem occurs under light loads, pad flutter is not a consideration. Increasing Gearbox Performance Requirements Gearbox performance requirements have been steadily increasing. As gearbox transmitted torques have increased, gear mesh face widths have also increased to carry the load while keeping gearbox shaft center distances small and gear pitch line velocities down. Wider face widths lead to longer rotors. Additionally, carburized gearing is often used to meet these higher power and torque levels. The higher tooth surface hardness achieved by carburizing permits considerably more torque to be transmitted than would be possible with a through-hardened gear of the same physical size. Bearing technology advancements are permitting more load to be carried by smaller bearings with reduced oil flow and power loss. These factors combine to result in heavier couplings for torque transmittal without a corresponding increase in the rotor diameter. Longer rotors and heavier couplings decrease the rotors critical speeds. At the same time, operating speeds are increasing. All of these factors apply to this gearbox. Further compounding the problem, the flexible couplings on this gearbox utilize hydraulic taper fits, resulting in even heavier couplings and higher overhung moments, further reducing the critical speeds. Figure 20. Pinion Mode Shape with Coupling Hub Attachment.
Figure 21. Predicted Pinion Response, Starvation Model with K = 70 Percent and C = 15 Percent of Full Film Values, Cb = 9.0 mils, Ks = 20.0106 lbf/in, with Integral Flange.
Another possibility to eliminate this problem is to use a properly

cooled flooded bearing. The solution for the low speed shaft on this gearbox included switching to a flooded bearing design. While the operating temperatures increased compared to the evacuated design, they were still within the customers specification. Also, it was estimated that the gearbox power loss increased by roughly 15 percent. A flooded design can be successful in severe application if proper cooling features are employed. An example is shown in Figure 18, green line, solid green triangles. This was the magnetic bearing rotor high speed balance application discussed previously in the ULTRA HIGH SPEED APPLICATION section. After some initial runs up to 435 f/s with the evacuated design, the bearings were changed to a flooded design with special cooling features. The special flooded tilting pad bearings ran successfully up to a surface velocity of just over 500 f/s. As expected, the flooded design ran slightly hotter. For example, at 435 f/s, the flooded bearing ran 20F hotter compared to the evacuated bearing.
PROPOSED SOLUTIONS AND RECOMMENDATIONS

While the exact cause of the problem is unclear, some solutions and recommendations are suggested below. Keep in mind that, while gearbox mechanical tests are run at loads in the range of 5 to 10 percent of full load, this light load condition for centrifugal compressor trains is not realistic in the field. Specifically, operation at 10 percent of full load and at maximum speed is unrealistic for centrifugal compressor trains as the minimum load at maximum speed would be much greater than 10 percent. The solutions and recommendations follow: moment is highly recommended. From the pinion mode shape illustrated in Figure 20, the pinion critical is clearly controlled by the coupling end overhang moment. This pinion coupling employed a shrunk-on hub. An integral flange attachment would greatly reduce the overhung moment. From Figure 13, N1 = 14,700 rpm and A1 = 11.3 with a coupling hub attachment. This analysis used the artificially degraded bearing stiffness and damping properties to match the test results. Replacing the hub attachment with an integral flange attachment with a lower overhung moment, results in Figure 21. Now, N1 = 16,200 rpm and A1 = 9.0 and this problem would not have materialized. For this gearbox, there were no engineering reasons to use a hub attachment instead of an integral flange.
When
The use of an integrally flanged coupling with a low overhung
using an evacuated housing design, properly size the bearing oil flow. The final configuration for this pinion bearing ended up with 55 gpm of total per bearing oil flow. While this is 114 percent of the minimum lubricating flow for a full film at zero load, it is a staggering 192 percent of the minimum flow for a full film at full load. The authors suggest sizing the flow for 150 percent of the minimum flow for a full film at full load.
Use a realistic gear case support stiffness value in the forced

response analysis.
Include the pivot stiffness in the bearing dynamic analysis. When evacuated housing bearings are used, a conservative
analytical separation margin should be employed.
CONCLUSIONS
The
subject gearbox pinion rotor experienced a vibration problem during no-load mechanical testing.
The location of the pinions critical speed was well below predicted and the amplification factor was well above predicted. Increasing oil flow increased the critical speed location but had a minor effect on reducing the amplification factor.
Acknowledge that operation at 10 percent of full load at maximum speed is unrealistic for field operation of centrifugal compressor trains. Relax the no-load mechanical test acceptance criteria. This will allow the bearing designers to design for full load and not for the no-load mechanical test. It would permit the elimination of otherwise unnecessary oil flow and power loss, and thus reduce the operating cost of the equipment, without reducing reliability.
Increasing oil flow had a large influence on the critical speed
location. This seems to indicate that starvation was the cause of the problem. However, there are contrary issues that seem to negate this conclusion. The major issues are outlined below: With all of the experience shown in Figure 1, it would seem likely that this problem would have manifested itself previously since all of the bearings were similar in design, had evacuated housings, and were designed to operate partially starved at no-load. However, not all of the gearboxes were unbalance tested, and this problem may have been overlooked on past gearboxes. A similar problem did not occur during the magnetic bearing rotor high speed balance with a similar lightly loaded, evacuated tilting pad bearing design in an extreme starvation condition. An 85 percent decrease in full film bearing damping is necessary to analytically match the test data. - Analytical starvation modeling does not predict this dramatic decrease in bearing damping necessary to match the test stand results. Laboratory test results at no-load for a similar but smaller bearing did not show a dramatic decrease in bearing damping or stiffness as starvation increased during no-load testing.
This gearbox performed flawlessly during the full load string test. No vibration problems were experienced. The bearing temperatures were all well within specifications.
In summary, the subject gearbox pinion and bull gear rotors experienced vibration problems during no-load mechanical testing with evacuated housing tilting pad journal bearings. The bearings were operating at essentially zero load in a partially starved condition. Test results indicate that the location of both rotor critical speeds were well below predicted with the amplification factors well above predicted. Increasing the oil flow increased the location of both critical speeds thereby solving the vibration problem. From this, one may conclude that bearing starvation was the problem cause. However, independent bearing testing in a starved condition at zero load did not induce a significant bearing stiffness or damping decrease. Thus, it is concluded that starvation alone did not cause the problem. The most probable cause was starvation in conjunction with another gearbox and/or test stand related phenomena: either mesh air impingement, mesh oil impingement, or air entrainment in the lubricating oil.
From the above, it is concluded that increasing oil flow helped

to solve the problem but was not the cause. It may have been a contributor, but not the sole cause of the problem.
NOMENCLATURE
A1 Cb C Cxx, Cyy Dj K Kp Ks Kxx, Kyy m N1 Pin Q Tmax Wp = First critical speed amplification factor (rpm) = Bearing diametral clearance (mils) = Bearing principal damping (lbf-s/in) = Horizontal, vertical principle bearing damping, kN-s/m (lbf-s/in) = Journal diameter (in) = Bearing principal stiffness (lbf/in) = Pivot stiffness (lbf/in) = Gear case support stiffness (lbf/in) = Horizontal, vertical principle bearing principle stiffness, MN/m (lbf/in) = Pad preload = First critical speed (rpm) = Oil inlet pressure (psi) = Oil flow (gpm) = Maximum bearing operating metal temperature (F) = Pivot load (lbf)
Other possible causes or contributors include:

Mesh oil and air impingement on the oil film. Mesh oil impingement is more likely to be a cause for concern with single helical gearboxes whereas the problem gearbox has a double helical mesh. Air impingement from mesh windage is obviously present in all gearboxes. Air entrainment in the lubricating oil.
The use of an integrally flanged coupling with a low overhung

moment is highly recommended. If one were used on this application, this problem would not have occurred.
An unbalance test is recommended for all gearboxes with pinion speeds above 8000 rpm to locate both the bull gear and pinion rotor critical speeds. Consider the use of a properly cooled flooded bearing.
This was the resulting configuration for the low speed shaft bearings. Operating temperatures will increase. Power loss will increase. Proven successful in severe applications if properly designed with appropriate cooling features.
REFERENCES
API Standard 613, 2003, Special-Purpose Gear Units for Petroleum, Chemical and Gas Industry Services, Fifth Edition, American Petroleum Institute, Washington, D.C. Ball, J. H. and Byrne, T. R., 1998, Tilting Pad Hydrodynamic Bearing for Rotating Machinery, US Patent No. 5,795,076, Orion Corporation, Grafton, Wisconsin. Brockwell, K., Dmochowski, W., and DeCamillo, S. M., 1994, Analysis and Testing of the LEG Tilting Pad Journal BearingA New Design for Increasing Load Capacity, Reducing Operating Temperatures and Conserving Energy, Proceedings of the Twenty-Third Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 46-56. Gardner, W. W., 1994, Tilting Pad Journal Bearing Using Directed Lubrication, US Patent No. 5,288,153, Waukesha Bearing Corporation, Waukesha, Wisconsin.
When
using an evacuated housing design, properly size the bearing oil flow. A design flow equal to 150 percent of the minimum lubricating flow for a full film at full load is a suggested target.
Use a realistic gear case support stiffness and include the pivot
stiffness in the bearing and forced response analyses.
When
evacuated housing bearings are used, a conservative analytical separation margin should be employed.
Bearing manufacturers, gear vendors, compressor manufacturers,
and the end users all need to work together to help prevent similar problems and to provide the best possible system. To this end:
10
Harris, J. and Childs, D., 2008, Static Performance Characteristics and Rotordynamic Coefficients for a Four-Pad Ball-in-Socket Tilting Pad Journal Bearing, Proceedings of ASME Turbo Expo 2008: Power for Land, Sea and Air, GT2008-50063. He, M., 2003, Thermoelastohydrodynamic Analysis of Fluid Film Journal Bearings, Ph.D. Dissertation, University of Virginia, Charlottesville, Virginia. Kirk, R. G. and Reedy, S. W., 1988, Evaluation of Pivot Stiffness for Typical Tilting-Pad Journal Bearings Designs, ASME Journal of Vibration, Acoustics, Stress and Reliability in Design, 110, (2), pp. 165-171. Nicholas, J. C., 1994, Tilting Pad Bearing Design, Proceedings of the Twenty-Third Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 179-194. Nicholas, J. C., 2003, Tilting Pad Journal Bearings with Spray-Bar Blockers and By-Pass Cooling for High Speed, High Load Applications, Proceedings of the Thirty-Second Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 27-38. Nicholas, J. C. and Wygant, K. D., 1995, Tilting Pad Journal Bearing Pivot Design for High Load Applications, Proceedings of the Twenty-Fourth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 33-48. Nicholas, J. C., Gunter, E. J., and Allaire, P. E., 1979, Stiffness and Damping Coefficients for the Five Pad Tilting Pad Bearing, ASLE Transactions, 22, (2), pp. 112-124.
Tanaka, M., 1991, Thermohydrodynamic Performance of a Tilting Pad Journal Bearing with Spot Lubrication, ASME Journal of Tribology, 113, (3), pp. 615-619. Tao, L., Diaz, S., San Andrs, L., and Rajagopal, K. R., 2000, Analysis of Squeeze Film Dampers Operating with Bubbly Lubricants, ASME Journal of Tribology, 122, (1), pp. 205-210.
BIBLIOGRAPHY
Nicholas, J. C., Whalen, J. K., and Franklin, S. D., 1986, Improving Critical Speed Calculations Using Flexible Bearing Support FRF Compliance Data, Proceedings of the Fifteenth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 69-78.
ACKNOWLEDGEMENT
The authors recognize Lufkin Industries for their assistance and support, Texas A&M Turbomachinery Laboratory for the tilting pad journal bearing test data, and GE Oil & Gas, Nuovo Pignone, for the magnetic bearing rotor, on-lamination, high speed balancing test data. Also, the authors thank Marty Maier, Dresser-Rand, for suggesting that the maximum amount of hot oil carryover must traverse the bearing pads minimum film thickness. Finally, thanks to Dr. John Kocur, ExxonMobil, for providing some of the starvation bearing data.
JOURNAL BEARING VIBRATION AND SSV HASH

by Scan M. DeCamillo
Manager, Research and Development Kingsbury, Inc. Philadelphia, Pennsylvania
Minhui He
Machinery Specialist
C. Hunter Cloud
President
and James M. Byrne

Machinery Consultant BRG Machinery Consulting, LLC Charlottesville, Virginia 22903 USA
Scan M. DeCamillo is Manager of Research and Development for Kingsbury, Inc., in Philadelphia, Pennsylvania. He is responsible for design, analysis, and development of Kingsbury fluid film bearings for worldwide industrial and military applications. He began work in this field in 1975 and has since provided engineering support to industry regarding application and performance of hydrodynamic bearings. Mr. DeCamillo has developed performance and structural bearing analysis tools during his career, establishing design criteria used in many publications and specifications. He has patents and has authored several papers on bearing research, which is currently focused on advancing hydrodynamic bearing technology in high-speed turbomachinery. Mr. DeCamillo received his B.S. degree (Mechanical Engineering, 1975) from Drexel University. He is a registered Professional Engineer in the State of Pennsylvania and a member of STLE, ASME, and the Vibration Institute. Minhui He is a Machinery Specialist with BRG Machinery Consulting LLC, in Charlottesville, Virginia. His responsibilities include vibration troubleshooting, rotordynamic analysis, as well as bearing and seal analysis and design. He is a member of STLE, and is also conducting research on rotordynamics and hydrodynamic bearings. Dr. He received his B.S. degree (Chemical Machinery Engineering, 1994) from Sichuan University. From 1996 to 2003, he conducted research on fluid film journal bearings in the ROMAC Laboratories at the University of Virginia, receiving his Ph. D. (Mechanical and Aerospace Engineering, 2003).
11
C. Hunter Cloud is President of BRG Machinery Consulting, LLC, in Charlottesville, Virginia, a company providing a full range of rotating machinery technical services. He began his career with Mobil Research and Development Corporation in Princeton, NJ, as a turbomachinery specialist responsible for application engineering, commissioning, and troubleshooting for production, refining, and chemical facilities. During his 11 years at Mobil, he worked on numerous projects, including several offshore gas injection platforms in Nigeria, as well as serving as reliability manager at a large US refinery. Dr. Cloud received his B.S. (Mechanical Engineering, 1991) and Ph.D. (Mechanical and Aerospace Engineering, 2007) from the University of Virginia. He is a member of ASME, the Vibration Institute, and the API 684 rotordynamics task force. James M. Byrne is currently a member of the BRG Machinery Consulting team, in Charlottesville, Virginia. BRG performs research and analysis in the fields of fluid film bearings, magnetic bearings, and rotordynamics. Mr. Byrne began his career designing internally geared centrifugal compressors for Carrier in Syracuse, New York. He continued his career at Pratt and Whitney aircraft engines and became a technical leader for rotordynamics. Later Mr. Byrne became a program manager for Pratt and Whitney Power Systems managing the development of new gas turbine products. From 2001 to 2007, he was President of Rotating Machinery Technology, a manufacturer of tilting pad bearings. Mr. Byrne holds a BSME degree from Syracuse University, an MSME degree from the University of Virginia, and an MBA from Carnegie Mellon University.
12
ABSTRACT
Peculiar, low-frequency, radial vibrations have been observed in various turbomachinery using tilt-pad journal bearings. Unlike discrete subsynchronous spikes that often indicate a serious problem, the vibrations are indiscrete and of low frequency and amplitude. The low level shaft indications have raised concern in witness tests of critical machinery, even in cases that comply with American Petroleum Institute (API) limits, owing to uncertainty regarding the cause and nature of the vibrations. This paper presents shaft and pad vibration data from various tilt-pad journal bearing tests that were undertaken to investigate and better understand these subsynchronous indications. The vibration characteristics are defined and compared under the influence of speed, load, oil flow, and bearing orientation. Results are presented for conventional and direct lube tilt-pad bearing designs, along with discussions of parameters and methods that were successful in reducing and eliminating these low level vibrations. The test results indicate that the low-frequency shaft indications are caused by pad vibration. Hypotheses and analyses are presented and discussed in relation to the test observations.
that these low-frequency vibrations have been encountered in turbines, compressors, and gearboxes, using conventional and direct lube tilt-pad bearing designs. The signature has also been documented in separate research investigating stability (Cloud, 2007). Accurate stability prediction is a major topic of concern in the industry. Kocur, et al. (2007), highlight a large spread in predicted bearing stiffness and damping coefficients among computer codes, and associated ramifications regarding stability assessment of critical turbomachinery. At the same time, there are not many codes available that address direct lube performance and dynamic coefficients (He, 2003; He, et al., 2005). Researchers are presently attempting to sort through some difficult questions. What is the source of the low-frequency vibrations? Are they attributable to direct lube bearings, pad flutter, starvation, high speeds, or low loads? Do they affect machine stability, safety, or reliability? This paper presents a chronology of tests, investigations, and theoretical analyses aimed at providing answers to some of these questions. It is desired that the information be of value to researchers, OEMs, users, and other personnel involved with hydrodynamic bearings and vibration in turbomachinery.
INTRODUCTION
Many early technical papers report on the inherent stability of tilt-pad journal bearings in overcoming oil whirl and oil whip limitations of fixed geometry bearings. There are, however, other vibration phenomena associated with tilt-pad bearings that are topics of past and present research. In the late 70s, extensive babbitt fatigue cracking of upper, unloaded pads was a major problem in large, conventional tilt-pad journal bearings. These were flooded designs, which incorporate end seals to restrict oil outlet and flood the bearing cavity. A well-referenced study by Adams and Payandeh (1983) attributes the damage to self-excited, subsynchronous pad vibration. Trends for larger and higher speed turbomachinery impose greater demands on bearings and rotor dynamics. Direct lube bearings have evolved to reduce the higher power losses, oil flow requirements, and pad temperatures associated with higher surface speeds. Direct lubrication is not a new concept as designs have been used in special applications with a long history of reliability. Use of direct lube journal bearings became more prevalent as machine size and speed increased, eventually spawning several papers in the early 90s investigating their steady-state performance. These include research by Booser (1990), Tanaka (1991), Harongozo, et al. (1991), Brockwell, et al. (1992), and Fillon, et al. (1993). The references document pad temperature reductions derived from the efficient evacuation of hot oil from the bearing cavity. Direct lube bearings are therefore typically designed for evacuated operation, accomplished by opening up end seal and oil outlet restrictions. A direct application of oil to the journal surface prevents oil from bypassing the films, which can occur in conventional bearings when outlet restrictions are removed. Use of nozzles to spray oil on the journal surface between pads is one method of direct lubrication. An additional pad temperature advantage is gained by direct lubrication features. Literature on vibration characteristics of direct lube journal bearings is not as prevalent. DeCamillo and Clayton (1997) present rotordynamic data for large, 18 inch (457 mm) generator bearings. The tests showed comparable vibration response for conventional and direct lube designs. Edney, et al. (1996), provide similar information for a small, high-speed, multistage steam turbine with 4 inch (102 mm) diameter journal bearings. Peculiar, low-frequency, radial vibrations were observed during these steam turbine tests, but levels were low and acceptable. However, similar vibrations in a high-speed compressor during acceptance tests in 1999 did encroach upon acceptable API limits, and significant time and resources were expended to address this issue (DeCamillo, 2006). Personal experience and discussions among original equipment manufacturers (OEMs) and users over the past few years indicate
SSV HASH, DEFINITION

Figure 1 is an example radial vibration spectrum from a high-speed compressor test using direct lube tilt-pad journal bearings. The main subject of this paper is the low-frequency shaft vibrations indicated in the figure. Unlike discrete subsynchronous spikes that often indicate a serious problem, the vibrations are indiscrete and of low-frequency and amplitude. The plot is paused to capture the random, broadband frequencies. In order to distinguish the subsynchronous vibration (SSV) characteristics under consideration, the term SSV hash is defined for use in this paper:
SSV Hash: A vibration signature characterized by low-frequency,

low amplitude, broadband subsynchronous vibrations that fluctuate randomly.
Figure 1. Example Radial, Low-Frequency, Broadband, Vibration. Experience with turbomachinery bearing sizes 3.88 to 8.00 inches (100 to 200 mm) in diameter has observed SSV hash frequencies typically ranging up to 30 Hz with amplitudes on the order of 0.1 to 0.2 mils (.0025 to .0050 mm) peak-to-peak.
INITIAL INVESTIGATIONS
Delayed acceptance of a high-speed compressor due to SSV hash in 1999 prompted a research project performed on a high-speed test rig described in detail in a separate reference (Wilkes, et al., 2000). Pertinent information is provided here for convenience. The rig has a test shaft driven by a variable speed gas turbine through a flexible coupling (Figure 2). The test shaft is approximately 5 feet (1.5 m) long and 5 inches (127 mm) in diameter, supported at either end by a pivoted shoe journal bearing. Two orthogonal proximity probes are mounted inboard of each journal bearing, 45 degrees off top-dead-center, to record radial shaft vibration. A spectrum analyzer was used to acquire fast Fourier transform (FFT) vibration
13
signatures, presented in paused plots in Figures 3 through 5. The y-axis represents peak-to-peak amplitude in mils.
Figure 2. High-Speed Test Rig Journal Bearings and Shaft Schematic.
Figure 3. Evacuated Discharge Configuration, 3.0 gpm (11.4 l/min).
Fortunately, it was possible to duplicate the low-frequency vibration signature in the test rig, which allowed a parametric study of many journal bearing designs and configurations over the course of this initial investigation. The data presented in Figures 3 through 5 are for direct lube, five-pad, leading-edge-groove journal bearing tests at a shaft speed of 10,000 rpm and a low, 20 psi (0.14 MPa) projected load. The bearings have a nominal diameter of 5 inches (127 mm) and an axial length of 2.25 inches (57 mm). The pads are steel backed with a babbitt surface and have a 60 degree angle and a 60 percent offset pivot. The assembled bearing diametric clearance is 0.009 inch (0.23 mm) and the nominal preload is 0.15. Tests were run with ISO VG 32 turbine oil supplied at 120F (49C). The SSV hash indicated in Figure 3 was recorded for an oil flow of 3.0 gpm (11.4 l/min) for the direct lube bearing in its as-designed, evacuated oil discharge configuration. Tests found that increasing the oil flow tended to reduce the amplitudes but did not entirely eliminate the SSV hash signature. Elimination required the installation of floating oil seals as well as an increase in oil flow to 6.0 gpm (22.8 l/min), the results of which are shown in Figure 4. This solution unfortunately required higher oil flow and operated with higher power loss and pad temperatures than the original design. Methods were therefore pursued to address these performance issues, and results began to suggest that the low frequencies may be due to air entering the oil film. Based on this hypothesis, a design was conceived that might eliminate SSV hash while maintaining some direct lube benefits. The pads were modified with narrow circumferential SSV grooves, cut in the babbitt near the edges of the pads (Figure 5), to capture and redirect side leakage toward the leading edge of the next pad (Wilkes and DeCamillo, 2002). In this way, additional oil is made available to the oil films without increasing the bulk oil flow to the bearing. The SSV groove pads were installed and tested in the original evacuated condition (oil seals removed) with results shown in Figure 5 for 3.0 gpm (11.4 l/min), and tested as low as 2.0 gpm (7.6 l/min) with negligible SSV hash indications. Of several methods pursued during the course of this initial investigation, the grooves were the only solution successful in eliminating SSV hash in an evacuated configuration. The low oil flow and power loss of the original design were maintained, with a slight penalty in pad temperature due to the introduction of warm, side leakage oil back into the oil film. Another observation of this initial investigation was the tendency for an increase in synchronous amplitudes when SSV hash was eliminated, noticeable by comparing both the flooded solution (Figure 4) and the evacuated groove solution (Figure 5) with the original open discharge configuration data of Figure 3.
Figure 4. Flooded Discharge Configuration, 6.0 gpm (22.8 l/min).
SUBSEQUENT TESTS
Although solutions were obtained in initial investigations, operating conditions were limited in load and speed. A subsequent series of tests was initiated in 2005 to further investigate the low-frequency vibrations. The same test rig was modified to incorporate a radial load system in the bearing housing at the free end of the shaft, shown schematically in Figure 6. The system incorporates a hydraulic cylinder that loads through a dowel to a single journal bearing loader shoe on top of the shaft. This pushes down on the shaft and loads the test bearing. The applied load is measured by a load cell below the hydraulic cylinder. A new data acquisition system was used to acquire the data, and additional proximity probes were installed to monitor the vibration of an upper and a lower journal bearing pad. The pad proximity probes were mounted in the bearing casing and targeted the trailing edge of the back of the pads at the locations depicted schematically in Figure 7.
Figure 5. Evacuated Discharge Configuration, 3.0 gpm (11.4 l/min), SSV Grooves.
14
Figure 6. Test Rig Radial Load System Schematic, Free End of Shaft. Figure 10. Direct Lube Leading-Edge-Groove, Evacuated Configuration.
Figure 7. Bearing Orientation and Pad Numbering. The test bearings consist of a conventional center pivot bearing with labyrinth end seals (Figure 8) and evacuated direct lube center and offset pivot bearings. Three direct lube methods were tested: spray nozzles (Figure 9), leading-edge-grooves (Figure 10), and between-pad-grooves (Figure 11). Each bearing was tested for load-on-pad and load-between-pad orientation over a range of loads, oil flows, and speeds. Figure 11. Direct Lube Between-Pad-Groove, Evacuated Configuration. Bearing geometry, test oil viscosity, and inlet temperature are the same as in earlier investigations except that the assembled diametric clearance is 0.0072 inch (0.18 mm) and the nominal preload is 0.25. Loads indicated in the following sections and figures refer to the applied radial load. For data labeled no-load, the journal bearings are supporting the weight of the shaft, which gives a 20 psi (0.14 MPa) projected unit bearing load. Each test bearing is installed around the free end of the shaft without disturbing the coupling or the coupling end journal bearing. The same coupling end journal bearing is used in all tests. Vibration data were recorded while varying speed to 14,000 rpm, load to 400 psi (2.8 MPa), and oil flow to 30 gpm (114 l/min), holding two of the parameters constant while the other was varied. The speed ramp data are used in this paper, presented in color-coded FFT waterfall diagrams with a vertical scale denoting peak-to-peak amplitude in mils. With Figure 12 as an example, each waterfall diagram contains data for three separate run-up/run-down speed ramps from 5000 to 14,000 to 5000 rpm (83 to 233 to 83 Hz), one for each of three oil flows. The back plane of the diagram contains a projection of all data. Vibrations below 1 Hz are cut off to delete spurious 0 Hz indications from obscuring the low frequencies under investigation and the data are uncompensated, the main focus being the low-frequency vibrations. As a guide, SSV hash amplitudes above 0.1 mils (0.0025 mm) peak-to-peak are levels that have caused concern in acceptance tests.
Figure 8. Conventional Bearing with Labyrinth End Seals.
Figure 9. Direct Lube Spray Nozzles, Evacuated Configuration.
15
Figure 12. Conventional Center Pivot, LBP , No-Load. Shaft SSV Hash Trends The volume of data for the different bearing designs allowed for an assessment of trends in shaft SSV hash. This section describes those that were fairly common for all test bearings. Conventional bearing speed ramp data in Figures 12 through 15 are used for reference.
Figure 15. Conventional Center Pivot, LOP , 400 psi (2.8 MPa). Comparing data among the figures, it can be noticed that most SSV hash indications occur at low flow and low load. The word indications is carefully used in this generalization. Although there were more indications at low flow and low load for most bearing designs, SSV hash amplitudes were sometimes higher at other operating conditions, as evidenced in results later in this paper. Other common trends derived from the study are that SSV hash levels for no-load data were similar for load-between-pad (LBP) and load-on-pad (LOP) orientations (e.g., Figures 12 and 13). A curious result was that while SSV hash decreased with applied load for load-between-pad orientation (e.g., Figure 12 versus 14) there was only a small change with load for load-on-pad orientation (e.g., Figure 13 versus 15). Consequently, load-on-pad orientation had higher SSV hash levels than load-between-pad orientation at higher loads for all test bearings. It is worth noting that initial investigations and field reports were for high-speed applications with relatively light rotors, and so SSV hash was formerly associated with high-speed and low-load. The test data in these subsequent series of tests indicate that this is not exactly true. Although indications were more pronounced at low loads, tests indicate there can be undesirable levels of SSV hash in the case of load-on-pad orientation at higher loads, for example Figure 15 at 400 psi (2.8 MPa). There were also many indications at lower speeds in the test data, again using Figure 15 as an example. Conventional Versus Direct Lubrication Figure 16 displays no-load data for the conventional center pivot test bearing with load-on-pad orientation, the same used in Figure 13, but zoomed in on the lower frequencies and amplitudes under investigation. Figure 17 contains comparable, center pivot, between-pad-groove direct lube data. A condition mentioned earlier is noticeable in the direct lube data, i.e., there are more SSV hash indications at low flow but with higher amplitudes at an intermediate condition, 8 gpm (30 l/min) and 11,000 rpm in this particular test. The figure also supports observations from field experience and initial investigations where SSV hash decreases, but is not necessarily eliminated at higher flows. Comparisons between conventional and direct lube results (Figures 16 and 17) are fairly straightforward. Direct lube data have noticeable SSV hash indications at 8 gpm (30 l/min) whereas conventional bearing levels are negligible. Conversely, very high subsynchronous vibrations are noticeable in conventional bearing data at 4 gpm (15 l/min) where the direct lube design has significantly lower SSV hash indications. These data indicate that flooding and oil flow are not the only key parameters influencing SSV hash. Conventional bearing tests with labyrinth seals (Figure 16), and initial tests with oil seal rings
Figure 13. Conventional Center Pivot, LOP , No-Load.
Figure 14. Conventional Center Pivot, LBP , 400 psi (2.8 MPa).
16
(Figure 4), both required some increase in oil flow to eliminate the signature, whereas higher oil flow is not as effective in evacuated designs (e.g., Figure 17). This suggests cavity pressure may be another influential parameter that, unfortunately, was not measured. Calculated pressure drops across oil discharge restrictions indicate that cavity pressures between 0.5 and 2.0 psig (0.04 and 0.14 bar) were present when SSV hash was noticeably reduced. Also fitting this scenario are the lower SSV hash indications in direct lube designs compared to the conventional bearing data at low flow (Figure 17 versus 16), which may be attributable to the more direct application of oil to the oil films.
Figure 18. Direct Lube Spray, Offset Pivot, LOP , No-Load.
Figure 16. Conventional Center Pivot, LOP , No-Load.
Figure 19. Direct Lube Spray, Center Pivot, LOP , No-Load.
Figure 17. Direct Lube Between-Pad-Groove, Center Pivot, LOP , No-Load. Center Versus Offset Pivot Center versus offset pivot tests were conducted for several direct lube bearing configurations. Comparisons have been difficult to quantify. For example, spray nozzle offset pivot data (Figure 18) have high amplitudes and a broad band of SSV hash midway through the 4 gpm (15 l/min) data speed ramps at approximately 12,000 rpm, indicated in the figure. At the same conditions, spray nozzle center pivot data (Figure 19) have less broadband indications, and amplitudes are actually highest at approximately 6000 rpm. In 8 and 12 gpm data (30 and 45 l/min), the offset pivot has slightly less SSV hash. Similar comparisons for between-pad-groove tests (Figure 17 versus 20) show lower SSV hash levels in the offset pivot data at 4 gpm (15 l/min), and only subtle differences at the higher flows.
Figure 20. Direct Lube Between-Pad-Groove, Offset Pivot, LOP, No-Load. The results were unexpected in that offset pivot designs require more oil flow to the films, and it was suspected that this would be clearly reflected in SSV hash levels. The data indicate there are factors other than bulk oil flow that need to be taken into account regarding SSV hash and pivot offset.
17
Variations Among Direct Lube Designs Three offset pivot, direct lube bearings were tested to compare shaft SSV hash characteristics among different methods of direct lubrication. All three have an evacuated discharge configuration. The methods include spray nozzles, leading-edge-grooves, and between-pad-grooves. Photographs of the direct lube features are provided in Figures 9 through 11, and corresponding vibration data are contained in Figures 19 through 21.
This lack of correlation prompted another series of tests in 2006 with all pads monitored by proximity probes, installed and positioned as described earlier, and made possible by the acquisition of new high-speed vibration equipment. Figure A-1 (refer to APPENDIX A) organizes shaft and pad vibration data for a leading-edge-groove bearing test at 200 psi (1.38 MPa), for a load-on-pad configuration as an example. The waterfall diagrams are arranged so that data for the orthogonal shaft probes and all five pads can be viewed in relation to one another. The same scales are used for all waterfall diagrams in the figure. It is important to note that the shaft was removed and refurbished between 2005 and 2006 tests and a different coupling end journal bearing installed, so data are not comparable to information from earlier tests. Referring to Figure A-1, it can be noticed that the subsynchronous vibrations in unloaded pad 2 at 4 gpm (15 l/min) do not correlate with either of the orthogonal shaft probes (labeled +45 degrees and 45 degrees in the figure). A unique observation is that the side pads 3 and 5, which were not monitored in previous tests, have the strongest subsynchronous vibrations and show excellent correlation with the shaft SSV hash indications. Pad/Shaft Correlation General Trends The observations from Figure A-1 relate to a specific bearing and set of operating conditions. Comparisons of data for all five pads, over a broad range of operating conditions for the different bearing designs allowed for a more precise assessment of pad/shaft interaction. The following observations and trends were determined to be fairly common for all test bearings. All shaft SSV hash indications observed in these tests were confirmed to correlate with vibrations from at least one of the five pads. The converse is not true. There were many instances where subsynchronous vibrations from individual pads did not appear in any shaft data. An important observation is that the shaft SSV hash indications most often correlated with one or more of the side pads, depending on bearing type and orientation. The orientation and pad number schematic in Figure 7 is helpful in explaining the following observations and trends. In the case of load-between-pad orientation under some light downward loading, side pad 1, side pad 3, or both tended to correlate with the shaft SSV hash indications. Upper pad 2 had more and higher subsynchronous vibration than the side pads in many cases, but rarely correlated with the shaft indications. Noted earlier, shaft SSV hash decreased with applied load in the case of load-between-pad orientation, which is reasonable considering the high horizontal stiffness of this orientation with the shaft supported between the two bottom pads. At the same time, pads 1, 2, and 3 become less strongly coupled to the shaft and may still experience subsynchronous vibrations, but with hydrodynamic forces too weak to affect the shaft. For a load-on-pad orientation under some light downward loading, the side pads (1, 2, 3, and 5), individually or in combinations, tended to correlate with the shaft SSV hash indications. The correlation varied with bearing type and operating conditions, and is also likely influenced by any slight skew in applied load vector and differences in manufacturing heights of the individual pads for this particular orientation. Noted earlier, shaft SSV hash did not decrease as much with applied load for load-on-pad orientation, which is reasonable considering the weak horizontal stiffness for this orientation. Pad 3 and pad 5 tended to correlate more with the shaft SSV hash indications as load was applied. At the same time, pads 1 and 2 become less strongly coupled to the shaft and may still experience subsynchronous indications, but with hydrodynamic forces too weak to affect the shaft. SSV Grooves, Recent Tests The SSV groove modification (Figure 5), developed during initial investigations in 1999, has since been successfully applied in
Figure 21. Direct Lube Leading-Edge-Groove, Offset Pivot, LOP , No-Load. In general, all three direct lube methods display noticeable SSV hash indications. Variations in amplitudes and frequencies make it difficult to generalize the results. The spray nozzles produced the highest levels of SSV hash. The leading-edge-groove operated with lower SSV hash amplitudes but at higher frequencies, and between-pad-groove SSV hash levels were lower for all conditionsin comparison with spray nozzle data. Results comparing between-padgroove and leading-edge-groove data vary with operating conditions. SSV hash levels at 4 gpm (15 l/min) are negligible and lower than the leading-edge-groove. At higher flows, between-pad-groove SSV hash amplitudes are higher but occur at a lower frequency. Distinct differences in shaft SSV hash signatures are obvious among the designs. Some speed dependence is noticeable in spray nozzle data (Figure 18), and more so in leading-edge-groove data (Figure 21), which appears to be suppressed at higher speeds. Between-pad-groove characteristics (Figure 20) do not show variations with speed, and seem sensitive to a particular speed band depending on flow. There are also many conditions throughout Figures 18 through 21 where the evacuated direct lube bearings operate with negligible SSV hash indications. While the reasons for the variations are still under investigation, the fact that there are differences provides valuable information. Since the pad geometry is the same for the different bearings, the direct lube feature itself is influencing the shaft SSV hash and, because these are evacuated designs, the influence is most likely occurring at the entrance or leading edge of the oil film. Pad/Shaft Correlation Investigation A review of 2005 vibration data from the upper and lower pad proximity probes (depicted in Figure 7), noted more subsynchronous indications in the upper pad, with the highest amplitudes occurring at high-load and low-flow conditions in many cases. This is reasonable considering that the upper pad clearance increases as the shaft is pushed down under load, away from the upper pad. The upper pad oil film consequently generates lower hydrodynamic forces and requires more oil flow to fill the gap. Unfortunately, an unexpected result of the investigation was that there were only a few random correlations between upper pad and shaft SSV hash for the broad range of operating conditions and bearing configurations tested.
18
many compressor, turbine, and gearbox applications. At the time, development was based on hypotheses and shaft vibration data. Upgrades in the test rig capability and in high-speed data acquisition equipment has more recently allowed an investigation of the grooves influence on pad vibration. SSV grooves were machined in the leading-edge-groove pads used for results presented in Figure A-1 and were tested over the full range of operating conditions for load-on-pad and load-between pad orientation. Test results for the same conditions and orientation as Figure A-1 are presented in Figure A-2 (refer to APPENDIX A) as an example comparison. The tests found that subsynchronous levels in all pads were significantly reduced or eliminated over the full range of test operating conditions and configurations, and confirmed the effectiveness of the method in eliminating shaft SSV hash in evacuated, direct lube bearing designs.
Pad Responsiveness Since test results confirm that the individual pad motions are involved in this SSV hash phenomenon, the tilting-pad bearings full coefficients predicted by the modeling algorithm are of primary interest. For a tilting-pad bearing with Npad pads, the full coefficients consist of 5Npad+4 stiffness and 5Npad+4 damping coefficients. Described in detail by Shapiro and Colsher (1977) and Parsell, et al. (1983), these full coefficients relate pad and shaft motions to forces and moments on the shaft and pads. For rotordynamic purposes, the magnitude of each individual coefficient is overlooked, focusing instead on their combined or reduced coefficients (four stiffnesses and four dampings). Closer scrutiny and understanding of these full coefficients are required when pad dynamics become the focus. Table 1 presents the stiffness full coefficients for a 60 percent offset, LOP, five-pad bearing with 12 gpm supply flowrate where: Fx and Fy x and y M1 M5 1 5 = = = = Horizontal and vertical forces on the shaft Horizontal and vertical shaft displacements Moments on pads 1 through 5 Tilt angle changes on pads 1 through 5
THEORETICAL INVESTIGATIONS
To help shed some light on these SSV hash measurements, the bearing geometry and operating conditions corresponding with the data in Figure A-1 were modeled using an algorithm developed by He (2003). This algorithm includes the effects of supply flowrate in its determination of a bearings steady-state (eccentricity, temperatures, power loss, etc.) and dynamic (stiffness and damping) performance based on the theories and experiments put forth by Heshmat (1991). Since the measurements presented in these SSV hash tests demonstrate the importance of supply flow on SSV , any theoretical investigation must account for supply flowrate effects. Supply Flowrate Effects The model applies to any bearing configuration where the supply flowrate is not sufficient to ensure a full film across all the pads surfaces. In this case, the model will predict a partial starvation at the inlet region of some of the pads. Unlike the full film situation where a continuous lubricant film starts at the pad leading edge, in a partially starved situation, the continuous film is formed downstream at some point where the clearance is sufficiently reduced. This physical phenomenon was observed and studied by Heshmat (1991) on a two axial groove, sleeve bearing. Figure 22 illustrates predicted starvation effects for pad number 3, corresponding with the test data pad numbers in Figure A-1. With a supply flowrate of 8 gpm to the bearing, the film is able to generate pressure across the pads entire arc length. Reducing the bearing flow to 4 gpm lowers overall pressure distribution, but the pad still produces pressure across almost all of its arc length. The pad is clearly partially starved at 3 gpm where no film pressure is produced at the inlet film region. Decreasing the flow to 2 gpm expands the starved inlet region. At this flow level, pressure levels are very small and the pivot is nearly centered within the positive pressure region. Partial starvation has effectively reduced the pads pivot offset, which may explain unexpected results in center versus offset pivot tests.
Table 1. Stiffness Full Coefficients for the 60 Percent Offset LOP Bearing, 11,475 rpm, 200 psi.
Each value in Table 1 represents a stiffness, either Fx or M divided by x, y, or depending on the position in the table. Examining a few of these Table 1 coefficients will help explain the physical meaning behind them. Tilt angle changes of the fourth pad (4) produce the largest vertical force on the shaft (Fy), since it has the highest value of Ky (3.261e6 lbf/rad). This is due to the fourth pad being on the bottom, supporting the majority of the load. Conversely, tilt angle changes on the second pad, located in the upper half of the bearing, produce the smallest vertical force on the shaft (Ky = 0.499e6 lbf/rad). The fourth pad is also the most difficult to tilt with the highest tilting stiffness K value (1.614e6 lbf-in/rad), while the second pad is the easiest to tilt (K = 0.245e6 lbf-in/rad). Tilting motions of the other four pads do not directly cause any moments on the loaded pad. The pads can only effectively communicate to each other through the shaft. Some of the full coefficients along with the pads polar inertia (Ip) form the equation of motion for pad tilting. A simplified version of this equation, using a pads tilting stiffness K and damping C, is given by:
where Mext is an external moment being applied. This equation can be used to examine the frequency response characteristics of each pads tilting motion. Each pads frequency response function (FRF) H() is given by:
Figure 22. Partial Starvation Effects on Pad Film Pressure.
An FRFs magnitude indicates how responsive a pad is to external moments trying to excite it. Examining the predicted FRFs of each pad as a function of supply flowrate provides for a direct comparison with the pad waterfall diagrams of the test data.
19
Figure 23 presents the predicted H() of pad 3 in Figure A-1 as an example, focusing on the low frequency region below 60 Hz. Regardless of flow, the frequency response remains relatively flat as a function of excitation frequency with no peaks present. This is because the pads tilting motions remain overdamped for all the flowrates examined. An overdamped system (critical damping greater than 100 percent) has no damped natural frequency, resulting in no peaks in its response.
Figure 24. Tilting Stiffness of Individual Pads Versus Supply Flow. When a pad becomes unloaded, it applies no pressure force on the shaft and its full coefficient stiffness and damping terms go to zero. The pad then moves like a rigid body and its tilting FRF is dictated by the polar inertia. The upper pads (1 and 2) become unloaded first as flow is reduced (Figure 24). This is because their larger film thicknesses require the most flow to maintain a full film along their entire surface. With smaller film thicknesses, the third and fifth pads remain loaded until below 2 gpm. These two pads tilting stiffness values are low in magnitude at reduced flowrates, resulting in the increased responsiveness observed in Figure 23.
Figure 23. H() with Decreasing Supply Flow. It is unclear as to the exact cause of the 8 Hz peak in Figure A-1s waterfall diagram for pad 3. One explanation is that there may be a predominant excitation at this frequency. Heshmat (1991) observed a pulse excitation at similar low frequency levels in his starvation experiments on a sleeve bearing. Another possibility may be that the pad is actually underdamped, which the model has not been able to predict. In this case, a broadband excitation would result in a definitive peak at the pads damped natural frequency. Nothing indicates that the peak is associated with an unstable, self-excited phenomenon, since the pads tilting damping C remains greater than or equal to zero. All indications are that the observed vibrations are forced in nature. Both the measurements in Figure A-1 and the predicted FRF in Figure 23 show that the third pads overall responsiveness does dramatically increase as supply flow is reduced. According to the predictions in Figure 23, this pad responds very little to excitations until the bearing flow is reduced below approximately 4 gpm. This increased sensitivity correlates well with the predicted pressure profiles in Figure 22. Below 4 gpm, the predictions in Figure 22 show that the pad becomes partially starved at the leading edge with low overall film pressures. These lower film pressures allow the pad to respond to external moment excitations more easily. According to the H() equation, the low frequency responsiveness of an individual pad is largely dictated by its tilting stiffness, K. Figure 24 presents the calculated tilting stiffness of all five pads as supply flow is varied. With the exception of the fourth pad, reducing supply flow causes a decrease in all the pads tilting stiffnesses. Below 5 gpm, the fourth pads stiffness begins to increase as it supports more of the entire bearing load. The strength of this tilting stiffness means the fourth pad does not easily respond to excitations, which correlates well with its relatively low vibrations in Figure A-1.
CONCLUSIONS
A series of tests and analyses were performed to investigate a peculiar, low-frequency, low amplitude, broadband subsynchronous vibration, termed SSV hash, that has been witnessed in different types of turbomachinery using tilting pad journal bearings. Based on a study of test results for conventional and direct lube designs over a broad range of speeds, loads, and oil flows, the following shaft SSV hash trends were found to be fairly common for all test bearings:
There were more SSV hash indications at low flow and low
and load-on-pad bearing orientations.
load, although amplitudes were sometimes higher at other operating conditions.
SSV hash levels at light loads were similar for load-between-pad Shaft SSV hash decreased with applied load for load-between-pad
orientation, but there was only a small change with load for load-on-pad orientation. levels than load-between-pad orientation at higher loads.
Consequently, load-on-pad operation produced higher SSV hash

The following general trends for pad/shaft vibration correlation were found fairly common for all test bearings based on tests of bearing designs with vibration measurements of all five pads:
All shaft SSV hash indications were confirmed to correlate with

vibrations of at least one of the five pads.
The converse is not true. There were subsynchronous vibrations in The

side pads most often correlated with shaft SSV hash indications. The top upper pad for load-between-pad orientation had the highest subsychronous indications in many cases, but rarely correlated with any shaft data.
individual pads that did not appear in measured shaft vibration data.
20
The following specific observations were noted in test data comparisons of bearing types and geometries:
directly from the test results and analyses. As such, the discussions are subject to debate, which is welcome. Possible Source of Excitation Test results and analyses indicate that the shaft SSV hash indications are caused by pad vibration with response characteristics indicative of a forced vibration. Since there were no means in the SSV hash test apparatus to visualize the flow in the bearing, possible sources of excitation can only be surmised from the test results presented in this paper. Instead, attention is directed to Heshmat (1991), who did observe a periodic phenomenon in his visualization experiments of oil streamlets in an axial groove journal bearing. The phenomenon is reported as a pulse that periodically transformed the starved region of the leading edge film into a pattern of oil streamlets. This pulse occurred every 0.5 to 1.0 seconds and its frequency varied with the degree of starvation. This phenomenon is consistent with many of the observations from these SSV hash investigations. Test data and analyses both show an increased sensitivity to SSV hash at reduced flowrates. A periodic pulse excitation can explain dominant SSV hash amplitudes at intermediate operating conditions, i.e., peaks noticeable in Figures 17 through 22, as well as difference in SSV hash signatures among test bearings, which may be attributable to the influence of the different supply methods on the excitation. Pad Vibration and Damage Pad flutter, babbitt fatigue, and pivot fretting are often brought up as topics of concern regarding SSV hash. The babbitt damage studied by Adams and Payandeh (1982) was a major concern in large, conventional, flooded bearings, attributed to self-excited, subsynchronous vibrations of unloaded pads with frequency ratios approximately 0.50 that of running speed. The term pad flutter is often used to describe this motion, and spragging is often used to describe more violent, full clearance vibrations with forces sufficient to fatigue the babbitt and fret the pivots. The pad vibrations associated with SSV hash do not conform to these characteristics. Subsynchronous amplitudes are at least an order of magnitude less than the bearing clearance, and frequency ratios are more on the order of 0.10 on average. Babbitt fatigue is not considered a concern in regard to the SSV hash characteristics described in this paper. Fretting is a more difficult phenomenon to assess analytically. Vibrations certainly contribute to fretting, but there are many sources of vibration in turbomachinery. The only information that can be offered from the SSV hash test series in this respect is that there were no indications of fretting, or babbitt fatigue, in individual test pads after 600 hours of operation. Bearing Clearance, Preload, Pivot Offset It would seem that preload, via reduced bearing clearance, would reduce shaft SSV hash and that increasing the pivot offset should make it worse. It was therefore unexpected when center versus offset pivot tests showed lower levels for offset pivot operation. Hindsight from the test results and analyses indicate other factors are involved, e.g., proximity of the individual pads to the shaft, the magnitude of their hydrodynamic forces, sensitivity to excitation, etc., which make it difficult to generalize cause and effect. Although preload and clearance tests were not complete at the time of this paper, scenarios where preload might invoke pad vibration and shaft SSV hash can be envisioned when considering the other influences. With multiple factors and complex pad/shaft interactions, cause and effect scenarios for clearance, preload, offset, and other geometry should be performed by analysis rather than generalizations. The analyses presented in the theory section of the paper are useful tools for assessing new designs, as well as evaluating changes to existing designs.
It
is incorrect to associate SSV hash only with high-speed, low-load applications. Indications were noted over a broad range of operating conditions.
Shaft SSV hash has been observed in conventional bearing designs at low oil flow. The data indicate that merely flooding the cavity is insufficient to suppress shaft SSV hash. At these conditions, direct lube designs operate with lower levels of shaft SSV hash, attributed to the more direct application of oil to the film. At intermediate test oil flows, shaft SSV hash is more pronounced
in evacuated, direct lube bearings and can be reduced, but not necessarily eliminated, by increasing oil flow. Amplitudes appear sensitive to certain operating conditions, around which are conditions where the evacuated direct lube bearings operated with negligible SSV hash indications.
There were mixed results in SSV hash comparisons of center Comparisons among three direct lube methods noted distinct
versus offset pivot direct lube test data. The offset pivot pads in most cases operated with the same to slightly less SSV hash indications than the center pivot pads. differences in shaft SSV hash signatures, the results suggesting that the differences originate at the entrance or leading edge of the oil film. Overall, the various tests and comparisons indicate there are factors other than flooding and bulk oil flow that contribute to SSV hash behavior. There were two solutions determined from test data that eliminated the SSV hash signature:
The first required flooding the bearing cavity, using labyrinth or
floating seals and increased oil flow. The solution was effective for direct lube designs as well, but resulted in higher power loss, flow requirements, and pad temperatures comparable to a conventional, flooded design.
The
second was a modification consisting of patented SSV grooves cut in the babbitt near the edges of the pad, to capture and redirect side leakage toward the leading edge of the next pad. This solution was successful in eliminating SSV hash in an evacuated configuration. Low oil flow and power loss were maintained, with a slight penalty in pad temperature due to the introduction of warm, side leakage oil back into the oil film.
In both cases, the elimination of SSV hash increased synchronous

amplitudes. The following conclusions are derived from the theoretical investigation of SSV hash:
A tilting pad bearings full coefficients can be used to assess the

dynamics of individual pads.
Using the full coefficients of one of the test bearings, theoretical

investigations suggest that the observed SSV hash is likely a forced vibration phenomenon.
Theoretical predictions indicate that a pads responsiveness at

low frequencies increases when there is insufficient oil to provide for a full film.
At lower supply flowrates, predicted partial starvation at the

DISCUSSIONS
leading edge progresses, which reduces the pads tilting stiffness, making it more responsive to excitation.
Many questions arose over the course of the tests and analyses and during review of the initial drafts of this paper. The following discussions comment on topics that are not necessarily derived
21
SSV Hash: Issue, Nonissue, Allowable Levels The many types of turbomachinery and the wide array of sizes and operating conditions suggest caution against generalizations. There are applications running with SSV hash with no reported problems, and DeCamillo and Clayton (1997) and Edney, et al. (1996), provide rotordynamic data showing some level of flow reduction is possible without affecting rotor response. On the other hand, Edney, et al. (1996), do report problems when flow is reduced too much, which is logical and also supported by the theoretical analyses. Personal experience with SSV hash for turbomachinery bearing sizes 3.88 to 8.00 inches (100 to 200 mm) in diameter generally falls within API guidelines. In comparisons with API 617 (2002) nonsynchronous limits, for example, levels are typically less then 0.2 mils (.0050 mm) peak-to-peak. Actually, SSV hash more often is a consideration in overall vibration limitations, as nonsynchronous frequencies are typically below 0.25 times the maximum continuous speed and are indiscrete. Noted earlier, elimination of SSV hash tended to increase synchronous amplitudes in tests. This is brought up neither as an issue or nonissue, but to provide information that
overall vibration levels may not decrease as much as anticipated when SSV hash is eliminated. Issues more often arise when other specifications further limit allowable SSV indications or, as stated earlier, because of uncertainty regarding the cause and nature of the vibration. In this case, solutions and test results in this paper can be used to address the situation. A more challenging issue is complying with multiple spec limitations in more severe applications. It seems that many of the parameters that reduce SSV hash tend to increase power loss, flow requirement, and pad temperature, or produce undesirable rotor response. Flooding and pressurized bearing cavities, rotordynamic preference for center pivot load-on-pad configurations, and tighter clearances are some examples. There is certainly the need for more research and development, experimental and theoretical, on the subject of SSV hash including its effects on dynamic coefficients, rotor response, and stability. In the meantime, the authors hope that the tests and analytical investigations presented in this paper will provide a useful reference for topics related to SSV hash.
APPENDIX A
Figure A-1. Direct Lube Leading-Edge-Groove, LOP , 200 psi (1.38 MPa), No SSV Grooves.
22
Figure A-2. Direct Lube Leading-Edge-Groove, LOP , 200 psi (1.38 MPa), with SSV Grooves.
REFERENCES
Adams, M. L. and Payandeh, S., 1983, Self-Excited Vibration of Statically Unloaded Pads in Tilting-Pad Journal Bearings, ASME Journal of Lubrication Technology, 105, pp. 377-384. API 617, 2002, Axial and Centrifugal Compressors and Expander-Compressors for Petroleum, Chemical and Gas Industry, Seventh Edition, American Petroleum Institute, Washington, D.C. Booser, E. R., 1990, Parasitic Power Losses in Turbine Bearings, STLE Tribology Transactions, 33, pp. 157-162. Brockwell, K., Dmochowski, W., and DeCamillo, S., 1992, Performance Evaluation of the LEG Tilting Pad Journal Bearing, IMechE Seminar Plain BearingsPlain Bearings Energy Efficiency and Design, MEP, London, United Kingdom, pp. 51-58. Cloud, C. H., 2007, Stability of Rotors Supported on Tilting Pad Journal Bearings, Ph.D. Dissertation, University of Virginia, Charlottesville, Virginia. DeCamillo, S. and Clayton, P. J., 1997, Performance Tests of an 18-Inch Diameter, Leading Edge Groove Pivoted Shoe Journal Bearing, Proceedings of the 2nd International Conference on Hydrodynamic BearingRotor System Dynamics, Xian, China, pp. 409-413. DeCamillo, S., 2006, Current Issues Regarding Unusual Conditions in High-Speed Turbomachinery, Keynote
Presentation, 5th EDF & LMS Poitiers Workshop Bearing Behavior Under Unusual Operating Conditions Proceedings, pp A.1-A.10. Edney, S. L., Waite, J. K., and DeCamillo, S. M., 1996, Profiled Leading Edge Groove Tilting Pad Journal Bearing for Light Load Operation, Proceedings of the Twenty-Fifth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 1-16. Fillon, M., Bligoud, J. C., and Frene, J., 1993, Influence of the Lubricant Feeding Method on the Thermohydrodynamic Characteristics of Tilting Pad Journal Bearings, Proceedings of the 6th International Congress on Tribology, Budapest, Hungary, 4, pp. 7-10. Harangozo, A. V ., Stolarski, T. A., and Gozdawa, R. J., 1991, The Effect of Different Lubrication Methods on the Performance of a Tilting Pad Journal Bearing, STLE Tribology Transactions, 34, pp. 529-536. He, M., 2003, Thermoelastohydrodynamic Analysis of Fluid Film Journal Bearings, Ph.D. Dissertation, University of Virginia, Charlottesville, Virginia. He, M., Cloud, C. H., and Byrne, J. M., 2005, Fundamentals of Fluid Film Journal Bearing Operation and Modeling, Proceedings of the Thirty-Fourth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 155-175.
23
Heshmat, H., 1991, The Mechanism of Cavitation in Hydrodynamic Lubrication, STLE Tribology Transactions, 34, (2), pp. 177-186. Kocur, J. A., Nicholas, J. C., and Lee, C. C., 2007, Surveying Tilting Pad Journal Bearing and Gas Labyrinth Seal Coefficients and Their Effect on Rotor Stability, Proceedings of the Thirty-Sixth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 1-10. Parsell, J. K., Allaire, P. E., and Barrett, L. E., 1983, Frequency Effects in Tilting-Pad Journal Bearing Dynamic Coefficients, ASLE Transactions, 26, (2), pp. 222-227. Shapiro, W. and Colsher, R., 1977, Dynamic Characteristics of Fluid-Film Bearings, Proceedings of the Sixth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 39-53. Tanaka, M., 1991, Thermohydrodynamic Performance of a Tilting Pad Journal Bearing with Spot Lubrication, ASME Journal of Tribology, 113, pp. 615-619.
Wilkes, J. J., DeCamillo, S. M., Kuzdzal, M. J., and Mordell, J. D., 2000, Evaluation of a High Speed, Light Load Phenomenon in Tilting-Pad Thrust Bearings, Proceedings of the Twenty-Ninth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 177-185. Wilkes, J. and DeCamillo, S., 2002, Journal Bearing, United States Patent No. 6,361,215 B1, Mar. 26, 2002.
ACKNOWLEDGEMENT
The authors would like to thank colleagues at Kingsbury, Inc., for their special efforts and help in acquiring and preparing data and information for this technical paper. The authors would also like to especially acknowledge John Kocur of ExxonMobil, Brian Pettinato of Elliott, and Thomas Soulas of Dresser-Rand, among others, for their help and expertise in discussions of the subject matter.
DESIGN AND MITIGATION TECHNIQUES FOR APPLICATIONS WITH POSSIBLE LIQUID CONTAMINATION OF THE SEALING GAS DRY GAS SEAL SYSTEM
by James F. McCraw
Senior Consultant, Rotating Equipment BP America, Inc. Houston, Texas
Vladimir Bakalchuk
Manager, Type 28 New Products
and Richard Hosanna

Type 28 Engineering Manager John Crane Inc. Morton Grove, Illinois
James F . McCraw is a Senior Rotating Equipment Engineer with BP America, Inc., Operation Efficiency Group, in Houston, Texas. In his 23 years with BP and Amoco, he has worked in machinery engineering, project support, plant maintenance, and machinery operation. In his position he is responsible for new equipment specifications, technology applications, research and development, and failure analysis for a variety of centrifugal compressors, pumps, gas turbines, and gas engines. Mr. McCraw holds a BSEE from Finlay Engineering/University of Missouri and is a member of the National Society of Professional Engineers. Vladimir Bakalchuk is Manager, New Products, at John Crane Inc., in Morton Grove, Illinois. With more than 25 years of experience in various fields of the oil and gas industry, he has spent the better part of 15 years specializing in dry gas seal/rotating equipment. Mr. Bakalchuk predicated his career in various technical capacities in the Ukrainian Academy of Sciences, the Mechanical Engineering Department of the University of Calgary, Brown-Root-Braun, Nova Gas Transmission, Canadian Fracmaster, and Revolve Technologies. Mr. Bakalchuk holds a degree (Mechanical Engineering) from Lviv Polytechnical University, passed Ph.D. candidacy at the University of Calgary, and is a registered Professional Engineer in the Province of Alberta.
37
Richard Hosanna is Manager of T28 Gas Seal Engineering with John Crane U.S.A., in Morton Grove, Illinois. He joined John Crane in 1983 and has held various positions within the applications and design engineering group, including 20 years with dry gas seals. Mr. Hosanna has a B.S. degree (Industrial Technology) and is the holder of three U.S. Patents associated with the sealing industry.
ABSTRACT
A compressor in a natural gas gathering service experienced multiple seal failures on the discharge end. Synthetic oil mist contained in the sealing gas was identified as a significant source of contamination that caused the seal failures. Since similar dry gas seals in other compressors at the same location were operating satisfactorily, a study of a particular compressor was undertaken. The seals were instrumented with thermocouples to monitor the temperature distribution across the seals. This paper discusses the findings of the study, resulting modifications of the dry gas seals, seal controls, and the compressor itself. This paper also outlines techniques that have been developed to mitigate the situation without having to shut the unit down. Also presented is a design approach to the sealing and seal monitoring in processes where liquid ingress into the sealing area may occur, including seal material selection, control system philosophy, seal head, and bearing cavity design.
38
INTRODUCTION
An oil and gas production company purchased a centrifugal compressor to increase the gas sales volume. Their chosen configuration consisted of an electrical motor driven centrifugal booster, and special attention was given to the design of the compressors dry gas seal system (Figure 1). The system control philosophy was based on remote diagnostics of all critical and noncritical seal system operating parameters. The system had also been designed to prevent process gas release into the atmosphere in case of a catastrophic seal failure.
developed based on similar calculations. By design, the system in case of a catastrophic seal failure of a primary and/or secondary seal would prevent uncontrolled gas release into atmosphere.
OPERATING BACKGROUND
The compressor was successfully commissioned and started. After a month of operation, a seal failure occurred. The compressors drive end seal failed catastrophically. Debris originating from the seals SiC rotating face damaged the seal head cavity during the failure and scored the lands during seal removal (Figure 2). Prior to the failure, operating personnel had noticed a decrease in barrier seal supply flow. At factory disassembly, the nondrive-end seal exhibited signs of contamination by both liquid and particulate. Upon disassembly of the barrier seal, it was noticed that seal O-rings were thermally hardened. The degree to which the O-rings hardened was depending on O-ring location: the farther the location was from the bearing cavity, the lesser was the degree of O-ring hardening. Although this fact had been noticed, no related action was taken at that time.
Figure 1. Facility Flow Diagram.
SYSTEM DESIGN
The control system design (refer to APPENDIX A) incorporated two stages of filtration: coalescing and particulate. The particulate filters (AJ-100A/B) were placed downstream of coalescing (AJ117A/B) to prevent seal contamination by fibrous debris in case of a coalescing element rupture. The sealing gas supply (PDV-120) was based on the differential pressure control principle (PDY-120), with balance piston cavity pressure used as a reference. Since the pressure control principle has been utilized for primary seal supply and not flow control, each compressor end seal supply flow was monitored by flow transmitters (FIT-120/125) to guarantee adequate supply volume. Traditionally dry gas seal performance is evaluated based on its leakage. Since in this application primary vent flow consisted of roughly the sum of primary seal leakage and secondary seal supply flows, in order to enable efficient leakage monitoring and prevent primary seal leakage dilution, the secondary seal supply employed flow control (FY-160/166). The primary and secondary leakage flows (FIT-150/151/155/156) and corresponding leakage cavity pressures (PIT-150/151/155/156) were monitored to provide indication of seal health. All critical and noncritical operating parameters were monitored and controlled by a discrete control system. The selected dry gas seal design incorporated a hard silicon carbide (SiC) rotating face running against a soft carbon stationary face. The interfaces between the seal and compressor head and the shaft were sealed with fluorocarbon O-rings, while spring energized polymers served as the seals internal secondary sealing elements. A circumferential carbon ring pressure bushing was used to separate the dry gas seal from the bearing cavity. The seals were installed into a separate seal head, which in turn, was installed into the compressor body. The supply and leakage porting was cross-drilled and piping connections were located on a circle terminating at flanges. Special attention was given to prevention of uncontrolled gas release into the atmosphere in case of a seal failure. It was determined that, in case of a catastrophic primary seal failure, in order to depressurize primary seal leakage cavity and redirect gas flow from the primary seal leakage chamber into the flare system, an additional primary seal venting area was required. To accommodate the requirement, two vent ports were incorporated into the primary seal leakage annulus. Also, to increase the venting area, control system secondary seal supply piping was transformed into vent piping by two sets of pneumatically actuated ball valves. In case of a trip on high high flow, one set of valves (SDV-160/166) would close, isolating the supply system upstream, and the other set (BDV-161/167) would open, connecting the secondary seal gas supply and, thus, seal the primary leakage cavity to the vent system. The secondary seal cavity depressurization scheme was
Figure 2. Head Damage at the O-Ring Location Resulting from Rotating SiC Ring Catastrophic Failure. The compressor head was repaired, a new seal installed, and the unit was restarted. In time, the flow to the barrier seal on the drive end progressively declined and eventually was reduced to a few liters per minute. At a later date, the unit was shut down to evaluate the barrier seal condition. The pattern of O-ring hardening persisted. During the shutdown, a small volume of liquid was found in the seal gas particulate filters located downstream of the coalescing ones (Figure 3).
Figure 3. Liquid Accumulation in Particulate Filter Housing. An analysis identified the liquid as a mixture of water and triethylene glycol (TEG). The gas used for the sealing supply was provided to the coalescor at 100F. To filter the mist contained in the gas, it would have been necessary to increase the gas temperature by incorporation of a heater into the stream. But the panel valve temperature rating limited the supply gas temperature increase and prevented additional gas heating. Various modifications to the existing coalescing filter elements were undertaken. The modifications eliminated liquid accumulations in the filter bowl itself, but liquid continued to exit the piping drain valve downstream of the filter. The existing coalescor was replaced by a high efficiency coalescing filter element. The liquid coalesced in the new filter chamber was identified as a mixture of polyalkylene glycol (PAG), TEG, and water. The PAG was used as lubrication/sealing oil in upstream screw compressors. The second seal failure occurred in a manner similar to the first one a year later (Figure 4).
39
Figure 4. Catastrophic Seal FailureSiC Fragments that Damage the Seal Head. It is widely accepted that in order to maintain reliable operation, a dry gas seal requires clean and liquid-free gas. In reality, a seal is tolerant, to a certain degree, of liquid contamination by vaporizable liquids, while reacting poorly to nonvaporizable liquids. Incompressible fluid ingress onto the seal faces produces two detrimental effects: sealing film instability and heat generation. In steady-state compressor operation, if pressure differential across the compressor does not suddenly change, the rotor usually does not experience rapid axial movements (Figure 5). Thus, if a non vaporizable liquid ingress into the seal has occurred at a steady-state compressor operation, heat generation in liquid shear becomes more of a concern than sealing film instability. Generated heat, if not dissipated, leads to a thermal expansion of the rotating parts of the seal and generation of high stresses. The typical mode of such a failure would be a fracture of a rotating sealing face followed by its disintegration due to consequential impacts in rotation. Interestingly enough, the same site had other compressors with dry gas seals that have been successfully operating for years on the same sealing gas, supplied from a common header. Nevertheless, the dry gas seals in these units appeared to be tolerant of the sealing gas composition, operating up to and in excess of five years.
was rejected by a coalescor that consisted of PAG and TEG/water. PAG, viscous synthetic oil, is used in flooded screw compressors in the facility for sealing and bearing lubrication. The increase in fluid viscosity adversely affects the coalescing ability of the filters. A filtration element vendor designed a special high-efficiency element (99.9998 percent at 0.1 micron aerosol) to enable reduction in the synthetic oil carryover. The element utilized the liquids velocity and temperature (viscosity) to maximize its coalescing ability. The second filter in a duplex arrangement was repiped to operate in series with the first one (Figure 6). As a result, a maximum concentration of PAG and TEG equated to 40 ppmw and water concentration to 160 ppmw.
Figure 6. Prefilter Arrangement. Compressor Seal Cavity Investigation showed that the design of a cavity between the balance piston and dry gas seal provided a potential for liquid accumulation. The same applied to the primary seal supply and leakage annuluses, as the correspondent ports were located in the upper quadrants. In case liquids were introduced into the primary seal, the accumulations would occur in the primary seal supply and leakage cavities annuluses (Figure 7). To prevent the liquid accumulation, the cavities were outfitted with the drains (Figure 8).
Figure 5. Mixture of PAG and TEG in the Seal Supply Annulus.
MODIFICATIONS
The issue of the compressor seal reliability can be approached from two directions: elimination of the liquid in the seal gas supply or creation of an operating environment for the seals that ensures that the liquid ingress impact is minimized. The first approach is always preferred because it eliminates the cause of the seal failures. Unfortunately, in many situations, such a source of clean gas does not exist and its creation is economically prohibitive. Since the other compressors at the site were operating satisfactorily and no other source of sealing gas existed at the site, a decision was made to study and modify the dry gas seal system. The goal was to reduce to a minimum concentration of the nonvaporizable liquid in the seal supply gas and to eliminate all additional sources of heat generation in and around the seal. Dry Gas Seal Supply Gas Conditioning A number of proprietary filter elements were tested to determine an optimum choice. While testing one, a two-phase liquid sample
Figure 7. Seal and Seal Head Prior to Modifications.
Figure 8. Seal and Seal Head after Modifications.
40
Dry Gas Seal Supply Path Part of the study was to determine what differed with regard to the compressors operating successfully at the facility from the one experiencing seal failures. When a seal gas supply path was examined, it was noted that the sealing gas was delivered immediately to the sealing faces in the problem unit, while in the other units, it was first directed onto a solid surface acting as a knock-out plate and than injected between the seal and a process side labyrinth, thus allowing for oil knock-out with further drainage away from the faces. To modify the existing design, an additional process side labyrinth was introduced. Now, the sealing gas entering the cavity is injected on the outer diameter (OD) surface of the new process labyrinth, which acts as a knock-out (Figure 8). Separated liquid is drained back to suction through an introduced drain. Dry Gas Seals In both seal failures, the compressor seal cavity was damaged by shattered SiC rotating faces. To avoid the lengthy cavity repair process resulting from damage by SiC bits, the seals were retrofitted with ductile (stainless steel with tungsten carbide [WC] coating) rotating faces. Since heat generation was the reason for both seal failures, the drive-end (DE) dry gas seal and barrier seal were retrofitted with temperature monitoring devices. Compressor Bearing Cavity The issue with oil being found in the secondary vent line started shortly after the initial startup. The barrier seals O-rings were found to be hardened. To eliminate all possible sources of heat generation, the barrier seal was redesigned to reduce the windage and eliminate the whipping action of protruding screw heads. A swirl brake was incorporated into the bearing cavity to facilitate oil drainage.
ANALYSIS
The thermocouple readings validated that the secondary seal appears to be the dominant source of heat generation. The liquid ingress into the seal cavity was minimized, but some liquid continued to be accumulating on the secondary seal faces. By design, in order to conserve nitrogen and allow for an accurate reading of actual seal leakage, the intermediate labyrinth supply was flow controlled to 1 acfm (velocity of 5 to 7 fps). The original design philosophy was based on an assumption that no liquids pass through the primary seal and, therefore, gas velocity of 5 to 7 fps would be sufficient to sweep clean leakage gas into the primary vent. But, this velocity was insufficient to prevent ingress of liquid trapped in the leakage gas into the secondary seal. Due to the fact that the primary leakage cavity pressure is small (around 1 to 2 psig), the secondary seal leakage is extremely lowa couple of Nl/min. It is not sufficient to provide cooling and its flow is not sufficient to remove oil film off the secondary seal faces. In comparison, the primary seal was being cooled by more than 100 scfm of cool sealing gas and was operating at over 200 psig pressure. As long as the secondary seal operated at a low pressure with a small leakage, the above condition existed and liquids were not removed from the secondary seal faces, the amount of heat generated in this condition could not be completely dissipated. Essentially, the compressors drive end seal head acted as a heat trap.
MITIGATION
There are very few ways of bringing heat generation under control in such a situation. The most easily attainable solutions include completely stopping the oil mist from depositing on the seal faces or to achieve an N2 flow to the intermediate labyrinth (secondary seal supply) that would be sufficient to remove the heat generated by the secondary seal faces. Such flow should remove the heat and prevent the sealing face fracture and/or create enough leakage flow through the faces to generate a velocity sufficient to remove a majority of the oil film of the faces or separate the faces. To achieve a solution to the failure problem in this application, the flow to the barrier seal was increased in steps. Temperature measurements indicated no impact on the barrier seal cooling on the secondary seal temperature, confirming that the secondary seal was the source of the heat generation. Next, the flow to the intermediate labyrinth (secondary seal supply) was increased in steps, allowing time for temperatures to settle. After a flow of 6 scfm was achieved as a result of these stepped changes to the secondary seal supply, the secondary seal temperature dropped and remained stable at the 210 to 230F level (Figure 10).
OBSERVATIONS
After the modifications were completed, thermocouple readings were taken twice a day and compared to the drains content in various locations. The temperature measurements established that the primary seal temperature was not subject to a high rise. The temperature increase in the secondary seal was the highest and started occurring shortly after the startup. The barrier seal temperature closely followed the temperature of the secondary seal, though the rate of increase was not as high. Since the liquid carryover was minute, the correlation between temperature rise and liquid presence in the filter drains and other drains proved to be complicated. Either there was no direct correlation or it was too difficult to discern the rise in the secondary seal temperature and the misting of the drains. Half a year after the startup, the secondary seal temperature was approaching 300F with the barrier seal temperature trending high (Figure 9).
Figure 9. Seal Temperature Trends Prior to Mitigation.
Figure 10. Seal Temperature Trends after Mitigation.
41
CONCLUSIONS
Although providing clean and liquid free seal supply gas remains a preferred option, many lessons could be learned from the study above. In order to enable a reliable operation of a centrifugal compressor in processes with a possibility of liquid ingression into the seal, a dry gas seal system and compressor design should incorporate the following:
Sealing gas should be diverted from direct entry into the seal. A
of the seals should be equipped with drains located at the six oclock position.
knock-out type entry should be provided in the seal gas supply cavity.
All seal cavities and stagnant flow cavities on the process side Control systems should be designed to accommodate possible
future increases in supply flow to both primary and secondary seals.
Dry gas seals should be outfitted with temperature monitoring

instruments.
Bearing cavity design should provide for a proper drainage of

lubricating oil preventing pressure build up inside the cavity and untrained oil accumulation.
Dry gas seal design, operating conditions permissive, should
utilize ductile coated mating rings to prevent cavity damage in case of the seal failure.
42
APPENDIX A DRY GAS SEAL CONTROL PANEL P&ID

Figure A-1. Primary Seal Supply System.
43
Figure A-2. Secondary and Barrier Seal Supply System.
Figure A-3. Leakage Monitoring System.
PREDICTION OF SUBSYNCHRONOUS ROTOR VIBRATION AMPLITUDE CAUSED BY ROTATING STALL

by Masayuki Kita
Manager, Compressor Designing Section Hiroshima Turbomachinery Engineering Department
Shinji Iwamoto
Acting Manager, Compressor Designing Section Hiroshima Turbomachinery Engineering Department
Daisuke Kiuchi
Hiroshima Turbomachinery Engineering Department
and Rinpei Kawashita

Research Engineer, Vibration & Noise Control Laboratory Takasago Research & Development Center Mitsubishi Heavy Industries, Ltd. Takasago, Japan
Masayuki Kita presently is the Manager of the Compressor Designing Section within the Turbo Machinery Engineering Department, Mitsubishi Heavy Industries, LTD., in Hiroshima, Japan. He is engaged in the design and development of single shaft and integrally geared process centrifugal compressors and expandercompressors for use in petroleum, chemical, and gas industry services that handle air and gas in accordance with API Standard 617. Mr. Kita received B.S. and M.S. degrees (Mechanical Engineering) from Shizuoka University. Shinji Iwamoto presently is an Acting Manager of the Compressor Designing Section within the Turbo Machinery Engineering Department, Mitsubishi Heavy Industries, LTD., in Hiroshima, Japan. He is engaged in the design and development of centrifugal compressors for use in petroleum, chemical, and gas industry services that handle air and gas in accordance with API Standard 617. Mr. Iwamoto received B.S. and M.S. degrees (Mechanical Engineering) from Kyushu University.
97
Rinpei Kawashita is a Research Engineer in the Vibration & Noise Control Laboratory in Takasago Research & Development Center, Mitsubishi Heavy Industries, LTD., in Takasago, Japan. He has been a specialist of rotordynamics and engaged in research and development of steam turbines, gas turbines, centrifugal compressors, and other rotating machinery for four years. Mr. Kawashita received B.S. and M.S. degrees (Mechanical Engineering) from Kyushu University.
ABSTRACT
The prediction of rotating stalls is one of the most important key technologies for compressors, especially high-pressure compressors. The operating ranges of compressors are restricted by rotating stalls because they may cause severe subsynchronous vibration of the rotor. As a consequence, there are many papers that have reported on the subject, including the influence of vaneless diffuser geometry on rotating stalls, and predictions of when a rotating stall occurs through experiment and theory. But, so far, there is no way to predict accurately the amplitude of subsynchronous rotor vibration caused by a rotating stall. In this paper, the pressure fluctuation caused by rotating stall and the resulting vibration of the rotor were measured using a test rig.
98
The external force on the impeller during a rotating stall was estimated from the measured pressure fluctuation. The subsynchronous vibration amplitude was then calculated using rotordynamics analysis by adding this external force. Comparing the result of this rotordynamics analysis with the measured subsynchronous vibration, an influence factor of pressure fluctuation to rotor vibration, , was calculated. The accuracy of this factor was checked by comparison with the result of a shop test on a compressor. Using this factor, the subsynchronous rotor vibration level caused by a rotating stall can then be predicted at the design stage.
INTRODUCTION
Many kinds of studies on rotating stalls have been carried out because this phenomenon not only adversely affects the performance of the compressor, but also causes subsynchronous vibration of the compressor rotor, especially with high-pressure services. The compressor operating range may be restricted by a rotating stall, because it usually occurs prior to a surge. It is usually inevitable that a rotating stall occurs prior to a surge for all designs. But what must be avoided is severe rotor vibration due to a rotating stall, not the rotating stall phenomenon itself. The purpose of this paper is to establish a method of predicting the subsynchronous rotor vibration level caused by a rotating stall at the design stage of centrifugal compressors.
probes (P1 through P6) were installed to measure pressure fluctuation at the vaneless diffuser. The tests were performed by three parameters, which were considered to influence the severity of rotating stall and the flow point at which a rotating stall occurs. The first parameter is diffuser width. From many reports that examined influences of the diffuser width on a rotating stall, the reduction in the diffuser width has a strong influence on the point at which a rotating stall occurs. The ratio of the diffuser width to the impeller outlet width was changed from 0.83 to 0.4. The second parameter is the clearance of the labyrinth behind the impeller. A rotating stall occurs when the direction of flow reverses locally at the diffuser due to reduction of flow. If the clearance of the labyrinth behind the impeller increases, the flow at the diffuser increases with increased flow. It is thought that this makes the point at which a rotating stall occurs a lower flow level. The clearance of the labyrinth behind the impeller was changed from 0 to .039 inches (0 to 1.0 mm) diameter. The third parameter is the machine Mach number. In general, it is known that if the machine Mach number increases, the pressure fluctuation caused by a rotating stall becomes larger, as does the point at which the rotating stall occurs. The machine Mach number was changed from 0.3 to 0.7. Test conditions are summarized in Table 1. Table 1. Summary of Test Conditions.
TEST BY SINGLE IMPELLER TEST RIG

Test Rig Arrangement The tests were performed in a test rig, the rotor of which has one impeller and was supported by the two ball bearings as shown in Figure 1. The selected design flow coefficient of the impeller is 0.02, because most rotating stall problems occur with low flow coefficient impellers. The test rotor was driven by a motor through a speed increasing gear.
Test Results The six test cases were carried out as shown in Table 1. Figures 2 and 3 show the pressure fluctuation and vibration spectra during the rotating stall in Case 1. By comparing these figures, it was found that there was a correlation between the pressure fluctuation and the rotor vibration at around 15 Hz.
Figure 2. Spectra of Pressure Fluctuation.
Figure 1. Test Rig Arrangement. The two vibration probes (horizontal and vertical orientation) were installed as shown in Figure 1. Six pressure measurement
Figure 3. Spectra of Rotor Vibration.
99
Figure 4 shows the relationship between the pressure fluctuation caused by a rotating stall and the flow coefficient for all cases. Figure 5 shows the relationship between the subsynchronous vibration amplitude and the flow coefficient for all cases. From these results, the followings are found:
Figure 7. Relationship of Pressure Fluctuation Measured by Pressure Measurement Probe P1 and P2.
PREDICTION OF VIBRATION LEVEL

Figure 4. Relationship Between Pressure Fluctuation by Rotating Stall and Flow Coefficient. Prediction by Pressure Fluctuation Level Assuming the pressure fluctuation magnitude at the diffuser is known, to establish the effect of a rotating stall on subsynchronous rotor vibration level, it should be examined how much the external force from pressure fluctuation affects the rotor vibration. If the influence factor of pressure fluctuation as the external force to the rotor vibration is found, it will be included in the rotordynamics analysis as an external force, and the subsynchronous vibration amplitude caused by the external force can be estimated. Figure 8 shows the relationship between the pressure fluctuation and the subsynchronous vibration measured with the test rig for Case 1. This figure shows that the vibration amplitude is proportional to the pressure fluctuation. Furthermore, the influence factor of pressure fluctuation to the vibration, defined by the following equation, can be derived:
Figure 5. Relationship Between Subsynchronous Vibration Amplitude by Rotating Stall and Flow Coefficient.
As the flow reduces, the pressure fluctuation increases. However,

as it reaches a certain level it does not increase any further.
As the flow reduces, the number of cells during a rotating stall
is increased from two to three. For many cases, as the number of cells increases, the pressure fluctuation and subsynchronous vibration amplitude decreases.
Figures 4 and 5 show similar patterns of behavior, but in Figure

5, the effects of changing the test parameters are larger. Figure 6 shows the ratio of the frequency of a rotating stall, f, to the rotational frequency of the rotor, N. From this figure, it is found that the frequency of a rotating stall increases as the flow decreases. Furthermore as the number of cells increases from two to three, the frequency of a rotating stall also increases.
where, F S PAC Pdout
= External force = Impeller projection area = Pressure fluctuation at diffuser inlet (impeller outlet) = Dynamic pressure at diffuser inlet (impeller outlet)
Figure 8. Relationship Between Pressure Fluctuation and Subsynchronous Vibration (Case 1). Figure 9 summarizes this factor for all cases. The factor varies with the test conditions, but the maximum value is approximately 0.6. Figure 6. Ratio of Frequency of Rotating Stall to Rotational Frequency of Rotor. Figure 7 shows the relationship of pressure fluctuation measured by pressure measurement probes P1 and P2 shown in Figure 1. The amplitude of the pressure fluctuation is larger at P1, because the rotating stall occurs near the impeller exit. However, by reducing the flow, the ratio of P2/P1 increases. This means that the region of the pressure fluctuation cell spreads radially as flow reduces.
Figure 9. Summary of Factor for all Cases.
100
Prediction by Dynamic Pressure Difference In order to know the pressure fluctuation level during a rotating stall, a computational fluid dynamics (CFD) analysis or experimental study is necessary. However, this would require enormous time and cost. Therefore a more efficient way to predict the subsynchronous vibration level was proposed. As an alternative of , the external force was derived from the difference in the dynamic pressures between the impeller inlet and outlet at the design stage. Figure 10 shows the relationship between the difference of the dynamic pressures and the subsynchronous vibration measured with the test rig for Case 1. Applying the same concept as , by introducing defined by the following equation, the influence of the pressure fluctuation on the vibration can be estimated. Figure 12. Block Diagram for Measurement. where, P Test Results = Difference of dynamic pressures between inlet and outlet of impeller Figure 13 shows the performance curve for the low-pressure section of the compressor under test conditions. During the test, a rotating stall was observed at the measured test points #341 and #351.
Figure 10. Relationship Between Difference of Dynamic Pressure and Subsynchronous Vibration (Case 1). Figure 11 summarizes this factor for all cases. Although the factor varies with test conditions, the maximum value is approximately 0.06. Figure 13. Performance Curve of Test Condition. Figures 14 and 15 show 3D spectra diagrams of the pressure fluctuation and the rotor vibration measured at test point #351. At this point, pressure fluctuations were observed at frequencies around 11.7 Hz and 23.4 Hz. These frequencies were considered to be caused by a rotating stall. At the same time as this pressure fluctuation, the vibration amplitude of the same frequencies was increased.
Figure 11. Summary of factor for all Cases.
COMPARISON WITH SHOP TEST RESULT OF AN ACTUAL MACHINE

In order to confirm the validity of factor , the pressure fluctuation and subsynchronous vibration were measured using an actual compressor. By the same procedure, the influence factor of was derived. The result was compared with the factor derived from the single impeller test results. Test Arrangement The shop tests were performed using a back-to-back type compressor, which has four impellers in the low-pressure section and three impellers in the high-pressure section. The impeller flow coefficients were approximately 0.008 to 0.024. The test was carried out under the condition of partial-load around 350 kW. During the test, rotor vibrations were measured using two vibration probes for each end (drive and nondrive end). At the same time, the pressure fluctuations were measured at the exit of the discharge scrolls for each section. The test arrangement is shown in Figure 12.
Figure 14. 3D Spectra of Rotor Vibration and Pressure Fluctuation for Low-Pressure Section.
101
In particular, at the frequency of 11.7 Hz, there was a strong correlation between rotor vibration-1 and pressure fluctuation-1. And at the frequency of 23.4 Hz, there was a strong correlation between rotor vibration-2 and pressure fluctuation-2. From these results, it was concluded that the frequency of a rotating stall for the low-pressure section was 11.7 Hz, and the frequency of a rotating stall for the high-pressure section was 23.4 Hz. Figure 18 shows the waveform of the pressure fluctuation level and the rotor vibration level. In order to compare each level at the frequency of a rotating stall, a band pass filter for 11.7 Hz was used. From these figures, one can see that the pressure fluctuation and the rotor vibration level tend to vary in the same fashion.
Figure 15. 3D Spectra of Rotor Vibration and Pressure Fluctuation for High-Pressure Section. Figures 16 and 17 show coherence between pressure fluctuation and rotor vibration. From these figures, there was a clear correlation between pressure fluctuation and rotor vibration at frequencies of 11.7 Hz and 23.4 Hz in addition to synchronous frequency.
Figure 18. Time Dependence Waveform of Pressure Fluctuation and Rotor Vibration.
VIBRATION ANALYSIS BY ROTORDYNAMICS

Figure 19 shows the model of the compressor rotor for rotordynamics analysis. For the analysis, the maximum, normal, and minimum bearing coefficients were used by considering the bearing lube oil temperature of 118.4F 5F (48 C 5C) and manufacturing tolerance of bearing clearance (Cp).
Figure 16. Coherence Between Pressure Fluctuation and Rotor Vibration (Vibration-1).
Figure 19. Compressor Rotor Model. The rotating stall was expected to occur at the last impeller stage of each section. Therefore, the unit external force from the rotating stall was added on the impeller of Stage 4 or Stage 7. The results of the analysis are shown in Figure 20. From these results, the subsynchronous vibration level from a unit force of a rotating stall was estimated.
Figure 17. Coherence Between Pressure Fluctuation and Rotor Vibration (Vibration-2).
102
At the design stage, should be considered as more than 0.15 in order to judge on the safe side. As the proposed value, of 0.20 should be used. The accuracy of should be improved by many future shop tests.
CONCLUSION
The effect of the pressure fluctuation from a rotating stall on the rotor subsynchronous vibration was investigated. The pressure fluctuation level and the subsynchronous level were measured with a single impeller test rig, and also by using an actual compressor. Using these test results and the rotordynamics analysis, the influence factor was estimated as the factor that shows how much pressure fluctuation works as an external force on the rotor. The estimated was approximately 0.15. of 0.20 was proposed in order to be on the safe side during the design stage. The accuracy of should be improved by many future shop tests.
BIBLIOGRAPHY
Ferrara, G., Ferrari, L., and Baldassarre, L, 2006, Experimental Characterization of Vaneless Diffuser Rotating Stall Part V: Influence of Diffuser Geometry on Stall Inception and Performance (3rd Impeller Tested), ASME Turbo Expo 2006: Power for Land, Sea and Air, GT2006-90698. Figure 20. Result of Rotordynamics Analysis. The vibration amplitude in Figure 20 was calculated using the unit force. In this way, the expected subsynchronous vibration amplitude could be calculated using the vibration amplitude at 11.7 Hz or 23.4 Hz in Figure 20 multiplied by F. By comparing the expected subsynchronous vibration amplitude with the measured vibration amplitude at the shop test, the factor could be obtained. Figure 21 shows the summary of the factor, , obtained by the above procedure. The resultant was approximately 0.15. Fulton, J. W. and Blair, W. G., 1995, Experience with Empirical Criteria for Rotating Stall in Radial Vaneless Diffusers, Proceedings of the Twenty-Fourth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 97-106. Kushner, F., 1996, Dynamic Data Analysis of Compressor Rotating Stall, Proceedings of the Twenty-Fifth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 71-81. Nishida, H., Kobayashi, H., Takagi, T., and Fukushima, Y., 1988, A Study on the Rotating Stall of Centrifugal Compressors (1st Report, Effect of Vaneless Diffuser Width on Rotating Stall), Transactions Japan Society of Mechanical Engineers (B Edition), 54, (449), pp. 589-594. Nishida, H., Kobayashi, H., Takagi, T., and Fukushima, Y., 1990, A Study on the Rotating Stall of Centrifugal Compressors (2nd Report, Effect of Vaneless Diffuser Inlet Shape on Rotating Stall), Transactions Japan Society of Mechanical Engineers (B Edition), 54, (449), pp. 589-594. Senoo, Y. and Kinoshita, Y., March 1977, Influence of Inlet Flow Conditions and Geometries of Centrifugal Vaneless Diffusers on Critical Flow Angle for Reverse Flow, ASME Journal of Fluids Engineering, pp. 98-103. Figure 21. Summary of Factor for Each Bearing Coefficient at Point #341 and #351. There was a difference between of 0.06 obtained from the single impeller test and of 0.15 obtained from the shop test. The reasons for this difference might be explained as follows: 1. Rotating stall may occur in the stages other than the last impeller and the induced external force may be larger than the assumption in this study. 2. The estimated increment of dynamic pressure from inlet to outlet of the last impellers was smaller than the actual one. Senoo, Y. and Kinoshita, Y., 1978, Limits of Rotating Stall and Stall in Vaneless Diffuser of Centrifugal Compressors, ASME Paper No. 78-GT-23.
ACKNOWLEDGEMENT
The authors gratefully wish to acknowledge the following individuals for their contribution and technical assistance in analyzing and reviewing the results, and for their great suggestions and guidance for practical applications and tests: K. Miyagawa, K. Yamashita, J. Masutani, and M. Ishikawa, of Mitsubishi Heavy Industries.
SELECTION OF MATERIALS AND MATERIAL RELATED PROCESSES FOR CENTRIFUGAL COMPRESSORS AND STEAM TURBINES IN THE OIL AND PETROCHEMICAL INDUSTRY
by Phillip Dowson
General Manager Materials Engineering
Derrick Bauer
Materials Engineer
and Scot Laney

Materials Engineer Elliott Company Jeannette, Pennsylvania
ABSTRACT
Phillip Dowson is General Manager, Materials Engineering, with Elliott Company, in Jeannette, Pennsylvania. He has 35 years of experience in the turbomachinery industry. Mr. Dowson is responsible for the metallurgical and welding engineering for the various Elliott product lines within the company. He is the author/coauthor of a number of technical articles, related to topics such as abradable seals, high temperature corrosion, fracture mechanics, and welding/brazing of impellers. Mr. Dowson graduated from Newcastle Polytechnic in Metallurgy and did his postgraduate work (M.S. degree) in Welding Engineering. He is a member of ASM, NACE, ASTM, and TWI. Derrick Bauer is a Materials Engineer with Elliott Company, in Jeannette, Pennsylvania. He joined Elliott Company in 2002, and has been involved with materials related R&D projects, failure analysis, production and aftermarket support, and remaining life assessments. Mr. Bauer received his B.S. degree (2002) from the University of Pittsburgh and is currently working toward his M.S. degrees from the same institution. Scot Laney is a Materials Engineer with Elliott Company, in Jeannette, Pennsylvania. He joined Elliott Company in 2007, and has been involved with materials related R&D projects, failure analysis, and aftermarket support. He also has experience in the areas of high temperature oxidation/corrosion and protective coatings. Dr. Laney received his B.S. degree (2001), M.S. degree (2004), and Ph.D. degree (2007) from the University of Pittsburgh in Materials Science and Engineering. He is also a member of ASM.
189
In todays marketplace the selection of materials for the various components for centrifugal compressors and steam turbines is very competitive and an important factor in the overall cost and delivery of the product. This paper reviews the material selection for major components for compressors and steam turbines such as shafts, impellers, blading, bolting, seals, etc. This paper is not intended to cover reciprocating, screw compressors, and high temperature hot gas expanders. Various material properties are discussed for the manufacture and service exposure of major components such as impellers, shafts, bolts, blades, and casings. Due to the various aggressive corrosive, fouling, and erosive environments in which both compressors and steam turbines are operated in, coatings must frequently be used to prolong the life of the machine. The application of various coating systems will be reviewed and how effective the coating system is with respect to withstanding the particular environment. Also discussed are the repairs for long lead time components such as rotors and the application of design proven repair procedures utilizing a design for fitness for service approach.
INTRODUCTION
Over the past 15 years a great deal of progress has been made with respect to the application of materials and related processes applied to various components for centrifugal compressors and industrial steam turbines. This paper will review how materials are specified for the various components and which material properties/factors one must consider for the application. Both rotating and nonrotating components of centrifugal compressors and steam turbines will be reviewed for material selection. Due to the very competitive market, material selection has moved beyond simply finding the material with the most ideal properties. Material cost and delivery can be one of the most important factors in the overall cost and delivery of the product, and therefore have become major drivers when selecting materials. Most original equipment manufacturers (OEMs) are continuously reviewing new ways whether by material changes or processes to reduce costs or delivery in order to remain competitive. This becomes even more important with the prevalence of outsourcing and inventory concepts such as just in time. OEMs are now competing not only for sales, but also as customers in places like casting and forging shops. At the same time, the environments the components are exposed to are increasing in severity, leading to the
190
need for more specialized materials. The use of these typically more costly materials may be avoided by selection of various coatings that are resistant to the aggressive environment under consideration. As stated previously, the paper will review materials selections for:
Rotating Components The heart of a centrifugal compressor is the rotor, which consists of a series of impellers and a shaft. The impellers are designed to accelerate the process gas, which causes it to be compressed in the proceeding diaphragm, while the shaft provides the support and rotation to the impellers. Impellers Given the importance of the impellers, a great deal of attention is given to their manufacture. Generally, centrifugal compressor impellers are fully shrouded, consisting of a solid hub and cover separated by radial equally spaced blades. The attachment of the blades to the impeller hub and cover can be either done by welding, integral cast, brazing, riveting, electrodischarge machining, integrally machined to the hub and/or cover, or a combination of these methods. A fully shrouded impeller is shown in Figure 1. Today, for most impellers, the blades are integrally machined to the impeller hub or cover and then welded to the nonmachined blade hub or cover.
Centrifugal compressorsboth rotating and stationary components. Industrial steam turbinesrotating and stationary components. Application of coatings for protection against corrosion, erosion,
and fouling environments.
CENTRIFUGAL COMPRESSORS
The choice of materials for rotating and stationary components of centrifugal compressors requires selection based on a number of design and environmental factors. The design factors that require consideration are the properties of the materials together with the operating requirements of the unit. For example, operating a machine intermittently or with variations is usually considered a much more severe service relative to constant, long-duration service, and must be designed accordingly. The properties of materials that can be utilized for use in the design are as stated by Cameron and Danowski (1973) and listed in Table 1. Long time, high temperature properties such as creep or creep rupture are rarely encountered in centrifugal compressors and, therefore, are not considered in this selection process. However, in cases where the temperature of operation is determined to be significantly different from room temperature, temperature is an important factor, since the following properties are temperature dependent: strength of material, corrosion rate, coefficient of thermal expansion, and fracture toughness. Table 1. List of Material Properties Used in Design of Centrifugal Compressors.
Figure 1. Photograph of a Fully Shrouded Centrifugal Compressor Impeller. Throughout the compressor industry selection of impeller materials can vary depending upon service operating conditions. Since operating conditions govern the material selection requirements, a number of mechanical and chemical properties such as yield and tensile stresses, low temperature properties, corrosion resistance, hydrogen sulphide resistance, weldability, and machinability must be considered. For over 50 years, OEMs have utilized impeller materials such as AISI 4330/4340, 4130/4140, AISI 410/17-4PH and 13Cr4Ni and nickel (Ni) base alloy. Impellers for centrifugal compressors are typically made from low alloy steels or stainless steels, and both NACE MR0175 (2003) and MR0103 (2005) have identical requirements for common impeller materials. Impellers made from AISI 4140 or 4130 steel have to meet a maximum hardness requirement of 22 HRC in the base metal, weld metal, and heat-affected zone (HAZ). In order to achieve the necessary hardness requirement, AISI 4140 impellers have to be austenitized, quenched, and tempered after welding to eliminate the HAZ. UNS G43200 and modified versions of G43200 with higher carbon contents are acceptable for compressor impellers by both NACE MR0175 (2003) and MR0103 (2005), provided that they are heat treated per the NACE standards. These UNS G43200 have a maximum yield strength requirement of 90 ksi, but there is no hardness limitation given per the NACE standards. AISI 410 stainless steel meets both applicable NACE standards if it is austenitized and double tempered with the second temper being performed at a lower temperature than the first temper. Both NACE MR0103 (2005) and MR0175 (2003) allow the use of low carbon stainless steel CA6NM (13%Cr-4%Ni) for impellers at a maximum hardness of 23 HRC after the material has been austenitized and double tempered at the temperature ranges defined in the NACE specifications. Precipitation hardenable stainless steels UNS 17400 (17-4PH) and UNS15500 (15-5PH) also meet both NACE specifications at a maximum hardness of 33 HRC after it has been through a solution annealing followed by a double aging treatment.
From an OEM perspective, receiving quality information with respect to the anticipated operating parameters is vital to producing a compressor that meets the customers expectations. API Standard 617 (2002) provides some guidelines as to what information should be shared by both the OEM and the purchaser. For instance, paragraph 2.2.1.3 requires the purchaser to specify any corrosive agents that may be present. As an example of why this is important, there is a substantial difference between the materials that are used in an air compressor when compared to an application that may contain wet chlorine. NACE MR0175 (2003) and MR0103 (2005) define which ferrous and nonferrous alloys can be used in wet hydrogen sulfide service to resist sulfide stress corrosion cracking. The alloys allowed by these specifications must also be heat treated in accordance with the specifications and meet a maximum hardness limit. NACE MR0175 (2003) is labeled Metals for Sulfide Stress Corrosion Cracking and Stress Corrosion Cracking Resistance in Sour Oilfield Environments, which applies to a component or machinery used for petroleum production, drilling, flow lines, and field processing facilities exposed to wet hydrogen sulfide service. NACE MR0103 (2005) is titled Materials Resistant to Sulfide Stress Corrosion Cracking in Petroleum Refining Environments, and is thus more specific to compressor components. Compressors with wet hydrogen sulfide present in the process gas are usually required to meet one or both of these NACE specifications to avoid stress corrosion cracking during service.
191
Since the maximum hardness requirement is 33 HRC for these steels, they are often used when impellers with higher yield strengths are necessary in hydrogen sulfide service environments. Although other manufacturing techniques have been utilized in the past 20 years, most impellers are welded using various weld processes. Other manufacturing techniques that have been utilized are:
Cast impellers. Electrodischarge

impellers.
machining to shape the gas passage of
Riveted impellers. Brazing impellers. Electron beam welding. Combination of electron beam welding and brazing. One piece machine impellers.
Early on, shielded metal arc was the welding process utilized for joining the sections of impellers, whether it would be milled blades to cover/hub to the unmilled section or joining preformed blades to a machined hub and cover. This welding process is still used today. In the 1980s several manufacturers started to utilize automation with other welding processes such as gas tungsten arc welding (GTAW), gas metal arc welding (GMAW), and submerged arc welding (SAW). Generally, these processes were applied to simple two-dimensional (2D) configuration impellers with 2D curved blades. Special torches were designed for both open bladed and closed configuration welding. Due to technological advances in five axis milling, two-piece three-dimensional (3D) geometric compressor impellers with various types of three-dimensional curved blades were being designed and manufactured. These impellers were designed to increase performance of the wheel. Since the impellers are two-piece designed, that is, either the blades are milled to the hub or cover, material selection becomes more important for consideration of manufacturability. Brazing of impellers has been used since the 1950s but in the recent years its popularity has increased. Early brazed impellers were done in a dry hydrogen atmosphere. More recently, however, impellers are done in a vacuum furnace, which produces more acceptable results and consistency. The inspection of the braze joints has also improved over the years with the application of C-scan immersion ultrasonic testing. The authors company utilizes a special calibration block to represent the impeller braze joint. This special block was utilized for the application of C-scan ultrasonic immersion equipment. The Figure 2 shown is a printout of an acceptable C-scan ultrasonic immersion test. Brazed impellers are also fluorescent penetrant inspected to validate the soundness of the braze joint.
impeller increases. This is illustrated in Figure 3, which shows the hardenability curve for various section diameters of AISI 4140. To understand why material mechanical properties such as AISI 4140, 4340, and 4330 diminish from the surface to the center of the material, one must review the continuous cooling transformation diagram (CCT) for that particular alloy and an elementary understanding of heat transfer. Figure 4 shows the CCT for AISI 4140 (Atkins, 1980). Depending on the rate of cooling, whether by air, oil, polymer, or water, for various thicknesses of material, certain microstructure phases such as ferrite, pearlite, bainite, and martensite are formed. Each one of these phases has an effect on the strength, ductility, and toughness of the material. Table 2 (Cameron, 1989) shows mechanical properties of impeller materials. Consequently, special consideration must be applied when selecting these low alloy steels in heavy forgings from which the blades are milled. For example, during austenitizing treatment to a fabricated impeller followed by a rapid quench, special attention must be made to changes in section thickness, such as that which occurs where the blades are welded to either the hub or cover in order to minimize the risk of quench cracking at the blade to hub/cover location. In some cases, quenching by forced air-cooling or salt bath may be required to avoid the quench cracking phenomenon. Other high alloy materials such as AISI 410, 17-4PH, and 13Cr4Ni have good hardenability with less drastic cooling rates; requiring only air or forced air cooling to achieve the desired mechanical properties. Figures 5 (Atkins, 1980) and 6 (Arlt, et al., 1988) show the CCT diagrams for AISI 410 and 13Cr4Ni, respectively.
Figure 3. Hardenability Curves for Standard and Controlled Yield Strength AISI 4140.
Figure 2. Immersion C-Scan Ultrasonic Results from a Brazed Impeller Joint. Materials such as low alloy steels have limited hardenability. Hardenability also becomes an issue as the diameter of the Figure 4. CCT Curves for AISI 4140.
192
Table 2. Mechanical Properties of Impeller Materials.
processes and brazing application. These procedures were developed for welding and brazing applications requiring high strength, qualification to meet NACE MR0175-200 (2003) (metallic sulfide stress cracking resistant materials), and low temperature (down to 166F (110C) applications. The Figures 7 and 8 show the properties that can be achieved using the 13Cr4Ni material (Dowson, 2002).
Figure 7. Plot of Strength and Hardness Versus Tempering Temperature (4 Hour Hold) for 13Cr4Ni.
Figure 5. CCT Diagram for AISI 410.
Figure 8. Plot of Charpy Impact Energy Versus Tempering Temperature (4 Hour Hold) for 13Cr4Ni. The precipitation grades such as Armco 17-4PH, Armco 15-5PH, and Carpenter Custom 450 obtain their mechanical properties by solution treatment and precipitation treatments. By varying the precipitation treatments, one can obtain a range of tensile strengths. In comparing the Armco 17-4PH with Armco 15-5, whose chemistries overlap, 17-4PH is more prone to form delta ferrite than 15-5PH, which may result in a loss of ductility/toughness. The precipitation treatments for these grades are usually in the range of 1050F to 1175F. For application in hydrogen and hydrogen sulfide environments, 17-4 or 15-5PH material is given a double precipitation treatment at approximately 1150F to obtain the necessary maximum hardness of HRC 33 maximum for H2S environments and 120 ksi maximum yield strength for hydrogen environments. Special Impeller Materials There are other special materials that are utilized in certain specialized applications. In moist halogen machines, such as a chlorine machine, Monel K500 has been successfully used for impellers. Monel K500 was also used in oxygen machines, because of its resistance to sparking. Special welding procedures were developed to obtain good quality fabricated Monel K500 impellers. Titanium and titanium alloys have been used for wet chlorine and in special cases where low density makes the material attractive. Precipitation hardenable nickel base alloys such as UNS
Figure 6. CCT Diagram for 13Cr4Ni. The material 13Cr4Ni (UNS S42400) has been utilized for manufacturing impellers to meet a wide range of requirements such as high strength, controlled hardness H2S, low temperature, and good corrosion resistance. For low temperature applications using 13Cr4Ni, controlling the volume fraction of austenite in the structure increases the material toughness. This increase in toughness also can lead to a reduction in hardness. Dowson (2002) also substantiated the application of a two-stage tempering treatment to obtain the maximum requirement of HRC 23 maximum. For welding, matching welding consumables are always applied to higher alloy steels. The authors company has formulated a patented chemistry for the welding consumables of 13Cr4Ni to be applied for impellers. This patent chemistry enabled HRC 23 maximum hardness to be achieved consistently for H2S controlled hardness environments. Procedures were developed using 13Cr4Ni for the various operating conditions utilizing the various welding
193
N0 7716 and UNS NO 7725 have been applied to aggressive coal gasification applications with a great deal of success. For temperatures down to 320F, the 9%Ni steel has widely been accepted for impellers in compressors for boil off gas from liquid methane due to its high fracture toughness at these low temperatures. Special processing of these grades is required to obtain the required mechanical properties. For some applications, aluminum alloys have also been used successfully especially in connection with liquefied natural gas projects. Rotor Shafts Generally, shafts are manufactured from either rolled bars or forgings. In the size range where both are acceptable by API 617 (2002) (currently inches finished machined), the choice between rolled bar or a forging should be dependent on cost and availability, as there is little difference in the end result, if manufactured to relevant American Society for Testing and Materials (ASTM) procedures. Tolerances on rolled bar can be more tightly controlled, which results in less machining required than for forgings and have a slight advantage in terms of thermal stability. Conversely, depending on availability of the particular material and size, selecting rolled bar may require the purchase of an entire length of bar or an entire heat, while forged shafts are made to order. Shafts with a finished diameter greater than 8 inches are generally made from forgings. For shaft materials the standard AISI 4330 and 4340 can be heat treated to give the required mechanical strength and toughness values. Due to these materials having higher strength values, the lateral expansion parameter is utilized for ASME Boiler and Pressure Vessel Code Section VIII Division I (2007). The requirement is 15 mils, which was proposed by Gross and Stout (1957). These materials have been used successfully down to 150F with the tensile strength at 110 ksi minimum and yield strength of 90 ksi minimum. For temperatures lower than 150F, ASTM A470 Class 7 has been used with a great deal of success. Shafts that are used in H2S environment cannot be limited to the NACE requirements due to the need for the higher strength required at the drive end of the shaft. Consequently, the applied stress in the main body of the shaft, which comes in contact with the gas, is low (<10 ksi). This value is below the threshold stress value for sulfide cracking as a function of hardness (Figure 9) (Warren and Beckman 1957).
the thrust bearing, are most susceptible to this type of failure. The process of wire wooling begins when a small particle of foreign material enters the bearing or seal. Through a series of localized temperature increases, due to high coefficient of friction between particle and rotor, material transfers from the rotor to the particle, and by hardening mechanisms, the particle becomes a hard, black scab that is able to cut and spin material from the shaft. Shaft material also continues to transfer to the scab as the cutting is occurring, allowing the scab to grow and propagate the failure. The result is a deeply grooved shaft. The material cut from the shaft often has the appearance of wire wool. Figure 10 shows damage to the journal and thrust collar due to wire wooling and the corresponding black scab on the journal bearing. For a more detailed description of the mechanism, the reader is directed to Fidler (1971).
Figure 10. Photographs Showing Wire Wooling Failure of a Journal and Thrust Collar and the Corresponding Black Scab on the Journal Bearing. For reasons that are not clearly understood, shaft materials with higher Cr contents are more susceptible to wire wooling. The critical Cr content proposed by Fidler (1971) is 1.8 percent. Because of this, shafts should be made of 4140 or 4340 wherever possible. If a higher Cr content material must be used, steps must be taken to reduce the risk of wire wooling. Altering the surface of the shaft or increasing clearances are ways that can be used to reduce the risk. One way to alter the surface of the shaft is to harden it by nitriding or hardfacing with hard chrome or carbide coatings. Sleeving with a wire wooling resistant material, such as 4140, can also work. Increasing the clearances reduces risk by increasing the size of particle able to initiate the failure. All of these have their pros and cons, which must be evaluated when deciding which method yields the best overall results. Stationary Parts The bulk of the centrifugal compressor consists of stationary parts. The most obvious is the casing, which is the pressure containing shell that surrounds the rotor. Diaphragms are responsible for slowing down the gas after it is accelerated by the impellers, causing it to be compressed. Bolts are used to fasten the parts together. Split line seals fit between the casing pieces to maintain the overall pressure. Abradable and rub tolerant seals maintain the pressure between each individual stage. Piping is used for gas path connections. Casings Nearly all multistage centrifugal compressor casings are made from carbon or a low alloy material that is either cast or fabricated from castings, forgings, plate, or a combination of these. Occasionally higher alloy steels are required for aggressive corrosion conditions or to achieve the desired toughness for extreme low temperature (320F) operating conditions, such as those for boil off gas compressors in liquified natural gas service. For low temperature application down to 175F, low alloy steels such as shown in Table 3 are applied in propane and ethylene compressors. The table must be used with some caution and understanding of the effect chemical makeup will have on the ability to achieve the desired toughness.
Figure 9. Threshold Stress for Sulfide Cracking as a Function of Hardness. The selection of shaft materials is also important to prevent certain types of failures, in particular, wire wooling. Wire wooling is an infrequent, but devastating, failure mechanism that occurs during startup (not necessarily the initial startup) in the bearing and seal area shafts. Areas where oil films are thin and loads are high, such as at
194
Table 3. Minimum Temperature for Low Temperature Application of Low Alloy Steels.
Table 4. Results of Chemistries and Their Corresponding Toughness Results.
Compressor casings are typically manufactured from carbon steel castings or carbon steel plate that is formed and welded. NACE MR0103 (2005) limits the hardness of carbon steel castings and plate to 200 HBW maximum and NACE MR0175 (2003) limits the hardness to 22 HRC (237 HBW). Under both specifications, carbon steel castings are acceptable in the normalized and normalized and tempered conditions, and carbon steel plate is acceptable in the hot-rolled condition as long as the material meets the maximum hardness requirement. The hardness of the carbon steel is largely controlled by the carbon content, so keeping the carbon content low is essential to meeting the maximum hardness requirement. Welding of the carbon steel plates under NACE MR0103 (2005) is controlled by NACE MR0472 (2005) which also limits the maximum hardness in the base metal and the weld metal to 200 HBW while limiting the hardness of the HAZ to 248 HV 10. The maximum hardness of 22 HRC is required in the base metal, weld metal, and HAZ under MR0175 (2003). The chemical composition of the filler metal used during welding must be similar or identical to the base metal composition. The hardness may be verified by the welding procedure qualification when all of the welding parameters and filler metal composition defined by the procedure qualification are controlled and followed. A post weld heat treatment (PWHT) may be performed to ensure that the hardness values meet the required specifications. NACE MR0103 (2005) has a lower maximum hardness limit than MR0175 (2003) to compensate for nonhomogeneity of some weld deposits and normal variations in production hardness testing using a portable Brinell tester. During the last 15 years, compressor casings have become larger with increases in pressure rating. This has also been applied to applications where the compressor casings are subjected to low temperatures. Design engineers now require thicker sections due to the higher pressure ratings to prevent gas leakage at the various joints connections. For the thicker sections, special controlled chemistry and heat treatments are required to achieve the desired toughness for low temperature application. For low alloy nickel steels, the elements C, S, and P are controlled to ensure good toughness is attained. Table 4 shows results of chemistries and their corresponding toughness results. These chemistries not only apply for steel plate but also for forgings and castings. Since the application of argon oxygen decarburization (AOD) and calcium-argon injection, cleaner steels, with sulfur as low as 0.005 percent, are attainable. Because of the addition of calcium compounds, the inclusions that remain resist elongation during rolling; remaining spherical. Consequently, inclusion shape control steel plates show improved ultrasonic quality, macrocleanliness, and mechanical properties as can be observed in Figures 11, 12, and 13 (Lukens Fineline Steels).
Figure 11. Comparison of Sulfur Contents of 50 Heats of Conventionally Processed Steel and 50 Heats of Lukens Fineline Double-O Five.
Figure 12. Comparison of Charpy Impact Energy for Conventional and Fineline ASTM A633C Plate, 4 Inches Thick, Normalized.
Figure 13. Comparison of Charpy Impact Energy for ASTM A633C Plates, 1 to 2 Inches Thick, Normalized, Transverse Orientation.
195
The introduction of inclusion shape control steel plates has also minimized the susceptibility of lamellar tearing during the fabrication of casings. Generally, this phenomenon can occur in thicknesses greater than 3 to 4 inches. By achieving lower carbon, sulfur, phosphorus, and inclusion shape control structures, excellent toughness properties can be achieved. However, for the larger thicknesses, accelerated quenching during heat treatment using water as the media from the austenitizing temperature is required. To maximize the quench affect of the water, the following is required: control of water temperature 80F maximum, good agitation of the water, adequate amount of water for the weight of component (2 gallons per pound of metal) and time from furnace to water less than one minute. In some cases movement of the component in the water sideways or up and down is required (Figures 14, 15, 16, and 17) (Metals Handbook, 1981).
In Camerons (1989) paper, the changes in the 1987 ASME Boiler and Pressure Vessel Code Section 8 Division I as required by API 617 (2002) specified the required Charpy V-notch energy absorption values had been made a function of the plate thickness. Refer to ASME Boiler & Pressure Vessel Code - Section VIII (2007) to show the values required. Since most of the compressor casings are fabricated, the welding consumables have to satisfy the impact test requirement of the base material. Actual test results of the weldments and surrounding heat-affected zone depend strongly on the preheat, interpass temperature, thickness of individual weld bead, heat input rate, and post weld heat treatment. Manufacturers have developed specific welding consumables, processes, and procedures to obtain the required mechanical properties for the casings. Since the soaking time of post weld heat treatments can affect the toughness of the materials as well as the weldments, simulated PWHT is performed on the materials that are used for low temperature application. Figure 18 shows effect of PWHT time on toughness (Charpy V-notch impact value).
Figure 14. Surface Cooling Power of Moderately Agitated Water Versus Water Temperature.
Figure 18. Impact Energy Versus Temperature for SA516 Grade 60 after Various PWHT. Diaphragms Figure 15. Effect of Concentration on Cooling Rate. In the past, diaphragms were generally constructed from grey cast-iron materials. When higher strength is required, ductile irons or fabricated mild steel is applied. In all cases, the casting or fabrications are heat treated to produce low levels of internal stress and difficulty with instability in service is virtually unknown. In the last 10 years, the choice of diaphragm material has been mild steel with the blades either milled integral or welded and the two-piece construction bolted together. One of the main reasons for the change is that the mild steel materials can be repaired by welding more readily as compared to grey cast-iron. The use of ductile iron that has improved weldability over grey cast-iron can also be selected as a material choice for diaphragms. Sulphide Stress Cracking There is extensive literature on sulfide cracking in oil well casing materials going back to about 1950. Incidents involving centrifugal compressors, however, have been rare as reported by Kohut and McGuire (1968) and Moller (1968). However, a failure due to sulphide stress cracking (SSC) can be serious; leading to loss of service of a vital link in the production chain of a chemical or petrochemical plant. For SSC to occur, the following conditions must be fulfilled:
Figure 16. Effect of Temperature on Cooling Rate.
Figure 17. Effect of Agitation on Cooling Rate.
Hydrogen sulfide must be present. Water must be present in the liquid state. The pH must be acidic. A tensile stress must be present. Material must be in a susceptible metallurgical condition.
196
When all the above conditions are present, sulfide cracking can occur with the passage of time. As stated previously, all manufacturers follow the requirements specified by NACE. In reviewing the work done Treseder and Swanson (1968) showed the effect of pH on sulphide stress cracking, Figure 19 where Sc is an experimental stress value. This Sc value increased substantially as the pH increased from 2 to 5. Other work done by Keller and Cameron (1974) has shown the importance of pH. In their work, quenched and tempered welded AISI 4140 material stressed to 80 percent of yield strength failed at pH 2.5, while at pH 4.2 no specimens failed. The yield strength of the base material was reported to be 126 ksi. From the same paper the material AISI 4140 quenched and tempered with a yield strength of 83 ksi was welded and given a temper treatment of 1100F. In this case, at pH 2.5 all specimens failed. At pH 4.2, a large percentage of specimens failed, while at 6.5 pH there were no failures in those tests. It is important to note that these are controlled laboratory experiments. In practice, it is difficult to specify a threshold concentration of H2S in a gas below which sulfide cracking will not occur, because other components of the gas can cause the pH to vary. Chlorides, for example, can lower the pH well below 4.3, while other components may cause an increase.
(2005) defers to NACE RP0472 (2005) which allows a maximum hardness of 248 HV10 in the HAZ and a maximum hardness of 200 HBW in the base metal and weld metal. NACE MR0175 (2003) requires a maximum hardness of 22 HRC at all carbon steel piping weld locations. When matching filler metals are used, a post weld heat treatment is not required if all hardness values meet the maximum hardness requirement. The hardness values may be verified by the welding procedure qualification if all welding parameters and the filler metal composition are defined by the qualification and are followed during production per NACE MR0175 (2003). Under NACE MR0103 (2005), which defers to NACE RP0472 (2005), 5 percent of all production piping butt welds must be hardness tested. Splitline Seals/O-Rings Turbomachinery casings usually consist of two or more pieces, which are bolted together. With respect to centrifugal compressors, the seam created is sealed using elastomeric materials in the form of O-rings. The choice of material used for this seal is vital to the operation of the compressor, as it is typically the weakest link in the casing in terms of chemical resistance, temperature resistance, mechanical properties, longevity, etc. Due to the wide range of operating parameters encountered in a centrifugal compressor and wide range of reactions to those parameters by the various elastomers, there is not a single, one stop choice. Table 5 is a compatibility chart for some common gases and O-ring materials. In fact, choosing the proper material frequently consists of an evaluation and acceptance of several compromises, particularly in cases with complex process gases. As a simplified example, fluoroelastomers will work well in hydrogen and nitrogen separately up to 300F. If they are mixed at 250F, such that a small amount of ammonia can form, the situation is no longer clear cut. The fluoroelastomer is not compatible with ammonia, nitrile has a maximum temperature of 212F, and silicone is not compatible with hydrogen. Clearly a compromise must be made to accept what shortcoming will occur when the chosen material is exposed to the offending environmental parameter. Table 5. Compatibility of Split Line Seal Materials in Various Environments.
Figure 19. Critical Stress for Sulfide Cracking as a Function of pH. Bolting Bolting for use in hydrogen sulfide service environments is also defined by NACE MR0175 (2003) and MR0103 (2005). Both specifications allow the use of AISI 4140 steel and AISI 410 stainless steel at a maximum hardness of 22 HRC provided the material is quenched and tempered. Both AISI 4140 and AISI 410 bolting is commonly available from bolting suppliers; however, an extra tempering treatment is required to meet 22 HRC maximum hardness requirement and conform to ASTM A193 Grade B7M for the AISI 4140 steel or ASTM A193 Grade B6M for AISI 410 stainless steel. For higher strength bolting materials, UNS 17400 and UNS 15500 stainless steel may be used. NACE MR0175 (2003) allows these alloys to be used at a hardness of 33 HRC while NACE MR0103 (2005) limits the hardness to 29 HRC maximum for pressure-retaining bolting. Piping The piping of a compressor unit is typically manufactured from carbon steel piping. NACE MR0103 (2005) limits the hardness of carbon steel piping to 200 HBW while NACE MR0175 (2003) allows carbon steel piping up to 22 HRC (237 HBW). The carbon content of the carbon steel pipe has a large influence on the hardness, so the carbon content must be limited to comply with the maximum hardness limit. Under both NACE specifications, the piping must also be thermally stress relieved following any cold deforming by rolling, cold forging, or any other manufacturing process that results in a permanent outer fiber deformation greater than 5 percent. For carbon steel piping welds, NACE MR0103
Abradable and Rub Tolerant Seal Materials Selection of abradable material for application in centrifugal compressors is very important. There are a number of property factors that need to be considered when selecting the abradable material:
Abradability and erosion resistance Compatibility of the abradable material with the gas Temperature limits of the abradable material Coefficient of thermal expansion of the material
Abradability and erosion resistanceIn the paper presented by Dowson, et al., (1991), abradable material mica-filled trifluoroethanol (TFE), nickel graphite, and silicon rubber were found to have good to very good abradability. The abradable silicon aluminum polyester, although not as good as those abradables stated previously, did perform well at lower rates of interaction and higher rubbing
197
velocities. The authors company has applied the silicon aluminum polyester abradable material in various centrifugal compressor applications with great success. Also since the 1991 paper, nickel graphite abradable seals have been used extensively in various applications for centrifugal compressors. However, to achieve optimum abradability and erosion resistance, special control of the hardness needs to be achieved. In the early years of manufacture of these seals, it was found that high heat input spray processes such as plasma could reduce the percent graphite in the coating and thereby reduce the abradability. During testing of an abradable nickel graphite seal the labyrinth rotating teeth were found to gall with the seal causing excessive damage to the impeller seal eye location (Figure 20).
Temperature limits of the abradable materialThe temperature limits for the various abradables are shown in Table 6. Table 6. Temperature Limits.
Figure 20. Damage to Nickel Graphite Impeller Seal. Compatibility of the abradable material with the gasThe mica-filled TFE is generally impervious to corrosive attack from most gaseous mixtures and would be acceptable with all hydrocarbon gases, sour natural gas, and chlorine gaseous conditions. However, for wet gaseous conditions (water > 2 percent) certain design considerations must be addressed to account for water absorption leading to possible swelling of the seal during service. The aluminum silicone abradables would operate under the same gaseous conditions that are generally applied to conventional aluminum seals. Exposed to extreme sour natural gases and chlorine gaseous conditions, these materials would be attacked severely. The nickel graphite coating could be used for all hydrocarbon gases and most sour natural gases. However, for hydrocarbon gases that contain large quantities of carbon monoxide (CO), a reaction can occur between the base coat of nickel and CO leading to delamination of the nickel graphite abradable seal as shown in Figure 21.
Coefficient of thermal expansion of the materialWhen designing seals, the coefficient of thermal expansion of the material must be taken into account at the design. The mica-filled TFE material has a coefficient well above that of steel and, therefore, dimensional changes due to temperature must be calculated in the overall design of the compressor. However, since the sprayed abradables are only 0.1 inch thick bonded to a metallic substrate, the coefficient of the substrate would apply for design purposes. For silicon rubber abradable material, the material is flexible/soft and will deform elastically to accommodate any thermal strains caused by the substrate. Rub Tolerant Polymer Seals Rub tolerant labyrinth polymer seals are seals with reduced clearances and, if contact is made between the stationary seal and smooth rotating member, the stationary teeth will deflect during contact without wear or damage to the rotor or seals. The rub tolerant plastic seals that are used in centrifugal compressors today are the new thermoplastics, which have better resistance to elevated temperatures. The thermoplastics matrix materials are tougher and offer the potential of improved hot/wet resistance. Because of their high strains to failure, they are the only matrixes that offer the new intermediate modulus, high strength (and strain) carbon fiber to use their full strain potential in the composite. These materials include such resins as polyetheretherketone (PEEK), which is intended to maintain thermoplastic character in the final composite. Others, such as polyamideimide (PAI), which is originally molded as a thermoplastic and is then postcured in the final composite to produce partial thermosetting characteristics. The partial thermosetting characteristic of the PAI enables an improved subsequent temperature resistance (Engineered Materials Handbook: Composites, 1987). When considering thermoplastic materials in rotating equipment, one has to understand their thermal properties (Table 7). The two thermoplastic materials used exclusively as labyrinth seals in centrifugal compressors are PEEK with additives and PAI. For the PEEK materials as labyrinth seals the temperature limit is dependant on the glass transition temperatures Tg of the material. Addition of additives such as chopped or continuous wound carbon fibers or graphite powder or PTFE will not increase the Tg of the material. The Tg is the temperature at which the crystalline polymer changes to a viscous or rubbery condition. In other words, the material has a dramatic change in properties. Generally, for labyrinth seals from PEEK material, the operation temperature is limited to 290F.
Figure 21. Delamination of Nickel Graphite Seal Due to Reaction of CO with Ni.
198
Table 7. Properties of Thermoplastics.
available in the 1950s and 1960s and over the past 30 to 40 years materials development concentrated on optimizing the properties of these materials for their application to larger components, which in turn, increases the reliability of components and their fabrications. Rotors Mechanical drive steam turbine rotors are manufactured from low alloy steel forgings. Rotors consist of a shaft and a series of disks, which may be monoblock, in which the disks are integral parts of the shaft (A type), disks shrunk fit onto the shaft (B type), or sections of the rotor may be welded together (C type). Mechanical drive turbines are in general A or B type rotors, with the majority being A type. For the higher temperature application up to 1025F, the 1Cr1MoV steels are currently in wide use but are limited to 1025F. For higher temperatures, the 12%Cr steel forgings have good experience at 1050F, where they have an excellent combination of creep strength, ductility, and toughness. However, there is a drive to use the most creep resistant version of 12%Cr up to temperatures of 1100F. In order to provide an additional margin of safety, the allowable creep strength at 1100F needs to be equivalent to that currently used with 12%Cr steel at 1050F (Gold and Jaffee, 1984; Armor, et al., 1984). For developing improved alloys, a tentative goal for rupture strength has been established at 14.5 ksi for a rupture life of 100,000 hours at 1100F. Figure 22 shows a comparison of 100,000 hour rupture strengths for various low alloy steels, 12%Cr and other development rotor steels. The Figure 23 shows the Larson-Miller rupture curves for commercial and developmental 12 percent rotor steel (Newhouse, 1987). Other critical material properties for rotor integrity are toughness, resistance to crack, initiation under creep, thermal fatigue and resistance to subcritical crack propagation in creep and fatigue. Toughness is discussed in more detail below.
For PAI materials, since the material is partial crystalline in character with some amorphous features, the Tg of the material is higher. However, when the crystalline character of the polymer is decreased, its resistance to solvents and water decreases also. The higher the degree of crystallinity, the higher the modulus and the higher the resistance to solvents and water. In the case of PAI careful consideration must be given to attack from amines, ammonia, oxidizing acids, and stray bases. Due to its partial crystallinity in character, PAI is prone to moisture absorption. Due to the moisture absorption, dimensional changes can occur that need to be addressed at the design stage and in manufacturing/storage. Generally, thermoplastic resin suppliers provide some information on water absorption after a 24 hour immersion in water. A more severe test is the 24 hour boiling water test. Due to the high thermal expansion coefficient of both the PEEK and PAI materials, accurate calculations of the growth of the seals with respect to the diaphragms and rotating components need to be done to enable one to calculate the clearances. Also, in the manufacture processing of PAI or PEEK into a tubular form, there exists a size limitation for polymer labyrinth seals.
PAIIn PEEK
tubular bulk form the size limitation is 35 inches. However, larger sizes can be constructed up to 45 inches by segmenting the seals. with chopped carbon fibersGenerally supplied in molded bulk form with a size limitation of 30 inches.
PEEK
with 68 percent continuous wound carbon fiber Generally supplied in a tubular form and there is no size limitation. Abradable seals have been used successfully in centrifugal compressors for reducing clearances and improving the efficiency of compressors. The authors company has applied abradable material such as Ni graphite, aluminum silicon polyesters, and fluorosint to numerous labyrinth seal applications with great success. Careful consideration must be applied to ensure that rub tolerant polymer seals can be utilized in centrifugal compressor labyrinths. The tests that were done at the authors company indicated that the PAI material for labyrinth seals may not be suitable for temperatures greater than 100F using similar clearances to that of abradables. The PAI material at a temperature of 150F was found to wear dramatically when it came into contact with the rotating member. Other polymer materials, such as PEEK or carbon wound PEEK, may be more suitable (Dowson, et al., 2004).
Figure 22. Plot Showing 100,000 Hour Rupture Strength of Various Rotor Steels.
STEAM TURBINES
Steam turbines present some different problems relative to centrifugal compressors. Temperatures are usually higher, which affects the various temperature dependent processes like corrosion rate and degradation mechanisms, such as creep, must now be considered. The key components of a steam turbine are the rotor, blades (buckets), casing, and bolting. With the exception of 12%Cr blading alloys, low alloy materials have been conventionally used for the major components of steam turbines for many years. For example, 1CrMoV has been used for high temperature rotor forgings and 3NiCrMoV for low temperature forgings. High temperature pipe work and castings for valve chests and turbine casings utilize 2CrMo or CrMoV. These materials were
Figure 23. Larson-Miller Rupture Curves for Commercial and Developmental 12 Percent Rotor Steel.
199
Toughness In CrMoV rotors, ASTM A470 Class 8 specification limits the fracture appearance transition temperature (FATT) to 250F. The FATT values for the base material normally range from 185 to 260F. Since the advancement of fracture mechanics technology, it has now become possible to characterize toughness in terms of a critical crack size ac. The cold start sequence of a rotor is most critical and the critical flaw size ac based on analysis of temperature and stress can lead to catastrophic failure (Viswanathan and Jaffee, 1983). For the Gallatin rotor, the temperature at the time of the peak stress 74 ksi was 270F. The combined crack size reached its lowest value 0.27 inches at this temperature and stress (Figure 24).
An ASTM special task force on large turbine generator rotors of subcommittee VI of ASTM Committee A-1 on steel has conducted a systematic study of the isothermal embrittlement at 750F of vacuum carbon deoxidized (VCD) NiCrMoV rotor steels. Elements, such as P, Sn, As, Sb, and Mo were varied in a controlled fashion and the shifts in FATT, () FATT were measured after 10,000 hours of exposure. From the results, the following correlations were observed in Equation (2):
where: FATT is expressed in degrees Fahrenheit and the correlation of all the elements are expressed in weight percent. According to this correlation, the elements P, Sn, and As increase temper embrittlement of steels, while Mo, P, and Sn interaction decrease the temper embrittlement susceptibility. All available 10,000 hour embrittlement data are plotted in Figure 26 as a function of calculated FATT using Equation (2) (Newhouse, et al., 1972). A good correlation is observed between calculated and experimental FATT. The scatter for these data is approximately 30F for 750F exposure and 15F for the 650F exposure.
Figure 24. Illustration of Cold-Start Sequence and Associated Variations in Stress (), Temperature (T), and Critical Flaw Size (ac) as Functions of Time from Start. Variations in temperature, stress and material in homogeneity throughout the rotor dictate the critical flaw size ac. By using lower scatterband values of KIC, the critical flaw size can be converted (Schwant and Timo, 1985). Another method of estimating the lower-bound values of KIC for CrMoV steel is by using the expression (Jones, 1972):
where: KIC is expressed in MPa m T is in C Although equipment manufacturers have records of the FATT value of the rotor material prior to service, the effect of temper embrittlement during service increases the FATT value and decreases the KIC value (Figure 25) (Viswanathan and Jaffee, 1983).
Figure 26. Correlation Between Compositional Parameter N and the Shift in FATT of NiCrMoV Steels Following Exposure at 650F and 750F for 8800 Hours. Other correlations for determining the temper embrittlement susceptibility of steel, such as the J factor proposed by Watanabe and Murakami (1981) and factor proposed by Bruscato (1970), are widely used. These factors are given by:
Figure 25. Effect of Temper Embrittlement on Fracture Toughness of CrMoV Rotor Steel. The maximum temper embrittlement occurring at location of the rotor exposed to temperatures from 700F to 800F. Consequently, evaluation of FATT (or KIC at the concerned location) in the service exposed condition is critical for damage assessment (Dowson, et al., 2005).
The Figures 27 and 28 show relationship between increase of FATT and J factor and factor at 750.2F (399C) for a 3.5%NiCrMoV steel.
200
Steam Turbine Blades (Buckets) There are basically three groups of steam turbine blade material used by turbine manufacturers. These are various grades of 12 to 13 percent chromium (cr) steels with additions of Mo, W, Cb, and V, higher chromium precipitation hardening steels such as 17-4PH and titanium alloys. Table 8 gives a listing of the commercial available materials used in blading (ASTM A1028, 2003). Additional data regarding the heat treatment and mechanical properties can be found in Aerospace Structural Metals Handbook: Volume 2 (1988), and Briggs and Parker (1965) (Figures 30 and 31). Table 8. Composition and Mechanical Properties of Commercial Blading Materials.
Figure 27. Correlation Between Compositional Factor J and the Shift in FATT of NiCrMoV Steels Following Exposure at 650F and 750F for 8800 Hours.
Figure 28. Relationship Between Increase of FATT and . In the high temperature regions, creep and creep rupture are a concern. The traditional approach is to use a Larson-Miller plot as shown in Figure 29. The design stresses are generally based on the 105 hours smooth bar creep rupture stress divided by some appropriate safety factors.
Figure 30. Rotational Bending Fatigue Behavior of Type 403 at Various Temperatures.
Figure 29. Larson-Miller Stress Rupture Curve for 1Cr 1Mo V Rotor Steel.
Figure 31. Stress-Range Diagrams at Various Temperatures for Type 403 Heat Treated to 26-32 HRC.
201
In designing blades one must take into account the material properties, the operating conditions, and the quality/purity of the steam. The principal failure mechanisms that occur in blades are high cyclic failure and creep rupture. Blades are designed to prevent creep rupture failures; therefore, it is rare that steam turbine blades fail by this mechanism. In general, most blade failures are due to high cycle fatigue caused by a number of factors. Some of the factors are listed below:
Dynamic
stresses caused by nonsteady steam forces, nozzle wakes, thermal transient per revolution diaphragm harmonics, and flow instabilities
Strong exciting harmonics of rotational speed such as the nozzle

passing frequency corrosion
Steam purity and its effect on corrosion in fatigue and pitting

Speidel (1981) highlighted the satisfactory corrosion fatigue resistance of X21CrMoV121 steel in a good purity steam (Figure 32). However, the growth of fatigue cracks in 12 percent chromium steels is greatly enhanced by the presence of chloride solutions at low cyclic stress intensities and high mean loads (Figure 33). Speidel (1981) also illustrated how the Kth (threshold stress intensity) is greatly reduced by the presence of certain environments (Figure 34). Other work done by Batte and Murphy (1981) showed similar results on the fatigue strength reduction in various aqueous solutions (Figure 35). With these data, it can be concluded that the most concerning conditions for fatigue crack growth is a high mean stress with superimposed cyclic stresses in the presence of aqueous solutions. These conditions are the service conditions that steam turbine materials are subjected to. It is well known that shot peening enhances the fatigue resistance of metallic materials by introducing compressive stresses at the surface layer of the component. However, in a corrosive environment, pits may penetrate the compressive stress layer, thereby negating the benefit gained by peening (Figure 36). The importance of steam purity and corrosion resistance of the blade material are clear when considering that it is not feasible in most cases to avoid the stress state and exposure to aqueous solutions present in the turbine and that treatments such as peening are not reliable. Figure 33. Effect of Chloride Solutions on the Fatigue Crack Growth Rate of 12%Cr Steels.
Figure 34. Effect of Environment on the Threshold Stress Intensity Kth of 12%Cr Steels.
Figure 32. Fatigue and Corrosion Fatigue of X21CrMoV 121 in Air, Deaerated Water, and Aerated Hot Chloride Solutions.
Figure 35. Effect of Various Aqueous Environments on Fatigue Strength.
202
Nozzles and Diaphragms In steam turbines, stationary nozzles and diaphragms are selected based upon stress/temperature, oxidation, and corrosion. The blades of nozzles and diaphragms are made from wrought 13%Cr series stainless material. The blade holders are manufactured from mild steel, ductile iron, or low alloy steels. The choice of material depends on the method of construction and the operating design stress at temperature. During service, erosion due to wet steam can occur on the latter stages of the steam turbine diaphragms. Normally repairs in those areas are done by depositing weld material that has improved water/steam erosion resistance, i.e., AWS E309/ER309 or Inconel weld consumable. In some instances, a mechanical fix may be performed using an austenitic stainless or nickel base wrought material. High Temperature Bolting In a steam turbine, bolts in flange joints operating in the creep range at temperatures up to 1050F must be able to withstand steam pressures up to 2 ksi or in some cases even higher. The main requirement of bolts in the creep range is to maintain joints without relaxing below the required design stress limit, which may allow leakage. Consequently, the important property for bolts is the stress relaxation characteristics of the material. For a given joint, the load required to exceed the steam load is applied to the flange area by tensile loading of the bolts. During service, creep in bolt causes relaxation of this initial load. The elastic strains produced by initial tightening of the bolts are progressively converted to creep strains; elongating the bolt and thereby, reducing the effective load in the joint. For design, the final relaxation stresses must be in excess of the design stress to keep the joint tight. Depending on the material and duty requirements, bolts are usually tightened to a predefined cold strain, e.g., 0.15 percent from which the initial stress on the bolt can be calculated. Consequently, the bolts need to retain their design stress between each overhaul when the bolts are removed for maintenance of the machine and retightened on reassembly. Typical composition of bolt steel, their properties, and various national standards are summarized in Tables 10 and 11 (Everson, et al., 1988). For convenience, the Central Electricity Quenching Board of the United Kingdom has classified these compositions into grouping numbered one through eight. Several of these groups of materials show stress relaxation behavior after 30,000 hours as a function of temperature when subjected to a cold prestrain of 0.15 percent (Figure 37) (Branch, et al., 1973). In selection of bolt material, the compatibility of the thermal expansion coefficient of the bolts with respect to the joint materials, as well as its susceptibility to various fracture mechanisms, must be taken into account. When sufficient stress relaxation data are not available, one can apply the useful correlations that exist between the creep rupture data and relaxation stress at 0.15 percent strain (Figure 38). Table 10. Various National Bolting Material Specifications. The optimization of low alloy materials has taken place over a period of 50 years or more. The development and optimization of the 9 to 12%Cr have only just begun. A lot of work remains to be done to characterize these materials for design lives of up to 250,000 hours and to cast variation in their properties. The development of joints between the modified 9%CrMo and low alloy material has been done in order to optimize selection of welding consumables, welding parameters, and post weld heat treatment. Creep tests on such joints have indicated that in long-term tests, their rupture strength falls near the lower bound of the low alloy parent materials strength. Due to composition gradients, carbon depleted, ferritic zones can form on the low alloy side of the interface between the high and low alloy materials. Consequently, joint locations are designed such that the temperature and stress are lower so that creep and carbon diffusion in service will be very limited.
Figure 36. Effect of Environment on Fatigue Resistance of Shot Peened 12%Cr Steel. Casings The various composition of steels used for turbine casings are shown in Table 9. During the years the drive toward improving creep strength to accommodate the steadily increasing temperatures led to progressive changes in material from the C- Mo and 1CrMo to the 1Cr1MoV and 2Cr1Mo. In the late 1960s and early 1970s, numerous instances of reheat cracking (stress relief cracking) in the weld heat affected zones occurred in the 1Cr1MoV and the importance of rupture ductility was realized. The material CrMoV was standardized by some manufacturers who implemented stringent specifications relating to control of residual elements (particularly phosphorus, antimony, tin, copper, aluminum, and sulfur), deoxidation practices, and welding procedures. Casing designs were modified to eliminate manufacturing and in-service reheat cracking. Other manufacturers utilized the 2Cr1Mo steel due to its higher creep ductility, higher low cycle fatigue resistance, and better weldability. The current designs use either CrMoV or the 2Cr1Mo steel material. For steam temperatures of 1050F and above, and up to 1100F, the 9 to 12%Cr casting grade steels will be utilized. Table 9. Various Composition of Steels Used for Turbine Casings.
Table 11. United Kingdom Bolting Material Specifications.
203
Figure 37. Comparison of Stress-Relaxation Behavior after 30,000 Hours as a Function of Temperature for Various Bolt Materials Subjected to a Cold Prestrain of 0.15 Percent.
Figure 40. Effect of Double Vacuum Melting on Stress-Rupture Properties of 1Cr-Mo-V Steels at 1020F . Other improvements in notch ductility were in a new class of CrMoV containing titanium and boron with subsequent grain refinement. Further improvements in the rupture ductility of 1CrMoVTiB steel have been made by reducing the major embrittling elements. The application of very clean steel practices have shown beneficial effects on ductility and notch strength. This material will continue to be used as a very cost-effective option for temperatures up to 1049F (565C). Also nickel-based super alloys such as Nimonic 80A and IN901 have been used for some of the high temperature bolts where stress relaxation is an issue.
REPAIRS
In the rotating equipment industry, service of components requires either replacement with new or refurbishment of the components. The refurbishment can mean restore the component to its original dimension either by welding/mechanical fix or other sprayed/plating processes. All OEMs have developed repair procedures for most of the long lead delivery components, e.g., components such as impeller, shaft, casing, and diaphragms. API RP 687 (2001) specifies the minimum requirements for performing the repairs to rotors. Throughout the last 20 years hundreds of rotors and shafts for both compressors and steam turbines have been successfully weld repaired by various OEMs. Welded rotor restoration, like any other critical repair technology, requires a highly analytical approach to assure component and machine reliability. Most OEMs develop property data for weldments and heat affected zones for the various materials. This would include materials for both steam turbines and centrifugal compressors. The data that are generated, but are not limited, are as follows:
Figure 38. Relationship Between Relaxed Stress and Rupture Strength at the Same Duration for Times from 1000 to 30,000 Hours and Temperature from 885 to 1110F (475 to 600C). Fracture of a bolt can occur if the local creep strain reaches the creep ductility of the material from which the bolt was made. Consequently, low rupture ductilities lead to notch sensitive bolt material. Numerous failures of 1Cr-1Mo-V steel bolts were attributed to notch sensitive failures. By reducing the impurity elements of the bolt steel, appreciable improvements have been made to rupture ductility without compromising the creep strength (Figure 39). The residual element content can be reduced by careful scrap selection, avoidance of air melt, and use of double vacuum melting (Figure 40).
Room and elevated temperature tensile properties Impact and fracture appearance transition temperature data Creep/stress rupture data Fatigue properties Stress corrosion cracking (SCC) threshold limits
Several papers (LaFave, 1991; Dowson, 1995; Dowson and Wiegand, 1996) outline some of the material property data that can be generated. Figures 41, 42, and 43 show some generic examples of data that are typically used. These data allow for the weld metal properties to be compared against mechanical analysis results and company design criteria to assure that the proposed repair will be reliable. Shorter term testing such as weldability, tensile hardness, and toughness is routinely done on the base metal, weld metal, and HAZ for each engineered rotor repair.
Figure 39. Relationship Between Residual Element Content and Ductility.
204
Compressors Corrosion Corrosion is a common problem in several industries, including turbomachinery. It is an electrochemical attack that can occur as an even attack of the surface (general corrosion) or uneven attack (pitting), as well as lead to stress corrosion cracking. Coatings for corrosion protection work by acting as a barrier that separates the substrate material and the environment. Typically, the coating material is viewed as a sacrificial layer and has slower reaction kinetics than the base metal in the particular environment. For selection of materials for sour gas service, API 617 (2002) refers to NACE MR0103 (2005) and MR0175 (2003), which allow the use of coatings for general corrosion resistance, but not for protection against stress corrosion-cracking. This is due to the fact that localized defects in the coating can lead to a stress corrosion crack. With any coating, there is always a risk that the coating will be removed at a small location, either by erosion, localized pitting, or foreign object damage, which will expose the substrate to the corrosive environment. Diffusion coatings are too thin to provide a complete barrier between the substrate and the environment. Metallic, electrolytic coatings contain microscopic cracks that will allow the H2S to contact the substrate material. Hydrogen sulfide can diffuse through polymer coatings to reach the substrate while thermal spray coatings are porous and will not provide adequate protection. There appears to be a limited number of coatings advertised by the various OEMs that are applied solely for corrosion protection. Part of this is due to the fact that changing the base metal to a more corrosion resistant material, such as a stainless steel, often gives satisfactory results. A second reason for the lack of corrosion coatings is that corrosion is often accompanied by fouling in centrifugal compressors and therefore the antifoulant coatings, discussed below in more detail, typically are designed to prevent corrosion as well. One coating that the authors company has used with some success is a baked phenolic coating. This coating is used in certain applications where it performs better than stainless steel or where the increased price of using stainless steel cannot be justified by the end user. Fouling Fouling is a common problem in compressors and to some extent, steam turbines. Fouling refers to the build of solids, usually polymeric materials, on the internal aerodynamic surfaces of the machine. While it does not usually lead to catastrophic failure, it does gradually reduce the efficiency of the machine by increasing the mass of the rotor, altering the aerodynamics, and blocking flow paths. If left unchecked, fouling can block the flow path to the extent that production is stopped or cause imbalances that can damage the machine. Depending on the service, fouling substances may come from outside of the machine or be generated internally. External foulants may come from airborne salt, submicron dirt, and organic or inorganic pollutants in the process gas. A well-maintained filtration system usually helps to minimize this type of fouling (Meher-Homji, et al., 1989; Guinee and Lamza, 1995). In petrochemical compressors, the situation is much more complicated, as the foulants can be generated internally. For example, in ethylene cracked gas compression, fouling results from the polymerization reactions intrinsic to the compression process. Fouling has imposed significant cost on petrochemical production. A material that is required to resist fouling must have excellent release properties. Materials with a combination of low coefficient of friction and chemical inertness are usually used for this application. A common and widely known coating material for centrifugal compressors is polytetrafluoroethylene (PTFE [Teflon]). These are multicomponent, sprayed coatings designed for fouling and corrosion resistance. Wang, et al. (2003), showed a dramatic decrease
Figure 41. Typical Larson-Miller Curve Generated from Various Weldments.
Figure 42. Typical S-N Curve from Fatigue Testing of Weldments, Fully Reversed (R = 1), Endurance Limit Set to 108.
Figure 43. Stress Corrosion Crack Growth Rate as a Function of Stress Intensity of Weld Metal, HAZ, and Base Metal. When repairs to impellers are required, it is generally due to some form of mechanical damage or corrosion/erosion attack. If the impeller has failed by fatigue, the component is generally replaced. When corrosion attack has occurred, repairs will be made by applying corrosive resistant welding consumables. For erosion attack, various coatings can be applied to extend the life of the impeller.
COATINGS
Despite best efforts in choosing a suitable alloy, it is frequently impossible to find a single material that is ideal for the particular application. In this case, composite systems, i.e., coatings, may be used to exploit the properties of two or more materials. There are several reasons why coatings may be used, ranging from enhancing a particular property for a specific application to reducing cost by allowing for the use of less expensive substrate materials. Coatings are also often designed to be multifunctional, addressing multiple problem areas simultaneously. The petrochemical industry presents several opportunities for the usage of coatings due to the frequently harsh conditions encountered in this service.
205
in time required to release an applied foulant on samples coated with two PTFE type coatings offered by the authors company versus bare steel samples (compare E and E+ versus steel in Figure 44). Figure 45 shows an example of a compressor rotor with PTFE coated impellers. Unfortunately, PTFE coatings are removed by erosive liquids (example water washing) or solids. Electroless nickel (EN) has also been shown by Dowson (2007) to exhibit excellent release properties (523 in Figure 44), while remaining adherent in erosive conditions. In fact, the release properties are as good as or better than results from PTFE. EN is applied by submerging the part into a Ni and P containing solution where an autocatalytic process plates the part with a well-bonded, amorphous Ni-P alloy. The P in the alloy is believed to be responsible for the release properties, while amorphous nature aids in corrosion resistance.
Operating Status of the Electroless Nickel Coating The compressor with EN coating had been successfully operated since the 2001 turnaround until 2006. The rotor vibration has been monitored at a much lower level than the previous run periods, as shown in Figure 46, which indicates that the fouling has been effectively controlled by the EN coating. This improvement is attributed to the ability of EN to withstand the heavy washing operation as compared to the 3P coating. The oil washing and chemical injections have been kept in the same manner as in the previous operating periods. The Corunna plant feels that the EN has successfully functioned as an antifouling coating. When the unit was replaced in 2006 with a new machine, the customer requested the same successful EN coating to be applied to the rotor and internals.
Figure 44. Comparison of Fouling-Release Performance of Bare Steel and Coated Panels.
Figure 46. Vibration Data Before and After Application of Electroless Nickel to the Rotor. After the original unit was pulled out of service, examination of the rotor shows that the EN coating remained in tact after the five year operational period. Pictures of a coated impeller are given in Figures 47 and 48. The coated impeller is free of corrosion and foulant buildup, and the impeller serial number remains readable. This is proof of the remarkable antifoulant properties of the EN coating.
Figure 45. Centrifugal Compressor Rotor Coated with a PTFE Type Coating. Case Study of Electroless Nickel Coating The application of electroless nickel as an antifoulant coating is a relatively recent advance. A brief survey of antifoulant coating offerings by OEMs shows that the majority do not offer an electroless nickel coating. Due to the current lack of offerings, an update of a case study initially introduced by Wang, et al. (2003), is presented below in order to justify the use of electroless nickel as a corrosion and foulant resistant coating option. Background A major chemical plant in Corunna has a centrifugal compressor that used to be coated with the coating 3P for antifouling. The coating had suffered severe deterioration two times during the four year operation since 1997. Analysis indicated that the deterioration was related to the heavy washing injection, which contained aggressive chemical additives, and steam-cleaning operation. However, the washing and cleaning were essential to the plant because of the extensive fouling and efficiency drop. In order to withstand the injections, the compressor rotor and diaphragms were recoated with EN during a plant turnaround in September 2001. The compressor had been in service until 2006 with satisfactory performance.
Figure 47. Picture of First Stage Impeller with Electroless Nickel Coating after 5 Years of Service. The Coating Remains Intact and the Serial Number Is Clearly Visible.
206
these hard, erosion coatings, is the method of application. Carbide coatings are typically applied using techniques such as high velocity oxyfuel (HVOF), plasma spraying, or detonation gun. These processes have certain requirements, such as spray distance or line-of-sight, that may not be possible for the gas paths of smaller, closed impellers or other parts with poor accessibility. Steam Turbines Corrosion Under ideal conditions, corrosion is not a major issue in steam turbines. Stainless steel materials perform extremely well without coatings in the elevated temperature, steam environment. The corrosion product formed on the stainless steel is usually a thin, uniform layer that grows slowly with time. Corrosion becomes a problem when there is an underlying problem with the process. Steam purity is the most common problem. Impure steam carries with it elements that increase the corrosion process. Typically, the amount of impurities in the steam is not large enough to change the general corrosion rate; rather they cause localized attack that leads to pitting. These pits provide easy initiation sites for fatigue cracks, as shown in Figure 50. Figure 51 is a plot of stress amplitude versus cycles to failure by fatigue. The drastic reduction in stress amplitude required for a given fatigue lifetime illustrates the ease in which fatigue cracks form in corrosion pits.
Figure 48. Picture of First Stage Impeller with Electroless Nickel Coating after 5 Years of Service. The Gas Path of the Impeller Has No Foulant Buildup. Erosion Erosion can be a significant issue in some applications of centrifugal compressors. It can occur as solid particle or liquid droplet erosion. Solid particle erosion is usually caused by external contaminants, such as those mentioned above. Liquid droplet erosion can be due to the condensation during compression or intentional injection of liquid for cleaning. The particle or droplet is accelerated by the carrier gas. The resultant impact upon the metallic compressor components removes small amounts of metal; creating pits and microcracks at the surface. Typically, the compressor design is robust and the material removal rate is such that erosion on its own is not a major problem. Problems arise when erosion is combined with other factors such as corrosion and cyclic stresses. In corrosive environments, erosion can compromise protection schemes and cause an accelerated corrosion attack. An example of this was mentioned above, where erosion can remove some antifoulant coatings. When cyclic stresses are present, the microcracks created can easily serve as initiation sites for fatigue cracking. Figure 49 shows the leading edge and fracture surface of a compressor impeller blade that failed by fatigue initiated by damage from liquid droplet erosion.
Figure 50. Micrograph Showing a Crack Initiating from a Corrosion Pit.
Figure 51. Plot Showing the Effect of Corrosion and Pitting on Fatigue Life. Similar to compressors, there are not many coatings applied to steam turbine components for corrosion only. If there are problems with corrosion, usually other problems are occurring, such as erosion, which also must be addressed by a coating. The authors company has used chromium diffusion coatings in the past where pitting corrosion was a problem. Erosion Erosion is a serious problem in steam turbines. The gas path of a steam turbine is much more closed than that found in a centrifugal compressor, providing more area for erosive media to impact.
Figure 49. Stereomicrograph Showing a Centrifugal Compressor Impeller Blade that Failed Due to Liquid Droplet Erosion. Erosion can often be thought of as a type of wear; therefore, similar coatings are used to prevent both. Coatings used for erosion protection are usually hard coatings, such as tungsten or chromium carbide. Any surface exposed to the gas path should be coated in applications where erosion is a problem. One problem that is an issue with all coatings for compressors, but is particularly true for
207
Erosion-corrosion mechanisms are more prevalent due to the higher temperatures. The chances of cyclic stresses are also very high due to the complicated stress state found in turbine blades. Solid Particle Erosion The erosion occurs on the blade-vane leading edges caused by the exfoliation of scales from the boiler tubes mainly during transient conditions. For rotating blades, sprayed Cr3C2 coatings are applied using processes such as detonation gun, plasma, and high velocity oxygen fuel processes to protect against the scales. Other processes such as diffused boride coatings have successfully been applied to stationary nozzles. Figure 52 shows a photomicrograph of boride coating on AISI 422. The boride diffusion coatings applied by pack cementation and the Cr3C2 coatings applied by the spray processes mentioned above will continue to be the industry best choices. Figure 54. Images Showing Erosion of Blades from the Same Turbine after 400 Hours and 70,000 Hours. While not strictly coatings, various countermeasures have been utilized by OEMs, including water drainage devices in a cylinder wall, flame or induction hardening of the leading edge of the blade, and applying stellite or tool steel strips to the leading edge of a blade by welding or brazing. All of the above methods show some degree of success, with the stellite material providing the better performance in the more stringent water droplet environment. Fouling Figure 52. Cross-Section of a Boride Diffusion Coating on AISI 422. Liquid Droplet Erosion The damage of rotor blades in the latter low pressure stages of steam turbines by condensed water droplets can be a problem that has troubled designers and operators for many years. The damage consists of removal of material from the leading edge and adjacent convex surfaces of the moving blade. It is related to the wetness conditions in the low pressure regions and the velocity with which the surface of the blade strikes the water droplets. Stresses produced by impact of drops have been calculated by existing theories to be sufficiently high to initiate damage in the latter stages of blades in a steam turbine. Turbine experience and laboratory testing have shown that erosion rate is time dependent with three successive zones: a primary zone in which damage is initiated at slip planes with little or no weight loss, a secondary zone where the rate rises to a maximum, and finally a tertiary zone where the rate diminishes to a steady-state value. Moreover, it is this tertiary region that is important to designer and operator alike rather than the initial and secondary zones since it is this region that the turbine erosion shields operate for most of their lives (Figures 53 and 54). Fouling and corrosion can also be a problem in steam turbines not only causing material damage but also can gradually reduce the efficiency of the turbine. Industrial turbines, whether condensing or noncondensing, can encounter problems with deposits building up on the turbine airfoils. In a turbine, hydroscopic salts, such as sodium hydroxide, can absorb moisture when superheated steam becomes saturated and condenses in the latter stages of the turbine/Wilson line. Wet sodium hydroxide has a tendency to adhere to turbine metal surfaces and can entrap other impurities such as silica, metal oxides, and phosphates. Once these deposits have formed they can be difficult to remove. Build up of these deposits may be a cause of decrease in efficiency and possibly an increase in vibration. A smooth clean steam path will not collect deposits so easily as a dirty, previously contaminated surface. Consequently, a previously contaminated turbine will accumulate deposits more rapidly than a clean one. Therefore, it is desirable to prevent further deposit buildup and to remove the problems associated with the presence of the deposits by cleaning the turbine. The authors company has provided support to end user turbines for water washing of steam turbines (Watson, et al., 1995). The effectiveness of the water removal procedures mainly depends on the adherence of the deposits to the substrate. A second route is to coat the surface with a material that has superior antifouling or antistick/corrosion characteristics. This in turn is beneficial to the turbine blades by reducing the tendency for contaminants to stick to the blades and increase the effectiveness of the water washing. Titanium nitride coatings with a chromium undercoat (Cr-TiN) have also been used by steam turbine OEMs to coat turbine blades. This coating provides corrosion protection and can be used on all stages of a steam turbine rotor, however, the Cr-TiN only provides limited antifoulant benefits. The authors company has recently developed a proprietary coating, which is a corrosion resistant antifoulant coating designed for the later stages of the turbine rotor (where the deposit buildup is most severe). Testing has shown that this coating provides significant improvement in foulant release ability (Figure 55), excellent corrosion protection (passes over 1000 hours of ASTM B117 corrosion testing under a 5 percent salt solution), and erosion protection (Figure 56), and, while having little effect on the fatigue properties (Figure 57).
Figure 53. Erosion Rate of 630 DPH, 18W-6Cr-0.7C Tool Steel Comparator Specimens.
208
ACKNOWLEDGEMENT
The authors are grateful for the support from the Materials Engineering Department at Elliott Company and recognize Elliott Company for permission to publish this paper.
REFERENCES
API RP 687, 2001, Rotor Repair, American Petroleum Institute, Washington, D.C. API Standard 617, 2002, Axial and Centrifugal Compressors and Expander-Compressors for Petroleum, Chemical and Gas Industry Services, Seventh Edition, American Petroleum Institute, Washington, D.C. Aerospace Structural Metals Handbook: Volume 2, 1988, Department of DefenseMaterials and Ceramics Information Center. Armor, A. F., Jaffee, R. I., and Hottenstine, R. D., 1984, Advanced Supercritical Power PlantsThe EPRI Development Program, Proceedings of the American Power Conference, 46, p. 70. Arlt, N., Gumpel, P., and Sahni, P., 1988, Uber Einige Eigenschaften von Nichtrostenden Nickelmartensitischen Chromstahlen, Thyssen Edeist, Techn. Ber. 14, Band Heft 1, p. 71. ASME Boiler & Pressure Vessel Code - Section VIII, 2007, American Society of Mechanical Engineers, New York, New York, pp. 66-67. Atkins, M., 1980, Atlas of Continuous Cooling Transformation Diagrams for Engineering Steels, Revised U.S. Edition, Copyright by the American Society for Metals, pp. 168 and 188. Batte, A. D. and Murphy, M. C., September 1981, The Corrosion Fatigue of 12%Cr Blade Steels in Low Pressure Steam Turbine Environments, Corrosion Fatigue of Steam Turbine Blade Materials, Workshop Proceedings, Palo Alto, California. Branch, G. D., et al., 1973, High Temperature Bolts for Steam Power Plant, International Conference on Creep and Fatigue in Elevated Temperature Applications, Sheffield and Philadelphia, Institute of Mechanical Engineers, London, Conference Publication 13, pp. 192.1-192.9. Briggs, J. Z. and Parker, T. D., 1965, The Super 12% Cr Steel, Climax Molybdenum Co., New York, New York. Bruscato, R. M., 1970, Temper Embrittlement and Creep Embrittlement of 2 Cr-1 Mo Shielded Metal-Arc Weld Deposits, Welding Journal, 35, p. 148s. Cameron, J. A., 1989, Materials for Centrifugal CompressorsA Progress Report, Proceedings of the Eighteenth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 9-22. Cameron, J. A., and Danowski, F. M., Jr., 1973, Some Metallurgical Considerations in Centrifugal Compressors, Proceedings of the Second Turbomachinery Symposium, Gas Turbine Laboratories, Texas A&M University, College Station, Texas, pp. 116-128. Dowson, P., 1995, Fracture Mechanics Methodology Applied to Rotating Components of Steam Turbine and Centrifugal Compressor Rotors, Third International Charles Parsons Conference, Tyne, United Kingdom. Dowson, P., 2002, Development of 13Cr4Ni Material for Brazed/Welded Impellers for H2S and Low Temperature Service, Supermartensitic Stainless Steels 2002, Brussels, Belgium. Dowson, P., 2007, Antifouling and Corrosion Resistant Coatings for Cracked Gas Compressors and Steam Turbines, AICHE Spring National Meeting, Houston, Texas. Dowson, P., and Wiegand, R., May 1996, Welded Rotor Restoration Technical Justification & Fit for Service Assurance, Second International EPRI Conference and Vendor Exposition.
Figure 55. Comparison of Foulant Release Performance of Bare Steel Against Proprietary Coating and Cr-TiN Coated Samples.
Figure 56. Comparison of Bare AISI 403 Stainless Steel Against Proprietary and Cr-TiN Coated Samples after 10 Hours of Modified ASTM G32 Testing.
Figure 57. Results of R.R. Moore Fatigue Testing.
SUMMARY
An overview of materials and material related processes has been presented for centrifugal compressors and steam turbines. Special attention has been given to address some of the problems associated with material selection for the various components and the steps taken to prevent and/or minimize reoccurrence. What will the future bring to materials/or materials related processes? The application of composite materials for rotating components in compressors may be seen in the not so distant future. The application of refined existing processes to manufacture components to near net shape such as P/M net shape, HIP process, or metal rapid prototyping based on laser microwelding of metallic powders may also be seen.
209
Dowson, P., Ross, S. L., and Schuster, C., 1991, The Investigation of Suitability of Abradable Seal Materials for Application in Centrifugal Compressors and Steam Turbines, Proceedings of the Twentieth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 77-90. Dowson, P., Walker, M., and Watson, A., 2004, Development of Abradable and Rub Tolerant Seal Materials for Application in Centrifugal Compressors and Steam Turbines, Proceedings of the Thirty-Third Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 95-102. Dowson, P., Wang, W., and Alija, A., 2005, Remaining Life Assessment of Steam Turbine and Hot Gas Expander Components, Proceedings of the Thirty-Fourth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 77-92. Engineered Materials Handbook: Composites, Volume 1, 1987, American Society of Metals, Materials Park, Ohio. Everson, H., Orr, J., and Dulieu, D., 1988, Low Alloy Ferritic Bolting Steels for Steam Turbine Applications: The Evolution of Durehete Steels, in Advances in Materials Technology for Fossil Power Plants, R. Viswanathan and R. I. Jaffee, Editors, American Society for Metals, Metals Park, Ohio. Fidler, F., 1971, Metallurgical Considerations in Wire Wool Type Wear Bearing Phenomena, Wear, 17, pp. 1-20. Gold, M. and Jaffee, R. I., 1984, Materials for Advanced Steam Cycles, ASM Journal of Materials for Energy Systems, 6 (2), pp. 130-145. Gross, J. H. and Stout, R. D., April 1957, Ductility and Energy Relations in Charpy Tests of Structural Steels, Welding Research Supplement, 22, pp. 151s-159s. Guinee, M. J. and Lamza, E. W., 1995, Cost Effective Methods to Maintain Gas Production by the Reduction of Fouling in Centrifugal Compressors, SPE30400, pp. 341-350. Jones, G. T., 1972, Proceedings Institute of Mechanical Engineers, 186, p. 31-32. Keller, H. F. and Cameron, J. A., 1974, Laboratory Evaluation of Susceptibility to Sulfide Cracking, Carrier Corporation, NACE Paper #99. Kohut, G. B. and McGuire, W. J., July 1968, Sulfide Stress Cracking Causes Failure of Compressor Components in Refinery Service, Materials Protection, 7, pp. 17-22. LaFave, R. A., 1991, Submerged Arc Weld Restoration of Steam Turbine Rotors Using Specialized Welding Techniques, Proceedings of the Twentieth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, pp. 19-34. Lukens Fineline Steels, Form No. 675 4-85., Lukens Steel Company, Coatesville, Pennsylvania. Meher-Homji, C. B., Focke, A. B., and Wooldridge, M. B., 1989, Fouling of Axial Flow CompressorsCauses, Effects, Detection, and Control, Proceedings of the Eighteenth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 55-76.
Metals Handbook, Ninth Edition: Volume 4, 1981, American Society of Metals, Materials Park, Ohio, p. 35 and p. 43. Moller, G. E., 1968, Corrosion, Metallurgical and Mechanical Experiences of Petroleum Refinery Compressors, NACE Task Group T-8-1 Interim Report. NACE Standard MR0103, 2005, Materials Resistant to Sulfide Stress Cracking in Corrosive Petroleum Refining Environments, NACE International, Houston, Texas. NACE Standard MR0175, 2003, Petroleum and Natural Gas IndustriesMaterials for Use in H2S-Containing Environments in Oil and Gas, NACE International, Houston, Texas. NACE Standard RP0472, 2005, Methods and Controls to Prevent In-Service Environmental Cracking of Carbon Steel Weldments in Corrosive Petroleum Refining Environments, NACE International, Houston, Texas. Newhouse, D. L., et al., 1972, Temper Embrittlement of Alloy Steels, ASTM STP 499, pp. 3-36. Newhouse, D. L., July 1987, Guide to 12Cr Steels for High-and Intermediate-Pressure Turbine Rotors for the Advanced Coal-Fired Steam Plant, Report CS-5277, Electric Power Research Institute, Palo Alto, California. Schwant, R. C. and Timo, D. P., 1985, Life Assessment of General Electric Large Steam Turbine Rotors, Life Assessment and Improvement of Turbogenerator Rotors for Fossil Plants, Viswanathan, R., Editor, New York, New York: Pergamon Press, p. 3-25-3-40. Speidel, M. O., September 1981, Corrosion-Fatigue of Steam Turbine Blade Materials, Corrosion Fatigue of Steam Turbine Blade Materials, Workshop Proceedings, Palo Alto, California. Treseder, R. S. and Swanson, T. M., 1968, Factors in Sulfide Corrosion Cracking of High Strength Steels, Corrosion, 24, pp. 31-37. Viswanathan, R. and Jaffee, R. I., October 1983, Toughness of CrMo-V Steels for Steam Turbine Rotors, ASME Journal of Engineering Material Techniques, 105, pp. 286-294. Wang, W., Dowson, P., and Baha, A., 2003, Development of Antifouling and Corrosion Resistant Coatings for Petrochemical Compressors, Proceedings of the Thirty-Second Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 91-97. Warren, D. and Beckman, G. W., October 1957, Sulphide Corrosion Cracking of High Strength Bolting Material, Corrosion, 13, pp. 631t-646t. Watanabe, J. and Murakami, Y., 1981, Prevention of Temper Embrittlement of CrMo Steel Vessels by the Use of Low Si Forged Steels, American Petroleum Institute, Chicago, Illinois, p. 216. Watson, A. P., Carter, D. R., and Alleyne, C. D., 1995, Cleaning Turbomachinery without Disassembly, Online and Offline, Proceedings of the Twenty-Fourth Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 117-128.
STEAM TURBINE CORROSION AND DEPOSITS PROBLEMS AND SOLUTIONS

by Otakar Jonas
Consultant
and Lee Machemer

Senior Engineer Jonas, Inc. Wilmington, Delaware
Otakar Jonas is a Consultant with Jonas, Inc., in Wilmington, Delaware. He works in the field of industrial and utility steam cycle corrosion, water and steam chemistry, reliability, and failure analysis. After periods of R&D at Lehigh University and engineering practice at Westinghouse Steam Turbine Division, Dr. Jonas started his company in 1983. The company is involved in troubleshooting, R&D (EPRI, GE, Alstom), failure analysis, and in the production of special instruments and sampling systems. Dr. Jonas has a Ph.D. degree (Power Engineering) from the Czech Technical University. He is a registered Professional Engineer in the States of Delaware and California. Lee Machemer is a Senior Engineer at Jonas, Inc., in Wilmington, Delaware. He has 13 years of experience with industrial and utility steam cycle corrosion, steam cycle and water chemistry, and failure analysis. Mr. Machemer received a B.S. degree (Chemical Engineering) from the University of Delaware and is a registered Professional Engineer in the State of Delaware.
longer blades) resulted in increased stresses and vibration problems and in the use of higher strength materials (Scegljajev, 1983; McCloskey, 2002; Sanders, 2001). Unacceptable failure rates of mostly blades and discs resulted in initiation of numerous projects to investigate the root causes of the problems (McCloskey, 2002; Sanders, 2001; Cotton, 1993; Jonas, 1977, 1985a, 1985c, 1987; EPRI, 1981, 1983, 1995, 1997d, 1998a, 2000a, 2000b, 2001, 2002b, 2002c; Jonas and Dooley, 1996, 1997; ASME, 1982, 1989; Speidel and Atrens, 1984; Atrens, et al., 1984). Some of these problems persist today. Cost of corrosion studies (EPRI, 2001a, Syrett, et al., 2002; Syrett and Gorman, 2003) and statistics (EPRI, 1985b, 1997d; NERC, 2002) determined that amelioration of turbine corrosion is urgently needed. Same problems exist in smaller industrial turbines and the same solutions apply (Scegljajev, 1983; McCloskey, 2002; Sanders, 2001; Cotton, 1993; Jonas, 1985a, 1987; EPRI, 1987a, 1998a; Jonas and Dooley, 1997). The corrosion mechanisms active in turbines (stress corrosion cracking, corrosion fatigue, pitting, flow-accelerated corrosion) are shown in Figure 1.
ABSTRACT
This tutorial paper discusses the basics of corrosion, steam and deposit chemistry, and turbine and steam cycle design and operationas they relate to steam turbine problems and problem solutions. Major steam turbine problems, such as stress corrosion cracking of rotors and discs, corrosion fatigue of blades, pitting, and flow accelerated corrosion are analyzed, and their root causes and solutions discussed. Also covered are: life prediction, inspection, and turbine monitoring. Case histories are described for utility and industrial turbines, with descriptions of root causes and engineering solutions.
Figure 1. Corrosion Mechanisms Active in Steam Turbines. Purpose, Design, and Operation of Steam Turbines The steam turbine is the simplest and most efficient engine for converting large amounts of heat energy into mechanical work. As the steam expands, it acquires high velocity and exerts force on the turbine blades. Turbines range in size from a few kilowatts for one stage units to 1300 MW for multiple-stage multiple-component units comprising high-pressure, intermediate-pressure, and up to three low-pressure turbines. For mechanical drives, single- and double-stage turbines are generally used. Most larger modern turbines are multiple-stage axial flow units. Figure 2 shows a typical tandem-compound turbine with a combined high pressure (HP), intermediate pressure (IP) turbine, and a two-flow low-pressure (LP) turbine. Table 1 (EPRI, 1998a) provides alternate terminology for several turbine components.
211
INTRODUCTION
This tutorial paper discusses steam turbine corrosion and deposition problems, their root causes, and solutions. It also reviews design and operation, materials, and steam and deposit chemistry. References are provided at the end of the paper. With an increase of generating capacity and pressure of individual utility units in the 1960s and 70s, the importance of large steam turbine reliability and efficiency increased. The associated turbine size increase and design changes (i.e., larger rotors and discs and
212
treatment practices coming fast in the last 25 years, experience was short and limited, and problems developed, which need to be corrected and considered in new designs and redesigns. While the turbine seems to be a simple machine, its design, including design against corrosion, is complex. There are five areas of design that affect turbine corrosion:
Mechanical
Figure 2. Typical Tandem Compound, Single Reheat, Condensing Turbine. (Courtesy of EPRI, 1998a) Table 1. Alternate Terminology for Turbine Components.
design (stresses, vibration, stress concentrations, stress intensity factor, frictional damping, benefits of overspeed and heater box testing)
Physical shape (stress concentration, crevices, obstacles to flow,

surface finish, crevices)
Material selection (maximum yield strength, corrosion properties,

material damping, galvanic effects, etc.)
Flow and thermodynamics (flow excitation of blades, incidence

angle, boundary layer, condensation and moisture, velocity, location of the salt zone, stagnation temperature, interaction of shock wave with condensation)
Heat
transfer (surface temperature, evaporation of moisture, expansion versus stress, heated crevices) Recognition of these effects led to a formulation of rudimentary design rules. While the mechanical design is well advanced and the material behavior is understood, the flow excitation of blades and the effects of flow and heat transfer on chemical impurities at surfaces are not fully included in design practices. Selection of some combinations of these design parameters can lead to undesirable stresses and impurity concentrations that stimulate corrosion. In addition, some combinations of dissimilar materials in contact can produce galvanic corrosion. Design and material improvements and considerations that reduce turbine corrosion include:
Welded
rotors, large integral rotors, and discs without keywayseliminates high stresses in disc keyways.
Replacement
Steam enters from the main steam lines through stop and control valves into the HP section. The first (control) stage is spaced slightly apart from subsequent stages to allow for stabilization of the flow. After passing through the HP turbine, cold reheat piping carries the steam to the reheater (if present) and returns in the hot reheat piping to the integrated HP and IP cylinder to pass through the IP turbine section. The flow exits the IP turbine through the IP exhaust hood and then passes through crossover piping to the LP turbine and exits to the condenser through the LP exhaust. The typical modern steam turbine has a number of extraction points throughout all sections where the steam is used to supply heat to the feedwater heaters. During its expansion through the LP turbine, the steam crosses the saturation line. The region where condensation begins, termed the phase transition zone (PTZ) or Wilson line (Cotton, 1993; EPRI, 1997c, 1998a, 2001b), is the location where corrosion damage has been observed. In single reheat turbines at full load, this zone is usually at the L-1 stage, which is also in the transonic flow region where, at the sonic velocity (Mach = 1), sonic shock waves can be a source of blade excitation and cyclic stresses causing fatigue or corrosion fatigue (EPRI, 1997c; Jonas, 1994, 1997; Stastny, et al., 1997; Petr, et al., 1997). Design Because of their long design life, steam turbines go through limited prototype testing where the long-term effects of material degradation, such as corrosion, creep, and low-cycle fatigue, cannot be fully simulated. In the past, when development was slow, relatively long-term experience was transferred into new products. With new turbine types, larger sizes, new power cycles, and water
of higher strength NiCrMoV discs with lower (yield strength < 130 ksi (896 MPa) strength discs.
Repair welding of discs and rotors; also with 12%Cr stainless

steel weld metal.
Mixed tuned blade rows to reduce random excitation. Freestanding and integrally shrouded LP blades without tenon
crevices and with lower stresses. except for NaOH. SCC and CF.
Titanium LP bladescorrosion resistant in turbine environments Lower stress and stress concentrationsincreasing resistance to Flow path design using computerized flow dynamics and viscous Curved (banana) stationary blades that reduce nozzle passing
excitation. flowlower flow induced vibration, which reduces susceptibility to CF.
New materials for blade pins and boltingresistant against SCC. Flow guides and double-ply expansion bellowsreduces
impurity concentration, better SCC resistance.
Moisture
extraction to improve efficiency and reduce flowaccelerated corrosion (FAC) and water droplet erosion and use of alloy steels to reduce FAC. LP Rotor and Discs There are three types of construction in use for LP rotors:
Built-up (shrunk-on design) with forged shaft onto which discs

are shrunk and keyed,
213
Machined from one solid piece (most common), or The discs welded together to form the rotor (Figure 3).
The rotor and disc construction is governed by the practices of individual manufacturers, capabilities of steel mills, cost, and, during the last few decades, by their resistance to SCC. The solid and welded rotors do not have a problem with disc bore SCC. The three types, shown in Figure 3, have little effect on the SCC and CF susceptibilities of the blade attachments.
Blades are connected at the root to the rotor discs by several configurations (Figure 5). There are several types of serrated attachments: the fir tree configuration, which is inserted into individual axial slots in the disc; and the T-shape, which is inserted into a continuous circumferential slot in the disc. The finger type attachment is fitted into circumferential slots in the disc and secured by axially inserted pins. All of the blade root designs have geometries that result in higher local stresses at radii and stress concentrations that promote SCC and CF. The goal of the design should be to minimize these local tensile stresses.
Figure 3. Three Types of Rotor Construction. (Courtesy of EPRI, 1998a) LP Blades Blade and blade path design and material selection influence blade CF, SCC, pitting, and other forms of damage in many ways (Sanders, 2001; EPRI, 1981, 1997c, 1998a, 2001b; BLADE-ST , 2000). The main effects of the blade design on corrosion, corrosion fatigue strength, stress corrosion cracking susceptibility, and pitting resistance include:
Figure 5. Types of Blade Root Attachments. (Courtesy of EPRI, 1998a) The airfoil of blades may be of constant cross-section for short blades, and twisted for longer ones. The longest blades for the last few rows of the LP are twisted to match the aerodynamics at different radii and improve aerodynamic efficiency. The longer blades are usually connected at the point of highest vibration amplitude to each other by a tie or lashing wire, which reduces vibration of the airfoil. To reduce random excitation, mixed tuning of long rotating blades has been used in which adjacent blades have different resonant frequencies (EPRI, 1998a). Stationary blades in LP stages are typically arranged in diaphragms of cast or welded construction. In wet stages, diaphragms may be made with hollow blade vanes or other design features as a means of drawing off moisture that would otherwise lead to liquid droplet erosion (EPRI, 2001b). Recently, some stationary blade designs have also been leaned or bowed, improving flow and efficiency and lowering the excitation forces on the downstream rotating blades. Casings Turbine casings must contain the steam pressure and maintain support and alignment for the internal stationary components. They are designed to withstand temperatures and pressures up to the maximum steam conditions. Their design has evolved over the years and casings are now multiple pressure vessels (for example, an inner and outer casing in the HP and IP cylinder, or a triple casing) allowing smaller pressure drops and thinner wall thickness. These thinner cross-sections allow for a lower temperature gradient across the casing section and thus lower thermal stresses. The LP casing may also be a multiple part design with the inner casing containing the supports for the diaphragms and the outer casing directing the exhaust to the condensers. Design Recommendations for Corrosion Control New designs, redesigns, and failed components should be checked to determine if they meet allowable corrosion-related design specifications and other corrosion related requirements (Jonas, 1985c).
Vibratory stresses and their frequencies. Maximum service steady stresses and stress concentrations. Flow induced vibration and deposition. Mechanical, frictional, and aerodynamic damping.
Rotating LP turbine blades may be free standing (not connected to each other), connected in groups, or all blades in the row may be continuously connected by a shroud. Connections made at the blade tip are termed shrouds or shrouding. Shrouds may be inserted over tenons protruding above the blade tips and these tenons then riveted down to secure the shrouds, or they may consist of integrally forged or machined stubs, which, during operation, provide frictional damping of vibration because they touch (Figure 4). This design also eliminates the tenon-shroud crevices where corrosive impurities could concentrate. In some cases, long 180 degree shrouds are used or smaller shroud segments are welded together.
Figure 4. Typical Turbine Shrouds. (Courtesy of EPRI, 1998a)
214
In new designs and redesigns of turbine components, use should be made of the new design tools, including 3D finite element stress and vibration analysis and 3D viscous flow analysis, and consideration of condensation and impurity behavior. Blade resonance frequencies should be verified by telemetry. To ensure a corrosion-free design, a corrosion engineer and a chemist should be consulted during the design activity. The following should be considered in design of steam turbines: avoidance of SCC and CF should include an evaluation of the material corrosion properties and defects that influence susceptibility to SCC and CF, i.e., threshold stress (SCC), threshold stress intensity (KISCC and KthCF), crack growth rate ((da/dt)SCC and (da/dN)CF), corrosion fatigue limit, pitting rate, and pit depth limit. True residual stresses (micro and macro) should also be considered. Because SCC and CF initiate at surfaces, the maximum surface stresses must be controlled, usually by control of the elastic stress concentration factor, kt. The stresses should be the lowest in the salt zone region where corrosion is most likely.
hydrochloric, sulfuric, carbonic, and organic acids can form (Jonas, 1982). It is recommended that their composition and use be carefully controlled to minimize the risk of residual contamination and subsequent corrosion. For the compounds that can remain on turbine surfaces during operation, chlorine and sulfur levels should be restricted to low ppm levels in that compound. Of specific concern are: MoS2 (Molylube) (Turner, 1974; Newman, 1974), Loctite, thread compounds (Cu, Ni, graphite), and chlorinated solvents. Materials and Corrosion Data There is little variation in the materials used for blades, discs, rotors, and turbine cylinders, and only a few major changes have been introduced in the last decade. Titanium alloy blades are being slowly introduced for the last LP stages, and better melting practices to provide control of inclusions and trace elements are being evaluated for discs and rotors. Table 2 (Jonas, 1985a) lists common materials and the typical corrosion mechanisms for the various turbine components. Table 2. LP Turbine Components, Materials, and Related Corrosion Mechanisms.
Stresses (Jonas, 1985c)The mechanical design concepts for
Vibratory
stresses are rarely accurately known, except when telemetry on operating turbines is performed. The design approach should be to minimize flow excitation, tune the blades, provide maximum damping, and perform laboratory and shop stationary frequency testing. Heater box overspeed and overspeed testing during operation are generally beneficial in reducing operating stresses by local plastic deformation.
Heat
transfer and flow (EPRI, 1997c)Surface temperature resulting from heat transfer and flow stagnation should be considered along with its effect on thermodynamic conditions of the impurities and water film at surfaces (i.e., evaporation of moisture). Flow effects on blade vibration and deposit formation are complex, and there are over 15 flow blade excitation mechanisms to be considered.
Flow of moistureTo avoid flow-accelerated corrosion (EPRI,

1996; Kleitz, 1994; Jonas, 1985b; Svoboda and Faber, 1984) and water droplet erosion (Ryzenkov, 2000; Pryakhin, et al., 1984; Rezinskikh, et al., 1993; Sakamoto, et al., 1992; Povarov, et al., 1985; Heyman, 1970, 1979, 1992), the flow velocity of wet steam should not exceed the allowable velocity specific to the materials and moisture chemistry. Regions of high turbulence should be avoided or higher chromium steels should be used. Blade path moisture can be extracted.
CrevicesCrevices can act as impurity traps and concentrators,
facilitate formation of oxygen concentration cells, and may generate high stresses by the oxide growth mechanism. The worst crevices are those with corrosive impurities and metal temperature within the salt zone. Some disc bore and keyway and blade tenon-shroud crevices fall into this category.
Galvanic effectsWhen dissimilar materials are coupled together, corrosion of both materials can be affected by the associated shift in corrosion potentials into the stress corrosion cracking (SCC) or pitting regions. The more active of the two materials may suffer galvanic corrosion. In addition, in some environments, the potential shift could be into the region where one of the coupled materials is susceptible to stress corrosion cracking or pitting. InspectabilityIn designing turbine components, the question Chemical
of inspectability should be addressed. In particular, crevice and high stress regions should be reachable using available inspection techniques. compounds used during machining, cleaning, nondestructive testing (NDT), and other activitiesMany different chemical compounds are used during manufacture, storage, erection, and inspection of turbine components. Some of them contain chlorine and sulfur as impurities or as a part of the organic matrix. During thermal decomposition of the residues of these compounds,
LP rotors are typically constructed of forgings conforming to ASTM A293 (Class 2 to 5) or ASTM A470 (Class 2 to 7), particularly 3.5NiCrMoV . Shrunk-on discs, when used, are made from forgings of similar NiCrMoV materials conforming to ASTM A294 (Grade B or C), or ASTM 471 (Classes 1 to 3). The strength and hardness of turbine components must be limited because the stronger and harder materials become very susceptible to SCC and CF (EPRI, 1998a); particularly turbine rotors, discs, and blades cannot be made from high strength materials. The crack propagation rate increases exponentially with yield strength at high yield strength values and SCC starts being influenced by hydrogen embrittlement. Because of this sensitivity to high yield strength, practically all turbine discs, fossil and nuclear, with yield strength higher than ~140 ksi (965 MPa) have been replaced with lower strength materials. Figure 6 is a correlation of crack propagation rates versus yield strength for several operating temperatures (Clark, et al., 1981). This type of data has been used to predict the remaining life and safe inspection interval. There is also an upper temperature limit for LP rotor and disc steels, ~650F (345C) aimed at avoiding temper embrittlement (EPRI, 1998a).
Figure 6. Average Crack Growth Rates Versus Yield Strength for Several Operating Temperatures for NiCrMoV Disc Steel. (Courtesy of Clark, et al., 1981)
215
Since the 1930s, most LP turbine blades have been manufactured from a 12%Cr stainless steeltypically types AISI 403, 403-Cb, 410, 410-Cb, and 422, depending on the strength required. Types 403 and 410 have better SCC and CF resistance than Type 422, an important characteristic for use in the wet stages of the LP turbine. There are numerous specifically customized versions of these generic materials (Carpenter H-46, Jethete M152 (modified 403), etc.). The precipitation hardened stainless steels designated 17-4 PH, 15-5PH, and PH13-8Mo have been used for some fossil and nuclear LP turbine blades. The composition of 17-4PH is 17 percent Cr, 4 percent Ni, and 4 percent Cu. These steels may be difficult to weld and require post-weld heat treatment. The copper rich zones in the copper bearing stainless steels are often subject to selective dissolution, forming pits filled with corrosion products. These pits can be crack initiation sites. Titanium alloys, primarily Ti-6Al-4V , have been used for turbine blades since the early 1960s (EPRI, 1984d, 1984e, 1985c). There are numerous benefits to using titanium alloy blades, including the ability to use longer lasting stage blades, favorable mechanical properties in applications involving high stresses at low temperatures, excellent corrosion resistance, and resistance to impact and water droplet erosion damage. Drawbacks to titanium include higher cost, difficult machining, and low material damping. LP turbine casings are typically constructed of welded and cast components. Materials acceptable for lower temperatures, such as carbon steel plate, are used. Considering the typical steam turbine design life of 25 to 40 years and the relatively high stresses, these materials have been performing remarkably well. Turbine steels are susceptible to SCC and CF in environments such as caustic, chlorides, acids, hydrogen, carbonate-bicarbonate, carbonate-CO2, and, at higher stresses and strength levels, in pure water and steam. Corrosion Data Corrosion data should provide allowable steady and vibratory stresses and stress intensities for defined design life or inspection intervals. It is suggested that SCC data include threshold stress (SCC), threshold stress intensity (KISCC), crack growth rate (da/dt), and crack incubation and initiation times. Corrosion fatigue data should include fatigue limits for smooth and notched surfaces and proper stress ratios, crack growth data, and corrosion fatigue threshold stress intensities. Examples of the type of data needed are shown in Figures 7 to 9 for the NiCrMoV disc material and in Figure 10 for 12%Cr blade steel. Properly heat treated 12%Cr blade steel (yield strength 85 ksi, 600 MPa) is not susceptible to stress corrosion cracking and stress corrosion data are not needed. The data shown in Figures 7 to 10 can be used in turbine disc and blade design in which the allowable stresses and stress intensities should be below the threshold values for SCC and CF. The use of these data is outlined in (Jonas, 1985c).
Figure 8. Corrosion Fatigue (Goodman) Diagram for NiCrMoV Disc Steel; Tested to 108 Cycles. (Courtesy of Haas, 1977)
Figure 9. Air Fatigue Strength Reduction of NiCrMoV Disc Steel Caused by Pitting (Courtesy of McIntyre, 1979)Effects of Pit Density Were Not Investigated.
Figure 7. Stress Corrosion Behavior of NiCrMoV Disc Steel Versus Yield Strength for Good Water and Steam (Compiled from Published Data); KISCCThreshold Stress Intensity, SCC Threshold Stress, and da/dtqStage 2 Crack Growth Rate. (Courtesy of Jonas, 1985a)
Figure 10. Corrosion Fatigue (Goodman) Diagram for Three Stainless Steel Blading Alloys. (Courtesy of Atrens, et al., 1984)
216
The needed data are difficult to obtain because of the long time needed for testing and because of the great range of possible service environments and temperatures. There does not seem to be any universally accepted accelerated test or environment. The KISCC test in hydrogen sulfide gives a reasonable approximation of KISCC for low alloy steels, and ultrasonic fatigue tests give usable data to a high number of cycles that may be usable for turbine design. As a rule of thumb, the elasticity limit at temperature (usually between 0.4 and 0.6 of the 0.2 percent yield strength) can be used as a good estimate of the SCC threshold stress for low and medium strength materials in mildly corrosive environments. This is consistent with the oxide film rupture or strain induced cracking theory of stress corrosion cracking. Shot peening (EPRI, 2001b)has been used as a means of reducing high surface tensile stresses. At a sufficiently high shot peening intensity, a surface layer of residual compressive stresses is produced. One turbine vendor uses shot peening and other surface treatments extensively and they have almost no SCC of discs and corrosion fatigue (CF) of turbine blades. There is a concern that in corrosive environments, pits can grow through the compressive stress layer into the subsurface region with much higher tensile stresses. Other surface treatments for protection against corrosion, such as coatings and electroplating, have been evaluated (EPRI, 1987a, 1993a, 2001b; Jonas, 1989) and sometimes used. There are now several suppliers of steam turbine coatings (EPRI, 1987a, 1993a; Jonas, 1989). EnvironmentStressMaterial Turbine stress corrosion cracking and high- and low-cycle corrosion fatigue mechanisms are typically governed by a combination of environmental effects (steam chemistry, temperature, etc.), steady and vibratory stresses, and material strength, composition, and defects (Figure 11). It should be noted that even pure water and wet steam can cause cracking of turbine materials, particularly in the low alloy rotor and disc steels, and that medium and high strength materials are very susceptible to environmentally induced cracking in any environment, including pure water and steam.
Figure 12. Physical-Chemical Processes in LP Turbines.
Figure 13. Cross Section of an LP Turbine with the Locations of the Processes Listed in Figure 15-11. Steam Chemistry Steam chemistry or purity, together with the thermodynamics and flow design, determines corrosiveness of the deposits and liquid films on turbine component surfaces (Jonas, 1982, 1985a, 1985d; Jonas and Dooley, 1996, 1997; EPRI, 1984b, 1994a, 1997b, 1997c, 1999; Jonas, et al., 1993; Jonas and Syrett, 1987; Schleithoff, 1984). In fossil units, the LP turbine requires the lowest concentration of impurities in the cycle, that is, low parts per billion concentrations (1 ppb is 1 g/liter). Steam purity is controlled by the purity of makeup, condensate, and feedwater and in drum boilers by boiler water chemistry, boiler pressure, and carryover. As a minimum, steam purity should be monitored by isokinetic sampling and by analysis of sodium and cation conductivity (EPRI, 1986, 1994c, 1998b, 1998c, 2002a; Jonas, 2000). The corrosiveness of the steam turbine environment is caused by one or more of the following:
Figure 11. Three Components of Turbine Stress Corrosion and Corrosion Fatigue Cracking in Turbines. Turbine environment plays a major role in corrosion during operation and layup. The uniqueness of this environment is caused by the phase changes of the working fluid and the impurities carried by the steam (steam, moisture, liquid films, and deposits). Within the steam flow path and on the turbine component surfaces, the parameters controlling corrosion, such as pH, concentration of salts and hydroxides, and temperature, can change within a broad range. Even though steam impurity concentrations are controlled in the low parts per billion (ppb) range, these impurities can concentrate by precipitation, deposition, and by evaporation of moisture to percent concentrations, becoming very corrosive (Figures 12 and 13) (EPRI, 1994a, 1997b, 1997c, 1999; Jonas, et al., 1993).
Concentration of impurities from low ppb levels in steam to Insufficient High
percent levels in steam condensates (and other deposits) resulting in the formation of concentrated aqueous solutions pH control and buffering of impurities by water treatment additives such as ammonia velocity and high turbulence flow of low-pH moisture droplets (FAC) The situation is generalized in Figure 14, which is a Mollier diagram showing the LP turbine steam expansion line and thermodynamic regions of impurity concentrations (NaOH, salts, etc.) and resulting corrosion. Low volatility impurities in the salt zone are present as concentrated aqueous solutions. The NaCl concentration can be as high as 28 percent. Note that the conditions at the hot turbine surfaces (in relation to the steam saturation temperature)
217
can shift from the wet steam region into the salt zone and above. This can be the reason why disc stress corrosion cracking often occurs in the wet steam regions. The surfaces may be hot because of heat transfer through the metal or because of the stagnation temperature effect (zero flow velocity at the surface and change of kinetic energy of steam into heat).
Figure 14. Mollier Diagram with LP Steam Expansion Line and Thermodynamic Regions for Impurity Concentration and Corrosion Mechanisms. The impurity concentration mechanisms include:
Precipitation from superheated steam and deposition. Evaporation and drying of moisture on hot surfaces. Concentration on oxides by sorption. Nonhomogeneous nucleation of concentrated droplets and crystals
on surfaces. Dissolved impurities deposit from superheated steam when their concentration exceeds their solubilities, which sharply decreases as the steam expands. Depending on their vapor pressure, they can be present as a dry salt or an aqueous solution. In the wet steam region, they are either diluted by moisture or could concentrate by evaporation on hot surfaces. The region of passivity for iron and low alloy and carbon steels is narrow, falling within the pH range of 6 to 10. Since pH control in a power plant is mostly for the protection of the preboiler cycle and the boiler, it often does not match the needs of the turbine surfaces. The pH in the turbine depends on temperature, mechanical and vaporous carryover of impurities, and water treatment chemicals from the boiler and their volatility (distribution between the vapor and the surface film). When hydrochloric acid forms in cycles with ammonia all-volatile treatment (by decomposition of chlorinated organics, Cl leakage from polishers, or seawater or other cooling water leakage in the condenser), ammonium chloride forms in the water, and the acid may be neutralized. However, because of its volatility, ammonium chloride is transported with steam into the turbine where it hydrolyzes, forming NH3 gas and HCl. Deposits on turbine surfaces in units with sodium phosphate boiler water treatment (most drum boilers) are less corrosive (Jonas and Syrett, 1987; EPRI, 1984b; Jonas, 1985d). Sodium phosphate is a better neutralizing agent through the cycle; fewer acids are transported into the turbine and phosphate frequently codeposits with harmful impurities, providing in-situ neutralization and passivation. This is most likely the reason for lower frequency of turbine corrosion in systems with phosphate water treatments. Measurements of pitting potential of disc and blading alloys confirm the beneficial effects of sodium phosphate in the presence of NaCl. Besides the corrosion during operation, turbines can corrode during manufacture (corrosive products from machining fluids, exposure to tool tip temperatures), storage (airborne corrosive impurities, preservatives containing Cl and S), erection (airborne
impurities, preservatives, cleaning fluids), chemical cleaning (storage of acid in hotwells), nondestructive testing (chlorinated cleaning and NDT fluids), and layup (deposits plus humid air). Many of these corrosive substances may contain high concentrations of sulfur and chloride that could form acids upon decomposition. Decomposition of typical organics, such as carbon tetrachloride occurs at about 300F (150C). The composition of all of the above compounds should be controlled (maximum of 50 to 100 ppm S and Cl each has been recommended), and most of them should be removed before operation. Molybdenum disulfide, MoS2, has been implicated as a corrodent in power system applications (Turner, 1974; Newman, 1974). It can cause stress corrosion cracking of superalloys and steels by producing an acidic environment. Its oxidation products form low pH solutions of molybdic acid and even ammonium molybdate, which can form during operation, causing rapid attack of turbine steels. MoS2 has been used as a thread lubricant and in the process of disc-rotor assembling when the discs are preheated and shrunk on the rotors. Analysis of disc bore and keyway surfaces often reveals the presence of molybdenum and sulfur. In steam, MoS2 reduces the notch strength of disc steel, to about 30 percent of its strength in air. It has also been implicated in bolt and rotor shaft failures. Layup corrosion of unprotected turbines increases rapidly when the relative humidity of air reaches about 60 percent. When salt deposits are present, pitting during unprotected layup is rapid. Pit growth in turbine blade and rotor alloys in chloride-metal oxide mixtures in wet air is about as fast as in a boiling deaerated 28 percent NaCl solution. Turbine layup protection by clean dry air is recommended. Progress in controlling turbine corrosion through better control of the steam chemistry includes (Jonas, 1982, 1985d, 1994; EPRI, 1984b, 1986, 1994a, 1994c, 1997b, 1997c, 1998b, 1998c, 1999, 2002a; Jonas, et al., 1993, 2000; Jonas and Syrett, 1987; Schleithoff, 1984, Progress in..., 1981):
Decreasing concentration of corrosive impurities in makeup and

feedwater, lower air inleakage and condenser leakage, etc. excellent feedwater chemistry and clean boilers.
Oxygenated water treatment for once-through fossil units for Layup protection. Turbine washing after
impurities. chemical upsets to remove deposited
Reduction
or elimination of copper and its oxides and their synergistic corrosion effects by reducing oxygen concentration, operating with a reducing (negative oxidation-reduction potential [ORP]) environment and a low ammonia concentration, or by replacing copper alloys with steel or titanium.
PROBLEMS, THEIR ROOT CAUSES, AND SOLUTIONS

Steam turbine corrosion damage, particularly of blades and discs, has long been recognized as a leading cause of reduced availability (Scegljajev, 1983; McCloskey, 2002; Sanders, 2001; Cotton, 1993; Jonas, 1985a, 1987; EPRI, 1981, 1998a, 2001b; Jonas and Dooley, 1996, 1997; NERC, 2002). It has been estimated that turbine corrosion problems cost the U.S. utility industry as much as one billion dollars per year (EPRI, 1985b, 2001a; Syrett, et al., 2002; Syrett and Gorman, 2003; Jonas, 1986) and that the cost for industrial turbines, which suffer similar problems, is even higher. In this section, the main corrosion problems found in LP turbines and their root causes are summarized, and solutions to reduce or eliminate each problem are discussed. The field monitoring equipment shown in Figure 15 can be used to diagnose and prevent many common LP turbine corrosion and deposition problems (EPRI, 1997c, 2001b; Jonas, 1994; Jonas, et al., 2007). In addition, there are also monitors available to detect vibration, blade and rotor cracking, steam leaks, air inleakage, rotor position, and wear of bearings (Jonas, et al., 2007).
218
the next planned inspection or overhaul? If not, what is the safe inspection interval? If repeated failures are likely and repair would take a long time and lead to a large loss of production, spare rotors may be a good economical solution. Data collected by the North American Electric Reliability Council (NERC) for 1476 fossil units between 1996 and 2000 shows that LP turbines were responsible for 818 forced and scheduled outages and deratings, causing the utilities a 39,574 GWh production loss. The outages are often characterized as low frequency high impact events. (NERC, 2002) shows the components that were responsible for the failures as well as the MWh losses associated with the failures. Many of these outages are caused by corrosion (except for bearings). Figure 15. LP Turbine Troubleshooting Instrumentation Can Identify Specific Corrosive Conditions. (Courtesy of EPRI, 1997c) An overview of the low-pressure turbine corrosion problems together with erosion and other problems is given in Table 3. The problems are listed according to their priority and impact with disc rim blade attachment stress corrosion cracking being the highest impact problem today because of the long time required for weld repair or procurement of a new disc or new rotor (up to six months). Cracking of discs, corrosion fatigue of the rotor shaft, and fatigue or corrosion fatigue of long blades can become a safety issue because they can lead to perforation of the casing and other destructive events (EPRI, 1981, 1982a, 1998a; Jonas, 1977; Turner, 1974). It is estimated that inadequate mechanical design (high steady and vibratory stresses, stress concentration, and vibration) is responsible for about 50 percent of the problems, inadequate steam chemistry for about 20 percent, and nonoptimum flow and thermodynamic design for about 20 percent. Poor manufacturing and maintenance practices account for the remaining 10 percent of the problems. Table 3. LP Turbine Corrosion, Erosion, and Deposition Problems. Table 4. Forced and Schedule Outages and Deratings Caused by LP Turbine Components for the Years 1996 to 2000 (1476 Units, 168 Utilities).
Life Prediction and Inspection Interval Experience shows that pits and ground-out stress corrosion cracks can remain in-service for several years, depending on stress and environment. However, components containing high-cycle corrosion fatigue cracks should not be left in-service. Procedures for prediction of residual life and determination of a safe inspection interval have been developed for all major failure mechanisms including SCC, CF, fatigue, FAC, and creep. The procedures for SCC of turbine discs (Clark, et al., 1981; EPRI, 1989; Rosario, et al., 2002), low cycle corrosion fatigue, and FAC (EPRI, 1996) have been successfully applied because all variables influencing these mechanisms can be reasonably predicted or measured. However, life prediction for high cycle corrosion fatigue and fatigue has not been so successful because the vibratory stresses and the corrosiveness of the environment are usually not accurately known. Life prediction is based on results of inspection, fracture mechanics analysis of components with defects, and application of SCC and CF crack growth data. Time or number of load cycles to reach ductile or brittle fracture is predicted and a safety factor is applied to determine the time for the next inspection. In the procedure used by OEMs and Nuclear Regulatory Commission (NRC) for nuclear turbines for determining the inspection interval for turbine discs under SCC conditions, the safety factor of two was applied to the predicted time-to-failure. Stress Corrosion Cracking of Discs Stress corrosion cracking of LP turbine disc keyways and blade attachments have been the two most expensive generic problems in large steam turbines (Cheruvu and Seth, 1993; EPRI, 1982a, 1982b, 1984a, 1984c, 1985a, 1985d, 1987b, 1989, 1991a, 1997a, 1998d; Jonas, 1978; Speidel and Bertilsson, 1984; Clark, et al., 1981; Rosario, et al., 2002; Nowak, 1997; Kilroy, et al., 1997; Amos, et al., 1997; Turner, 1974; Newman, 1974; Parkins, 1972; Holdsworth, 2002). The keyway cracking problem has been resolved by redesigns of the shrunk-on or bolted-on discs, material replacement with lower strength material, and by elimination of the corrosive disc bore lubricants, based on molybdenum disulfide (MoS2), used in assembling the rotor. SCC of blade attachments of various designs is still a problem (EPRI, 1997d, 1998a, 2002c; Nowak, 1997).
When a corrosion problem is discovered during inspection or by equipment malfunction, the failure mechanism and the root causes are not always known. Even when the damage fits a description of a well-known problem (disc or blade cracking), replacement parts may not be readily available and the decision for what to do has to be made quickly. The main objectives in handling identified and potential problems are maintaining safety and avoiding forced outages. The questions should be asked: can we operate safely until
219
There are several corrosion damage mechanisms and many factors affecting discs (Figure 16). Typical locations and orientations of SCC cracks in LP discs are shown in Figures 17 and 18. There has also been SCC in pressure balance holes. Most incidents of disc rim attachment cracking have been found in nuclear units, however, there have also been problems in fossil units. In an independent survey, 13 of 38 (35 percent) boiling water reactors (BWRs) and 28 of 72 (39 percent) pressurized water reactors (PWRs) reported disc rim attachment cracking while 29 of 110 (26 percent) supercritical fossil units and only 20 of 647 (3 percent) subcritical units reported cracking (EPRI, 1997d). Disc rim blade attachment cracking occurred in multiple-hook (steeple), fir tree attachment designs (typically occurs in the corners of the hooks), in straddle mount dovetail and pinned-finger attachments, and in T-roots. It is always associated with stress concentrations.
MaterialsAll low alloy steels used for LP turbine rotors and discs are susceptible to SCC and CF in numerous turbine environments including pure water and wet steam. The strongest material factor influencing SCC is yield strength. At higher yield strength, SCC crack growth rate can be several orders of magnitude higher than for lower yield strength materials. The purity of the material and the steel melting practices mostly influence the fracture toughness, which determines the maximum tolerable crack size before a disc brittle fracture and burst. FractographySCC cracks of low alloy steels often initiate from pits and propagate intergranularly with branching. The initial part of the crack can be corroded and filled with magnetite. There could be beach marks or stretch marks, caused by overloading the crack during overspeed testing. Another type of beach mark can be caused by changes of the environment or by fatigue. Depending on the ratio of the steady and cyclic stresses, the disc cracks can be a mixture of intergranular and transgranular cracking. Figure 19 shows an SCC crack initiating at the radius of the upper serration of L-1 blade steeple. The intergranular crack initiated from a pit.
Figure 16. Crack Locations in Turbine Discs with Probable Impurity Concentration and Corrosion Mechanisms and Corrodents. (Courtesy of Jonas, 1985a)
Figure 19. SCC Crack in the Upper Steeple Serration-L-1 Blade Attachment. The factors that determine the SCC crack initiation time and propagation rate include material yield strength, surface stress, temperature, and the local chemical environment. Some of these relationships are shown in Figures 6 and 7. At yield strengths above ~135 ksi (930 MPa), these low-alloy steels show high SCC growth rates. Pitting often initiates SCC. When corrosive deposits are present, pitting during unprotected layup can be faster than pitting during operation. This is because during the layup, there can be 100 percent relative humidity and there is oxygen present. At high stresses, above the elasticity limit of the material, pitting is enhanced through the mechanism identified as stress induced pitting (Parkins, 1972). In some cases, blade attachment cracking is a combination of stress corrosion cracking and corrosion fatigue because of the effects of blade vibration. Root Causes SCC of discs (at keyways, bores, and blade attachments) is caused by a combination of high surface stresses, a susceptible material, and operational and shutdown environments. Design-related root causes are the most important and prevalent. They include high surface tensile stresses and stress concentrations, and use of high strength materials. Sources of stresses that contribute to SCC of discs include:
Figure 17. Typical Locations and Orientations of SCC Found in LP Turbine Discs. (Courtesy of EPRI, 1982a)
Basic centrifugal load caused by rotor rotation. Locally high

concentration of centrifugal loads caused by variation in the gaps (gauging) between blade and disc rim attachment.
vibratory stresses reduce the life of the cracked disc when the flaws reach a sufficient size that fatigue becomes a dominant mechanism. Steam chemistry root causes of SCC and CF cracking include: Figure 18. Typical Locations of Disc Rim Cracking. (Courtesy of EPRI, 1982a)
Residual machining stresses. Vibratory stressesinteraction of SCC and corrosion fatigue. Also,
Operating
outside of recommended steam purity limits for long periods of time; sometimes caused by organic acids from decomposition of organic water treatment chemicals.
220
Condenser leaksminor but occurring over a long period of time. Condenser leaksmajor ingress, generally one serious event,
and the system and turbine not subsequently cleaned.
Water
all fatigue cracking should be considered corrosion fatigue. The fatigue limit of all turbine materials in turbine environments including pure steam is lower than the air fatigue limit. Corrosion fatigue cracks often originate from pits.
treatment plant or condensate polisher regeneration chemicals (NaOH or H2SO4) leak downstream.
Improperly
operated condensate polisher (operating beyond ammonia breakthrough, poor rinse, etc.).
Shutdown environment: poor layup practices plus corrosive deposits.

Sodium hydroxide is the most severe SCC environment encountered in steam turbines. The sources of NaOH include malfunctioning condensate polishers and makeup systems and improper control of phosphate boiler water chemistry combined with high carryover. Many other chemicals can also cause SCC of low alloy steels. The chemicals used in turbine assembly and testing, such as molybdenum disulfide (lubricant) and Loctite (sealant containing high sulfur), can accelerate SCC initiation (Turner, 1974; Newman, 1974). Solutions In most cases where material yield strength is <130 ksi (895 MPa), the solution to disc SCC is a design change to reduce stresses at critical locations. This has been achieved by eliminating keyways or even disc bores (welded rotors) and by larger radii in the blade attachments. Higher yield strength (>130 ksi, 895 MPa) low alloy steel discs should be replaced with lower strength materials. The goal is to keep the ratio of the local operating stress to yield stress as low as possible, ideally aiming for the ratios to be less than 0.6. Minimizing applied stresses in this manner is most beneficial in preventing initiation of stress corrosion cracks. Once cracks begin to propagate, a reduction in stress may be only marginally effective unless the stress intensity can be kept below ~10 to 20 ksi-in1/2 (11 to 22 MPa-m1/2). This is because of the relative independence of the crack growth rate over a broad range of stress intensities. For many rim attachment designs, such levels of applied stress intensity are impossible to achieve once an initial pit or stress concentration has formed. An emerging solution to disc rim stress corrosion cracking is a weld repair with 12%Cr stainless steel. Another solution has been to shot peen the blade attachments to place the hook fit region into compression. Good control of the steam purity of the environment can help to prevent or delay the SCC. Maintaining the recommended levels of impurities during operation and providing adequate protection during shutdown can help minimize the formation of deposits and corrosive liquid films, and lengthen the period before stress corrosion cracks initiate. The operating period(s), events, or transients that are causing excursions in water and steam chemistry should be identified using the monitoring locations and instrumentation recommended in the independent water chemistry guidelines (EPRI, 1986, 1994c, 1998b, 1998c, 2002a; Jonas, et al., 2000) and special monitoring as shown in Figure 15 and elsewhere (EPRI, 1997c, 2001b; Jonas, et al., 2007). Corrosion Fatigue and Stress Corrosion Cracking of Blades LP turbine blades are subject to CF, SCC, and pitting of the airfoils, roots, tenons and shrouds, and tie wires (Holdsworth, 2002; EPRI, 1984c, 1984d, 1985c, 1985d, 1987c, 1991b, 1993b, 1994b, 1998d; Jaffe, 1983; Evans, 1993; Singh, et al.; BLADE-ST , 2000). Figure 20 depicts the typical locations on an LP turbine rotating blade that are affected by localized corrosion and cracking. In addition, the blade surfaces are also subject to fatigue, deposition, water droplet erosion, and foreign object damage. CF is the leading mechanism of damage. It is a result of the combination of cyclic stresses and environmental effects. There are always environmental effects in fatigue cracking in LP turbines (EPRI, 1984d, 1984e) and
Figure 20. Typical Locations of Cracking and Localized Corrosion on LP Turbine Rotating Blades. There Has Also Been SCC and CF Cracking in the Tiewire Holes. (Courtesy of EPRI, 1998a) Figures 21 and 22 show corrosion fatigue cracks of L-1 blade root and airfoil, respectively. Figure 23 illustrates pitting in the blade tenon-shroud area, which sometimes initiates corrosion fatigue cracking. Fractography shows that CF blade cracking often initiates from a pit, continuing for 50 to 150 mils (1.3 to 3.8 mm) by intergranular cracking and then proceeding as a flat fatigue fracture with beach marks and striations.
Figure 21. Corrosion Fatigue of L-1 Blade Attachment. (Courtesy of EPRI, 1998a)
Figure 22. Corrosion Fatigue of L-1 Blade Airfoil. (Courtesy of EPRI, 1981)
221
Root Causes High stresses and marginal steam chemistry acting together are the most frequent root causes. Corrosion fatigue cracks are driven by cyclic stresses with high mean stress playing a large role. Figure 10 shows how fatigue strength is reduced at high mean stresses. It has been estimated that the corrosion fatigue limit for long LP turbine blades in the area of high mean stress is as low as 1 ksi (7 MPa). Rough surface finish and pitting often shorten the time to crack initiation. Cyclic stresses are caused by turbine startups and shutdowns (low number of cycles, high cyclic stresses), by the turbine and blades ramping through critical speeds at which some components are in resonance (high amplitude, high frequency), and by over 15 causes of flow induced blade excitation during normal operation that include: Figure 23. Pitting in the L-1 Blade Tenon-Shroud Area. (Courtesy of EPRI, 1981) Damage by corrosion fatigue occurs in the last few rows of LP turbines, mostly in the phase transition zone (PTZ) (salt zone shown in Figure 14). The PTZ moves according to load changes, but is typically near the L-1 row in most LP fossil turbines (Figure 24). Units that increase cycling duty may be subject to worsened corrosion fatigue problems. As the unit is ramped up and shut down, the blades pass through resonance more frequently and the phase transition zone shifts, potentially affecting more stages during the transients. In addition, the steam purity can be significantly worse during transients than during steady-state operation, and if the unit is shut down as part of the cycling operation, significant degradation of the local environment can occur (deposits and humid air).
Synchronous resonance of the blades at a harmonic of the unit

running speed.
condensation interaction. separation.
Nonuniform flows. Blade vibration induced from a vibrating rotor or disc. Self-excitation such as flutter. Random excitation-resonance with adjacent blades. Shock waves in the transonic flow region and shock wave Bad blade design with wrong incidence flow angle and flow
Elevated concentrations of steam impurities (particularly of chloride, sodium, and sulfate) and the resulting deposition and concentration by evaporation of moisture are underlying causes of corrosion fatigue. When feedwater, boiler water, and steam impurity levels exceed recommended limits, cycle chemistry can be a contributor or even the root cause. Poor shutdown and layup procedures are primary contributors to aggressive environments that can lead to pitting and corrosion fatigue. High steam sodium or cation conductivity may indicate conditions that can lead to rapid accumulation of deposits and concentrated liquid films on blade surfaces. Solutions The solutions to blade corrosion fatigue problems include:
Design of blades that are not in resonance with running speed

Figure 24. Distribution of Blade Failures in U.S. Fossil Turbines by Row. (Courtesy of Power, 1981) Causes of blade and blade attachment failures are listed in Table 5 (Jonas, 1985a). To find the true causes of corrosion, it is essential to analyze the local temperature, pressure, chemistry, moisture droplet flow, and stress conditions. These analyses are often neglected. Table 5. Causes of Blade Failures in LP, IP, and HP Steam Turbines.
and its harmonic frequencies or with any of the excitation sources listed above.
Design with friction damping. Elimination of the sources of excitation. Reduction of mean and alternating stresses by design (lower stress
concentrations, etc.).
Better materials, such as by avoiding high strength alloys, using

materials with high material damping, or using titanium alloys.
Improvement of steam chemistry.

Long-term actions for dealing with corrosion fatigue begin with economic and remaining life assessments. Depending upon the severity of the problem and the costs to eliminate it, some solutions may not be practical for all circumstances. The available prevention strategies fall into four main categories: redesigning the blade to reduce resonance, redesigning the blade or attachment to reduce stress levels, improving steam purity, and/or changing the material or surface (better surface finish, shot peening) of the blade. Stress reduction options include:
Changing
the vibration resonance response of the blade by design modification (adding or reducing weight of the blade,
222
changing to free standing, grouped or fully connected shrouded blades, moving tiewires or tenons, shroud segment integrally machined with the bladeno tenons, etc.).
Solutions The first line of defense against pitting is controlling steam purity. This will improve the local environment produced during operation and decrease the amount of deposition. The chemistry guidelines established by an independent research and development firm (EPRI, 1986, 1994c, 1998b, 1998c, 2002a; Jonas, et al., 2000) particularly for cation conductivity, chloride, and sulfate, should be followed. There should be a layup protection of the LP turbines by dehumidified air, vapor phase inhibitors, or nitrogen. After a steam chemistry upset, such as a large condenser leak, the turbine should be washed online or after a disassembly. Doing nothing may result in multimillion dollar corrosion damage requiring rotor replacement. Flow-Accelerated Corrosion Flow-accelerated corrosion of carbon and low-alloy steels in the steam path two-phase flow has been less widespread in fossil plants than in nuclear plants; however, it has occurred at some locations (EPRI, 1996; 1998a; Kleitz, 1994; Jonas, 1985b; Svoboda and Faber, 1984) such as:
Changing the response of the blades by mixed tuning. Increasing the damping of the blades. Changing operating procedures; for example, avoiding off-design
operation such as very low load, high backpressure operation (which can cause stall flutter), and changing rotational speed during startups. Options to improve the turbine environment include:
Controlling impurity ingress (reduce air inleakage, plug leaking

condenser tubes, etc.). and startups.
Changing unit operating procedures, particularly for shutdowns Control of boiler carryover by drum design and water chemistry. Optimizing or changing feedwater and boiler water treatment to
reduce concentration of impurities in steam. Another option to improve the environment at surfaces is to improve the surface finish of blading. Deposition and subsequent concentration of impurities are a function of blade surface finish (EPRI, 2001b), and this improvement may help slow the accumulation of impurities. Improving the continuous monitoring of steam chemistry (sodium, cation conductivity) will help to verify improvements in the environment. If such changes are not sufficient, then changing to a more corrosion resistant material, such as a material with a higher chromium content or a titanium alloy, is generally recommended. It should be noted that higher strength materials are often more susceptible to stress corrosion cracking. It has been shown that 403SS, with yield strength above ~90 ksi (620 MPa), becomes susceptible to SCC. Pitting Pitting can be a precursor to more extensive damage from CF and SCC, although extensive pitting of blades can also cause significant loss of stage efficiency by deteriorating the surface finish (EPRI, 2001b). Pitting is found in a wide range of components. It occurs most prevalently during shutdown when moisture condenses on equipment surfaces and as a result, it can be found in stages of the turbine that are dry during operation. However, it can also occur during operation, particularly in crevices (crevice corrosion). In LP turbines, pitting is primarily found on the turbine blades and blade-disc attachments, particularly in the salt zone (Figure 14). There is little pitting on wet stages because corrosion impurities are washed away by the steam moisture. Pitting is frequently found in the blade tenon-shroud crevices because, once corrosive impurities enter, they cannot be removed. Titanium alloys are the most resistant to pitting corrosion, followed by duplex ferritic-austenitic stainless steels (Fe-26Cr-2Mo), precipitation hardened stainless steels (Fe-14Cr-1.6Mo) and 12%Cr steels. Root Causes Steam and deposit chemistries are the main causes of pitting, with chlorides and sulfates being the main corrodents. Copper and iron oxides accelerate pitting by providing the matrix that retains salts. Copper oxides also transport oxygen to the corrosion sites. Dissolved oxygen does not concentrate in the liquid films forming on blade surfaces during operation but does, however, accumulate in the liquid films and wet deposits that can form during unit shutdown if proper layup practices are not used. There have been cases of severe blade pitting requiring blade replacement on brand new rotors from which preservatives were stripped leaving them exposed to sea salt during prolonged erection periods.
Wet steam extraction pipes and extraction slots. Exhaust hood and condenser neck structure. Casings. Rotor gland and other seal areas. Disc pressure balance holes. Rim and steeples of last row disc. Rotor shaftlast disc transition. Leaking horizontal joint. Transition between the stationary blades and the blade ring.
While most cases of flow-accelerated corrosion damage are slow to develop and are found during scheduled inspections, FAC of piping and turbine casing horizontal joints can lead to leaks and FAC of rotors and discs can initiate cracking. Root Causes Root causes of FAC in the turbine (EPRI, 1996, 1998a; Jonas, 1985b) include:
Susceptible material: carbon steel or low-chromium steel. Locally high flow velocities and turbulence. High moisture content of the steam. Low levels of dissolved oxygen, excess oxygen scavenger. Low pH of moisture droplets. Water/steam impurities.
Solutions It is recommended that a comprehensive FAC control program be implemented, including an evaluation of the most susceptible piping and other components (EPRI, 1996). Areas of local thinning need to be periodically inspected and repaired. Approaches to piping repair include replacement with low alloy steels, weld overlay, and plasma arc and flame spraying to protect susceptible surfaces. The material applied should have high chromium content. If a component is replaced, the material of the new component should contain some chromium. For example, carbon steel pipe should be replaced with 1.25 percent or higher chromium steel. Little can be done about changing the moisture concentration in fossil turbines. Steam chemistry improvements through better control of feedwater and boiler water chemistry, such as reduction of organic acids, could result in increase of pH of the early condensate and less FAC (EPRI, 1997b, 1997c, 1999).
223
Treatment of feedwater, such as maintaining high pH levels (above 9.6) and elevated oxygen concentrations, can also reduce FAC in the turbine. Other Phenomena (Noncorrosion) Although not specifically corrosion related, there are other problems that occur in steam turbines including: deposition on blade surfaces; water droplet erosion of wet stage blades; low cycle thermal fatigue of heavy high temperature sections of rotor, casing, and pipes; solid particle erosion of turbine inlets and valves; and water induction and water hammer. Deposition on Blade Surfaces Deposits are the result of impurities in the feedwater, boiler water, and attemperating water being carried over into the turbine (Jonas and Dooley, 1997; EPRI, 1997b, 2001b; Jonas, et al., 1993; Jonas, 1985d). All impurities are soluble in superheated and wet steam and their solubility depends on pressure and temperature. The steam leaving the steam generator is at the highest steam pressure and temperature in the cycle. As it passes through the turbine, the pressure and temperature decrease, the steam loses its ability to hold the impurities in solution, and the impurities precipitate and deposit on the turbine blades and elsewhere. The main impurities found in turbine deposits are magnetite, sodium chloride, and silica. It takes only a few hours of a chemical upset, such as a major condenser leak or a boiler carryover event, to build up deposits, but it takes thousands of hours of operation with pure steam to remove them. The impact of deposits on turbine performance is the most pronounced in the HP section. Performance loss depends on deposit thickness, their location (steam pressure), and the resulting surface roughness (EPRI 2001b). Deposits will change the basic profile of the nozzle partitions resulting in losses caused by changes in flow, energy distribution, and aerodynamic profiles, as well as by surface roughness effects. These changes can result in large megawatt and efficiency losses. With replacement power typically over $100/MWh and costing as much as $7000 per MWh in the summer of 1998, the savings from reducing this deposition can be very high. In the LP turbine, deposits are often corrosive, they can change the resonant frequency of blades, increase the centrifugal load on blade shrouds and tenons and, in the transonic stages, they can influence the generation of shock waves. SolutionsOptimization of cycle chemistry (Jonas, 1982, Progress in..., 1981; EPRI, 1986, 1994c, 1998b, 1998c; 1985d, 2002a; Jonas, et al., 2000, 2007; ASME, 2002) is the easiest method for reducing impurity transport and deposition. The optimal cycle chemistry will result in reduced corrosion and minimized impurity transport. This is especially important if copper alloys are present in the system because the optimal feedwater pH for copper alloys and ferrous materials are not the same and the incorrect pH can result in high levels of iron or copper transport. Other methods for managing deposition on blade surfaces include:
Water Droplet Erosion In the last stages of the LP turbine, the steam expands to well below saturation conditions and a portion of the vapor condenses into liquid (EPRI, 2001b; Ryzenkov, 2000; Oryakhin, et al., 1984; Rezinskikh, et al., 1993; Sakamoto, et al., 1992; Povarov, et al., 1985; Heyman, 1970, 1979, 1992). Although the condensed droplets are very small (0.05 to 1 m, 2 to 40 in diameter), some of them are deposited onto surfaces of the stationary blades where they coalesce into films and migrate to the trailing edge. Here they are torn off by the steam flow in the form of large droplets (5 to 20 microns, 0.2 to 0.8 mils). These droplets accelerate under the forces of the steam acting on them and, when they are carried into the plane of rotation of the rotating blades, they have reached only a fraction of the steam velocity. As a result, the blades hit them with a velocity that is almost equal to the circumferential velocity of the blades, which can be as high as 640 m/s (2100 ft/s) in a fossil LP turbine. Water droplet erosion typically occurs in the last two to three rows of the LP turbines in fossil fired units. The damage is most common on the leading edge and tip of the blades and along the shroud. In turbines operating at low load for long periods, such as cycling and peaking units, reversed flow of steam caused by windage and activation of hood spray can erode the trailing edges of blades. Thin trailing edges with erosion grooves can become fatigue or corrosion fatigue crack initiation sites. Less frequent erosion damage locations include LP turbine glands and seals, stationary blades, blade attachment sections of discs, disc flow holes (impulse design), and the LP rotor at the gland. The effects of steam and early condensate chemistry on water droplet erosion are not known, but studies have found that NaCl in the droplets significantly reduces the incubation period for erosion. In addition, pH was found to have a strong impact on both the incubation period and the erosion rate. At higher pH, both the maximum and steady state erosion rates decrease while the incubation period increases (Ryzenkov, 2000; Povarov, et al., 1985). SolutionsThere are several options available for reducing the amount of damage from liquid droplet impact. There are two principal options: protection of the leading edge by a hard material and collection and drainage of moisture. Table 6 outlines these options. Table 6. Long-Term Actions for Reducing Moisture Erosion on LP Turbine Blades.
Specify good surface finishes (polished blades) on all new and

replacement blades.
Turbine Inspection and Monitoring Adequate turbine inspection methods are available to detect corrosion and deposition. These NDT methods include visual, magnetic particle, ultrasonic, dye penetrant, eddy current, and radiographic techniques. Modern turbine designs consider the accessibility of individual components by inspection probes. Monitoring techniques include stress and vibration monitoring (i.e., vibration signature), temperature and flow measurement, and water and steam chemistry sampling and analysis. (Jonas, 1982, 1985d, 1986, 1994; EPRI, 1984b, 1994a, 1994c, 1997b, 1997c, 1998c, 1999, 2002a; Jonas, et al., 1993, 2000; Jonas and Syrett, 1987; Schleithoff, 1984; Progress in..., 1981). An advanced expert system has been developed (EPRI, 1994c) for the use by station operators and chemists, which automatically determines the
Determine the effect of erosion and deposition on maximum
MW and efficiency with a valves wide open (VWO) test. The data from several VWO tests can be compared to determine the rate of MW loss and MW versus chemistry and operation. This can be used to optimize the system.
Turbine
washing to reduce deposition and MW losses and improve efficiency. Both, a turbine wash of an assembled turbine and a wash of a disassembled rotor can be used to remove soluble deposits. To remove corrosive salt deposits, several days of washing may be needed. A wash is usually completed when the concentration of corrosive impurities, such as sodium and chloride, in the wash water is less than 50 ppb.
224
problems and recommends corrective actions. There are also monitoring methods for online diagnosis of the environments on turbine surfaces and corrosion (Jonas, 1994; EPRI, 1994a, 1997b, 1997c, 1999; Jonas, et al., 1993).
MISSING KNOWLEDGE
It is estimated by the author that 70 percent of knowledge to solve and prevent corrosion problems in steam turbines is available. The percentage of available knowledge for understanding the effects of stress and environment is much lower than that for solving the problems, about 40 percent. The knowledge that is missing or needs improvement includes:
Threshold stress required to initiate SCC in blade attachments. Effects of steeple geometry (stress concentrations and size) on
SCC and CF.
Effects
measure.
of overloads during heater box and overspeed tests on stress redistribution and SCC in steeples, blade roots, and disc keyways.
Effectiveness of grinding out SCC and CF cracks as a corrective Effects of organic water treatment chemicals, and organic impurities on SCC, CF, and pitting and composition of water droplets. Effects of electrical charges carried by water droplets on corrosion. Effects of galvanic coupling of dissimilar materials, such as the
blade-steeple, on corrosion.
Effects of residues of preservatives and Loctite on SCC, CF,

and pitting of blade attachments.
Effects of blade trailing edge erosion on cracking. Accelerated stress corrosion testing. Effects of variable amplitude loading on CF crack
and propagation. to predict erosion. and disc steeples.
initiation
Effects of water droplet pH and composition on erosion and how Effects of shot peening to reduce stresses and SCC of blade roots Understanding
of the basic mechanisms of stress corrosion, corrosion fatigue, fatigue, and stress induced pitting.
CASE HISTORIES
Stress Corrosion Cracking in Finger-Style Dovetails Unit: The station consists of three 805 MW, once-through boiler, supercritical, coal-fired units that went into operation between 1974 and 1976. Each of the three units has two LP turbines (LPA and LPB). Problem: In 1995, in Unit 1, a blade failed in the L-1 row at a tiewire hole of the leading blade of a four-blade group (Nowak, 1997; Kilroy, et al., 1997). Three damaged groups of blades were removed for replacement. A wet fluorescent magnetic particle examination of the finger-style disc attachments found hundreds of crack-like indications, which were later identified as SCC. Cracks ran both axially and radially, and were deep (Figure 25).
Figure 25. Locations in Finger and Pin Attachments Where SCC Has Been Found. (Courtesy of EPRI, 1997)
Inspection: Inspection procedures and the acceptance criteria that would be applied were developed before the outage. Eventually NDE inspection found some damage in each of the 12 ends of all six rotors. Severe cracking was found in all four-rotor ends in Unit 2 and in Unit 3 on both ends of LPB. Next in severity were both ends of Unit 1 LPB and Unit 3 LPA; with a few indications in Unit 1 LPA. In the heavily cracked rotors, damage was most severe in Fingers 3, 4, and 5 with little or no cracking in Fingers 1 or 6. Cracking was evenly distributed between admission and discharge sides in Fingers 3 and 4 with somewhat more cracking on the admission side in Finger 5. Results of metallurgical analysis: A metallurgical evaluation confirmed the presence of extensive pitting. Cracks were found to be intergranular, highly branched, and oxide-filled. The metallurgical examination was unable to detect the presence of specific contaminants on the fracture surface. Sampling of deposits, which had occurred elsewhere in the cycle (crossover piping, bucket pins, and bucket fingers) during prior outages, had found indications of sodium, sulfur, and chloride and sodium hydroxide was found by x-ray diffraction in crossover piping. Analysis of samples from two rotors showed that the chemical composition was within the specification for ASTM A470 Class 7 material. Tensile strength averaged 126.5 ksi (870 MPa) for the two specimens; yield averaged 112.5 ksi (775 MPa). Review of cycle chemistry: Several cycle chemistry changes had been made over time in response to events such as improved technology and information, and upsets caused by condenser tube leaks, demineralizer breaks, and variations in boiler water makeup. The main water chemistry problem was operation with morpholine, which resulted in poor condensate polisher performance and high concentrations of sodium hydroxide in feedwater and steam. A heavy deposit of sodium hydroxide was found in the crossover piping. The unit had been changed to oxygenated treatment about the time of the discovery of the stress corrosion cracks, which has since resulted in better water and steam purity. Results of stress analysis: A finite element stress analysis showed that Fingers 3, 4, and 5 of the group were the most highly stressed. The maximum equivalent elastic stresses around the pin holes were ~222 ksi (1530 MPa), ~114 ksi (786 MPa) at the inner land, and ~89 ksi (614 MPa) at the outer land (closer to disc outer diameter [OD]). The differences between inner and outer land stresses agreed with the observation that field cracking was more severe at the inner land. However, no SCC was discovered in the flat portion of the disc fingers around the holes, where the stress was nearly twice as high. Closer examination found that the flat surface near the pinholes did not show cracks nucleating from the bottom of the pits, whereas intergranular cracks appeared in the pits along the ledge where the pits were linked to form continual flaws. Root causes: There were two root causes of this massive problem: high design stresses and improper feedwater chemistry using morpholine, which resulted in the presence of NaOH and other impurities in steam. One can also speculate about the contributions of the local stress concentration, poor surface finish, and residual machining stresses. Economic analysis: An economic analysis was performed and the costs considered included: replacement of rotors in-kind, replacement with an improved rotor steam path that would improve unit heat rate by 1.2 percent, rotor weld repair, outage duration costs, performance changes, reduced generation capacity from pressure plates, fuel pricing, and replacement energy costs. Actions: As a temporary fix, the affected blade rows were removed and nine pressure plates were installed out of the 12 possible locations. A pressure plate is a temporary device that provides a pressure drop when installed as a replacement for a removed blade row. A decision was made to purchase two new fully bladed rotors and to refurbish the existing rotors. For Unit 1, new rotors were purchased from the original equipment manufacturer (OEM). The two LP rotors removed from Unit 1 were weld repaired
225
using 12 percent chromium material applied with a submerged arc weld process, and installed in Unit 3. Material testing and analysis were used to determine the expected lifetime of the refurbished parts against damage by SCC, and both high- and low-cycle fatigue. Turbine and crossover pipes were cleaned to remove deposits. Massive SCC of Disc RimSupercritical Once-Through Unit after Only Five Years of Service Problem: Stress corrosion cracking of the L-1 stage disc was discovered during routine inspection (Figure 26).
Corrosion Fatigue of Numerous Modifications of L-1 Blades Problem: One type of LP turbine experienced corrosion fatigue failures of the L-1 blade airfoil, which was modified and redesigned over 10 times. Fatigue failures occurred in periods ranging from six weeks to over 10 years of operation. The presence of chloride in the blade deposits at concentrations above 0.25 percent caused pitting, which accelerated crack initiation. In one case, new 16 inch (40 cm) blades were forged in two dies and most of the blades forged in one of the dies (but none from the other die) failed within about six weeks. Root cause: The root cause of this fast CF failure was an off-design blade geometry caused by wrong die dimensions, which brought these blades into resonance. The failure acceleration was caused by corrosive impurities in steam, mainly chloride. Actions: Redesigned, better tuned blades were installed and improved control of water and steam chemistry was initiated. Massive Pitting of a Turbine (HP , IP , and LP) after Brackish Water Ingress
Figure 26. Massive SCC of L-1 Disc Caused by High Concentration of NaOH in Steam. Root cause: The SCC was caused by poor performance of the condensate polishers that operated in H-OH form through ammonia breakthrough when they released Na+. The sodium reacted with water forming NaOH and deposited in the turbine. Actions: Rotor replaced and operation of condensate polishers and monitoring was improved. Stress Corrosion Cracking of Bolted-on Discs in One Type of LP Turbine Problem: In the effort to accommodate longer L-0 blades, one type of LP turbine was originally designed with the bolted-on last disc made of NiCrMoV low-alloy steel and heat treated to a yield strength of up to 175 ksi (1200 MPa). This steel was found to be susceptible to SCC. Root cause: The root cause of this problem was design with a high strength material that is very susceptible to SCC. Solution: All of these discs in many power plants had to be replaced with a lower strength material because of the danger of SCC failures in all types of steam environments. Corrosion Fatigue of a Blade Airfoil Unit: A 400 MW reheat unit with a once-through boiler, seawater cooling, and mixed bed condensate polishers. Problem: During a three-month period, condenser cooling water leakage occurred periodically. Cation conductivity in the condensate and feedwater increased up to 2 S/cm for about 30 to 60 minutes per day. At the end of the three-month period, vibration was detected in the LP turbine (EPRI, 1998a) Damage: The turbine was opened and five broken freestanding L-2 rotating blades were found. According to the turbine design data, the broken blades were in the phase transition zone. The blades were broken at the transition between reddish deposits at the blade foot and clean metal in the upper part of the blades (phase transition on the blade). A laboratory investigation found chloride and sodium at the crack site and confirmed corrosion fatigue as the underlying mechanism. A calculation of vibration frequencies did not show abnormal conditions. Root cause: Improper water chemistry conditions caused by the condenser tube leak. Actions: The broken blades were replaced, the condenser leak was repaired, and a dampening (lacing) wire was introduced into the blade design. By doing so, both the environmental and stress contributors were reduced.
Problem: A separation of the welded end of the condensate sparger caused breakage of condenser tubes and ingress of brackish water into a once-through boiler cycle. Because of the poor reliability of the water chemistry instrumentation, the instrument readings were ignored and the trouble was noticed after there was almost no flow through the superheater because of heavy deposits. The turbine with sea salt deposits was left assembled in the high humidity environment and was only opened 11 days after the condenser tube failure. It was found that the whole turbine was severely rusted and pitted and it was eventually replaced. Mixed bed condensate polishers were not able to protect the cycle against the massive ingress of brackish water. They were exhausted within a few minutes. Root cause: The root cause of the impurity ingress was a failure of the condensate sparger. Poor water chemistry monitoring and control and the long delay in beginning turbine damage assessment and cleaning significantly contributed to the amount of damage caused. Actions: The rotors were replaced and new chemistry monitoring instrumentation was installed. L-0 Blade Corrosion Fatigue Cracking Caused by Trailing Edge Erosion Problem: After 14 to 18 years of service of one type of 400 MW turbine, there were five cases of L-0 33.5 inch (85 cm) long shrouded blade failures about 5 inches (12.7 cm) below the tip. The CF cracks originated at the eroded and thinned trailing edge. When the blade tips separated, the blade fragments had such kinetic energy that they penetrated over 2 inches (5 cm) of carbon steel condenser struts. The blade material was martensitic 12%Cr stainless steel with a Brinell hardness of ~345. All affected units were similar drum boiler units, some on all volatile treatment (AVT) and some on phosphate treatment. Steam chemistry in the affected units was good and did not play a role in the rate of erosion or cracking. Root cause: Erosion caused by frequent operation at low load with the hood sprays on and reversed steam flow, lack of proper early inspection, and blade design with thin trailing edge. Actions: Heavily eroded and cracked blades were replaced, shallow erosion damage was polished, and similar turbines were inspected for damage. Stress Corrosion Cracking of Dovetail Pins Problem: A dovetail pin, which penetrates both the rotating blade and the wheel dovetails, holds a bucket in place on the wheel of a rotor. In many plants during the 1970s and earlier, dovetail pins had suffered stress corrosion cracking, although no lost blades had resulted. The material originally used for the dovetail pins was similar to ASTM A681 Grade H-11 tool steel, with a chemistry of Fe-5.0Cr-1.Mo-0.5V-0.4C at a strength level of 250 to 280 ksi
226
(1715 to 1920 MPa). This material is still used in rare cases where the highest strength is required. Root cause: Use of high strength material, which is susceptible to SCC, combined with high bending stresses. Actions: The approach to solving the cracking problem was twofold: changing the material chemistry and strength and using steel ball shot peening to impart a compressive layer to the surface of the finished pins. The new high strength material is 5CrMoV low alloy steel at a strength level of 240 to 270 ksi (1645 to 1850 MPa). For lower stress applications, a new 1CrMoV low alloy steel with a strength of (170 to 200 ksi) is used. Results: No cracking has been observed in dovetail pins since the change in materials and the introduction of the shot peening practice. Turbine DestructionSticking Valves Problem: After only 16 hours of operation of a new 6 MW steam turbine installed in a fertilizer plant, an accidental disconnect of the electrical load on the generator led to a destructive overspeed. The overspeed occurred because high boiler carryover of boiler water treatment chemicals, including polymeric dispersant, introduced these chemicals into the bushings of all turbine control valves, gluing the valves stuck in the open position. Root cause: Poor control of boiler operation (drum level) together with the use of the polymeric dispersant that, after evaporation of water, becomes a strong adhesive. Controls of the electric generator allowed accidental disconnect. Actions: New turbine generator installed, boiler and generator controls fixed, turbine valves reused after dissolution of the bushing deposits in hot water.
Results of 1982 Survey, American Society of Mechanical Engineers, New York, New York. ASME, 1989, The ASME Handbook on Water Technology for Thermal Power Systems, American Society of Mechanical Engineers, New York, New York. ASTM 2001, 2001 Annual Book of ASTM Standards, Section 3: Metals Test Methods and Analytical Procedures, 03, Nondestructive Testing, West Conshohocken, Pennsylvania. ASME, 2002, Consensus for the Lay-up of Boilers, Turbines, Turbine Condensers, and Auxiliary Equipment, ASME Research Report, CRTD-Vol. 66, 2002. American Society of Mechanical Engineers, New York, New York. Atrens, A., et al., 1984, Steam Turbine Blades, Corrosion in Power Generating Equipment, Plenum Press. Baboian, R., Editor, 1995, Corrosion Tests and Standards Application and Interpretation, ASTM, West Conshohocken, Pennsylvania. BLADE-ST New York. Version 3, 2000, STI Technologies, Rochester,
Cheruvu, N. and Seth, B., 1993, Key Variables Affecting the Susceptibility of Shrunk-On Discs to Stress Corrosion Cracking, The Steam Turbine Generator Today: Materials, Flow Path Design, Repair, and Refurbishment, ASME, New York, New York. Clark, W. G., et al., 1981, Procedures for Estimating the Probability of Steam Turbine Disc Rupture from Stress Corrosion Cracking, ASME/IEEE Power Generation Conference, Paper No, 81-JPGC-PWR-31, New York, New York. Cotton, K., 1993, Evaluating and Improving Steam Turbine Performance, New York, New York: Cotton Fact, Inc. EPRI, 1981, Steam Turbine Blades: Considerations in Design and a Survey of Blade Failures, Palo Alto, California, CS-1967. EPRI, 1982a, Steam Turbine Disc Cracking Experience: Volume 1Literature and Field Survey, Volume 2Data Summaries and Discussion, Volume 3Stress Corrosion Cracking of Low-Alloy Steels, Volume 4Factors Determining Chemical Composition of Low Pressure Turbine Environments, Volume 5Characteristics and Operating Histories of U.S. Power Plants, Volume 6Description of Turbine Rotor Models, Volume 7Metallurgical Analysis of Cracked Discs from 10 U.S. Power Plants, Palo Alto, California, NP-2429-LD. EPRI, 1982b, Metallurgical Evaluation of a Failed LP Turbine Disc, Palo Alto, California, NP-2738. EPRI, 1983, EPRI Research Related to the Steam Turbine Generator, Palo Alto, California, NP-3288-SR. EPRI, 1984a, Properties of Turbine Disc Materials, Palo Alto, California, NP-3634. EPRI, 1984b, The Effects of Phosphate Environments on Turbine Materials, Palo Alto, California, CS-3541. EPRI, 1984c, Stress Corrosion Cracking in Steam Turbine Discs: Survey of Data Collection, Reduction, and Modeling Activities, Palo Alto, California, NP-3691. EPRI, 1984d, Corrosion Fatigue of Steam Turbine-Blading Alloys in Operational Environments, Palo Alto, California, CS-2932. EPRI, 1984e, The Effect of Variables on the Fatigue Behavior of Ti-6Al-4V , Palo Alto, California, CS-2934. EPRI, 1985a, Stress Corrosion Cracking in Steam Turbine Discs: Analysis of Field and Laboratory Data, Palo Alto, California, NP-4056. EPRI, 1985b, Survey of Steam Turbine Blade Failures, Palo Alto, California, CS-3891.
CONCLUSIONS
Steam turbines can be a very reliable equipment with life over
30 years and overhaul approximately every 10 years. However, about 5 percent of the industrial and utility turbines experience corrosion and deposition problems. Mostly due to LP blade and blade attachment (disc rim) corrosion fatigue or stress corrosion failures. high stresses, bad steam chemistry, and use of high strength materials.
The root causes of the blade and disc failures include design with Other
steam turbine problems include: low cycle thermal fatigue, pitting during unprotected layup and operation, loss of MW/HP and efficiency due to deposits, water droplet erosion, flow accelerated corrosion, solid particle erosion by magnetite particles exfoliated from superheater, turbine destructive over speed caused by the control valves stuck open because of deposits in the bushings, and water induction-water hammer.
All the problems are well understood, detectable, and preventable.
Monitoring, inspection, and defects evaluation methods are available. These methods include design reviews and audits of operation and maintenance, NDT, life prediction, vibration monitoring, vibration signature analysis, water, steam, and deposit chemistry monitoring and analysis, valve exercise, and control of superheater temperatures.
Steam cycle design and operation influences turbine problems
by causing high steady and vibratory stresses, by thermal stresses related to load and temperature control, and by water and steam purity and boiler carryover.
REFERENCES
Amos, D., Lay, E., and Bachman, S., 1997, Qualification of Welding Rotors with 12Cr Stainless Steel to Improve SCC Resistance, Proceedings: Steam Turbine Stress Corrosion Workshop, EPRI, Palo Alto, California, TR-108982. ASM International, 1989, ASM Metals Handbook, 9th Edition, 17, Nondestructive Evaluation and Quality Control. ASME, 1982, ASME Committee on Water in Thermal Power Systems, Industrial Subcommittee, Steam Purity Task Group,
227
EPRI, 1985c, Proceedings: Steam Turbine Blade Reliability Seminar and Workshop, Palo Alto, California, CS-4001. EPRI, 1985d, Stress Corrosion Cracking and Corrosion Fatigue of Steam Turbine Materials, Palo Alto, California, NP-4074M. EPRI, 1986, Interim Consensus Guidelines on Fossil Plant Chemistry, Palo Alto, California, CS-4629. EPRI, 1987a, Guide for Use of Corrosion-Resistant Coatings on Steam Turbine Blades, Palo Alto, California, CS-5481. EPRI, 1987b, Stress Corrosion Cracking of A471 Turbine Disc Steels, Palo Alto, California, NP-5182. EPRI, 1987c, Steam Turbine Blade Reliability Seminar and Workshop, Palo Alto, California, CS-5085. EPRI, 1989, Guidelines for Predicting the Life of Steam Turbine Discs Exhibiting Stress Corrosion Cracking, Palo Alto, California, NP-6444. EPRI, 1991a, Proceedings: Fossil Steam Turbine Disc Cracking Workshop, Palo Alto, California, GS-7250. EPRI, 1991b, Florida Power and Light Uses BLADE Program to Analyze Low Pressure Turbine Blade Failures, Palo Alto, California, 100229. EPRI, 1993a, Solid Particle Erosion Technology Assessment, Final Report, Palo Alto, California, TR-103552. EPRI, 1993b, Proceedings of the Steam and Combustion Turbine Blading Conference and Workshop1992, Palo Alto, California, TR-102061. EPRI, 1994a, Steam Chemistry and Corrosion, Palo Alto, California, TR-103738. EPRI, 1994b, BLADE Code Helps TVA Diagnose Blade Failure Quickly and Avoid Blade Replacements, Palo Alto, California, IN-10420. EPRI, 1994c, Cycle Chemistry Guidelines for Fossil Plants: Oxygenated Treatment, Palo Alto, California, TR-102285. EPRI, 1995, Low-Pressure Turbine Blade Design Evaluation, Palo Alto, California, TR-105839. EPRI, 1996, Flow-Accelerated Corrosion in Power Plants, Palo Alto, California, TR-106611. EPRI, 1997a, Proceedings: Steam Turbine Stress Corrosion Workshop, Palo Alto, California, TR-108982. EPRI, 1997b, Turbine Steam, Chemistry and Corrosion: Experimental Turbine Tests, Palo Alto, California, TR-108185. EPRI, 1997c, Steam, Chemistry, and Corrosion in the Phase Transition Zone of Steam Turbines: Volume 1: Key Results, Summary, and Interpretation; Volume 2: Individual Contributions of Participants, Palo Alto, California, AP-108184. EPRI, 1997d, Low-Pressure Rotor Rim Attachment Cracking Survey of Utility Experience, Palo Alto, California, TR-107088. EPRI, 1997e, Moisture Nucleation in Steam Turbines, Palo Alto, California, TR-108942. EPRI, 1998a, Turbine Steam Path Damage: Theory and Practice, Palo Alto, California, AP-108943. EPRI, 1998b, Cycling, Startup, Shutdown, and LayupFossil Plant Cycle Chemistry Guidelines for Operators and Chemists, Palo Alto, California, TR-107754. EPRI, 1998c, Interim Cycle Chemistry Guidelines for Combined Cycle Heat Recovery Steam Generators (HRSGs), Palo Alto, California, TR-110051. EPRI, 1998d, Proceedings: Workshop on Corrosion of Steam Turbine Blading and Discs in the Phase Transition Zone, Palo Alto, California, TR-111340.
EPRI, 1999, Turbine Steam, Chemistry and Corrosion-Generation of Early Liquid Films in Turbines, Palo Alto, California, TR-113090. EPRI, 2000a, Productivity Improvement Handbook for Fossil Steam Power Plants: Second Edition, Palo Alto, California, TR-114910. EPRI, 2000b, Corrosion of Low Pressure Steam Turbine Components, Palo Alto, California, 1000557. EPRI, 2001a, Cost of Corrosion in the Electric Power Industry, Palo Alto, California, 1004662. EPRI, 2001b, Steam Turbine Efficiency and Corrosion: Effects of Surface Finish, Deposits, and Moisture, Palo Alto, California, 1003997. EPRI, 2002a, Cycle Chemistry Guidelines for Fossil Plants: All-Volatile Treatment, Palo Alto, California, 1004187. EPRI, 2002b, Advances in Life Assessment and Optimization of Fossil Power Plants, Proceedings, Palo Alto, California, 1006965. EPRI, 2002c, Priorities for Research and Development for the Electric Power Industry, Palo Alto, California, 1007274. Evans, D., 1993, Comparison of Unstalled Flutter Predictions and Field Measurements for Steam Turbine Blades, Proceedings of the Steam and Combustion Turbine-Blading Conference and Workshop1992, EPRI, Palo Alto, California, TR-102061. Haas, H., June 1977, Major Damage Caused by Turbine or Generator Rotor Failures in the Range of the Tripping Speed, Der Maschinenschaden. Heymann, F., 1970, Toward Quantitative Prediction of Liquid Impact Erosion, ASTM STP 474, American Society for Testing and Materials. Heymann, F., 1979, Conclusions from the ASTM Interlaboratory Test Program with Liquid Impact Erosion Facilities, Proceedings of the Fifth International Conference on Erosion by Liquid and Solid Impact, Cavendish Laboratory, University of Cambridge, United Kingdom. Heymann. F., 1992, Liquid Impingement Erosion, ASM Handbook, Volume 18: Friction, Lubrication, and Wear Technology, ASM International. Holdsworth, S., 2002, Prediction and Prevention of Stress Corrosion and Corrosion Fatigue Cracking LP Steam Turbines, Advances in Life Assessment and Optimization of Fossil Power PlantsProceedings, EPRI, Palo Alto, California, 1006965. Jaffee, R., Editor, 1983, Corrosion Fatigue of Steam Turbine Blade Materials, New York, New York: Pergamon Press. Jonas, O., 1977, Major Damage to Steam-Turbosets, Panel Discussion at the 9th Allianz Forum, Technology and Insurance, 1976, and Der Maschinenschaden, 1977. Jonas, O., 1978, Tapered Tensile Specimen for Stress Corrosion Threshold Stress Testing, ASTM Journal of Testing and Evaluation, 6, (1), pp. 40-47. Jonas, O., September 1982, Beware of Organic Impurities in Steam Power Systems, Power, 126, (9). Jonas, O., February 1985a, Steam Turbine Corrosion, Materials Performance. Jonas, O., March 1985b, Control Erosion/Corrosion of Steels in Wet Steam, Power. Jonas, O., 1985c, Design Against Localized Corrosion, Second International Symposium on Environmental Degradation of Materials in Nuclear Power SystemsWater Reactors, Monterey, California.
228
Jonas, O., 1985d, Transport of Chemicals in Steam Cycles, Paper No.245, Corrosion/85, NACE. Jonas, O., 1986, Cost of Corrosion and Scale in Electric Utilities, 1985 Fossil Plant Water Chemistry Symposium: Proceedings, EPRI, Palo Alto, California, CS-4950. Jonas, O., 1987, Corrosion of Steam Turbines, Metals Handbook, 9th Edition, 13, Corrosion, ASM, International, Metals Park, Ohio. Jonas, O., May 1989, Developing Steam-Purity Limits for Industrial Turbines, Power. Jonas, O., 1994, On-Line Diagnosis of Turbine Deposits and First Condensate, 55th Annual International Water Conference, Pittsburgh, Pennsylvania. Jonas, O., 1995, Steam Generation, Corrosion Tests and StandardsApplication and Interpretation, ASTM, West Conshohocken, Pennsylvania. Jonas, O., 1997, Effects of Steam Chemistry on Moisture Nucleation, Moisture Nucleation in Steam Turbines, EPRI, Palo Alto, California, TR-108942. Jonas, O. and Dooley, B., 1996, Steam Chemistry and its Effects on Turbine Deposits and Corrosion, 57th International Water Conference, Pittsburgh, Pennsylvania. Jonas, O. and Dooley, B., 1997, Major Turbine Problems Related to Steam Chemistry: R&D, Root Causes, and Solutions, Proceedings: Fifth International Conference on Cycle Chemistry in Fossil Plants, EPRI, Palo Alto, California, TR-108459. Jonas, O. and Syrett, B. C., 1987, Chemical Transport and Turbine Corrosion in Phosphate Treated Drum Boiler Units, International Water Conference, Pittsburgh, Pennsylvania. Jonas, O., Dooley, B., and Rieger, N., 1993, Steam Chemistry and Turbine CorrosionState-of-Knowledge, 54th International Water Conference, Pittsburgh, Pennsylvania. Jonas, O., et al., 2000, EPRI ChemExpert: Cycle Chemistry Advisor for Fossil Power Plants, EPRI 6th International Conference on Cycle Chemistry in Fossil Plants, Columbus, Ohio. Jonas, O., et al., March 2007, Water Hammer and other Hydraulic Phenomena, Power. Kilroy, R., et al., 1997, A 12% Chrome Weld Repair Increases Stress Corrosion Cracking Resistance of LP Finger Type Rotor Dovetails, Proceedings of the International Joint Power Generation Conference, 2, Power. Kleitz, A., 1994, Erosion and Corrosion Phenomena in Wet Steam Turbines, ESKOM Power Plant Chemistry Symposium. Kramer L. and Michael, S., September 1975, Service Experience and Stress Corrosion of Inconel 600 Bellows Expansion Joins in Turbine Steam Environments, Materials Performance. McCloskey, T., 2002, Troubleshooting Turbine Steam Path Damage, Advances in Life Assessment and Optimization of Fossil Power PlantsProceedings, EPRI, Palo Alto, California, 1006965. McIntyre, P., 1979, U.K. Experimental Work on Turbine Disc and Rotor Cracking: An Overview, Working on SCC in Turbine Rotors and Discs, EPRI-CEGB, Leatherhead, England. NERC, 2002, pc-GAR (Generating Availability Report, Statistical Data 1982-2000), North American Electric Reliability Council. Newman, J., 1974, The Stress Corrosion of Turbine Disc Steels in Dilute Molybdate Solutions and Stagnant Water, CERL Lab Note No. RD/L/N 215/74. Nowak, 1997, Low Pressure Turbine Stress Corrosion Cracking Investigation at the Navajo Generating Station, Proceedings: Steam Turbine Stress Corrosion Workshop, EPRI, Palo Alto, California, TR-108982.
Parkins, R., January 1972, Stress Corrosion Spectrum, British Corrosion Journal, 7. Petr, V., Kolovratnik, M., Jiricek, I., and Jonas, O., 1997, Experimental Investigation of the Effects of Steam Chemistry on Droplet Nucleation, Moisture Nucleation in Steam Turbines, EPRI, Palo Alto, California, TR-108942. Povarov, O., et al., June 1985, Basic Regularities of Erosion of Steam-Turbine Blade Materials under Impingement of Water Droplets, Strojarstvo. Progress in Understanding Improves Outlook for Control of Turbine-Steam Chemistry, March 1981, Power. Pryakhin, V . A., et al., October 1984, Problems of Erosion of the Rotating Blades of Steam Turbines, Teploenergetika. Rezinskikh, V ., Vaiman, A., and Melamed, M., November 1993, The Mechanism of Damaging the Metal of Blades of Steam Turbines Working in the Zone of the Phase Transition, Teploenergetika. Rosario, et al., 2002, Evaluation of LP Rotor Rim-Attachment Cracking Using LPRimLife, Advances in Life Assessment and Optimization of Fossil Power PlantsProceedings, EPRI, Palo Alto, California, 1006965. Ryzenkov, V ., 2000, The State of the Problem and Ways for Increasing the Wear Resistance of the Power Equipment at Thermal Power Stations, Thermal Engineering, 47, (6). Sakamoto, T., Nagao, S., and Tanuma, T., 1992, Investigation of Wet Steam Flow for Steam Turbine Repowering, Steam Turbine-Generator Developments for the Power Generation Industry, ASME Power, 18. Sanders, W., 2001, Steam Turbine Path Damage and Maintenance Volumes 1 (February 2001) and 2 (July 2002), Pennwell Press. Scegljajev, V ., 1983, Parni Turbiny (Steam Turbines), Prague, Czechoslovakia: SNTL. Schleithoff, K., 1984, Influence of Water Steam Chemistry on Stress Corrosion Cracking of Steam Turbine LP Discs, Corrosion in Power Generating Equipment, Plenum Press. Singh, M., Matthews, T., and Ramsey, C., Fatigue Damage of Steam Turbine Blade Caused by Frequency Shift Due to Solid BuildupA Case Study, http://www.dresserrand.com/e-tech/tp110/tp110prt.htm Speidel, M. and Atrens, A., Editors, 1984, Corrosion in Power Generating Equipment, Plenum Press. Speidel, M. and Bertilsson, J., 1984, Stress Corrosion Cracking of Steam Turbine Rotors, Corrosion in Power Generating Equipment, Plenum Press. Stastny, M., Jonas, O., and Sejna, M., 1997, Numerical Analysis of the Flow with Condensation in a Turbine Cascade, Moisture Nucleation in Steam Turbines, EPRI, Palo Alto, California, TR-108942. Svoboda, R. and Faber, G., 1984, Erosion-Corrosion of Steam Turbine Components, Corrosion in Power Generating Equipment, Plenum Publishing Corporation. Syrett, C. and Gorman, J. A., February 2003, Cost of Corrosion in the Electric Power IndustryAn Update, Materials Performance. Syrett, C., Gorman, J. A., Arey, M. L., Koch, G. H., and Jacobson, G. A., March 2002, Cost of Corrosion in the Electric Power Industry, Materials Performance. Turner, D., 1974, SCC of LP Turbines: The Generation of Potentially Hazardous Environments from Molybdenum Compounds, CERL Lab Note No. RD/L/N 204/74.

Turbo Machinery Presentation Collection

Încărcat de

Informații document

Descriere originală:

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Turbo Machinery Presentation Collection

Încărcat de

Drepturi de autor:

Formate disponibile

Table of Contents

and Tim R. Griffin

PROCEEDINGS OF THE THIRTY-FIFTH TURBOMACHINERY SYMPOSIUM 2006

The unit is a Dresser-Rand (Ingersoll-Rand) Model E-148. Design conditions were:

44.14 psia 1200F 775,000 lbs/hr 15.56 psia 24,400 hp

Figure 1. Typical Refinery Cat Cracker Process.

Figure 2. Typical FCC Expander Configuration.

ROOT CAUSE FAILURE ANALYSIS

Figure 4. Flue Gas Flow.

Identification of potential abnormal loading scenarios

PROCEEDINGS OF THE THIRTY-FIFTH TURBOMACHINERY SYMPOSIUM 2006

Mass Flow ( Inlet Temperature) Inlet Pr essure

Figure 10. Rotor with Failed Blade.

Blades 13 and 47Additional blades found with crack near

PROCEEDINGS OF THE THIRTY-FIFTH TURBOMACHINERY SYMPOSIUM 2006

Leading edge Crack initiation site Max camber Trailing edge

PROCEEDINGS OF THE THIRTY-FIFTH TURBOMACHINERY SYMPOSIUM 2006

Table 1. Measure Strain from Bench Test.

Table 2. Calculated Strain from FEA Model.

Three other sets of identical blades operated successfully in this

SECOND PHASE OF THE RCA

PROCEEDINGS OF THE THIRTY-FIFTH TURBOMACHINERY SYMPOSIUM 2006

Monitor the expander for:

Mach number Efficiency Flow function Stack emissions/opacity Vibration

PROCEEDINGS OF THE THIRTY-FIFTH TURBOMACHINERY SYMPOSIUM 2006

and Kengo Hirano

FUEL GAS COMPRESSOR WITH COMMON HEADER

Table of Contents 26 PROCEEDINGS OF THE THIRTY-FIFTH TURBOMACHINERY SYMPOSIUM 2006

COMPRESSOR ADVANCED FLOW CONTROL

Table of Contents 28 PROCEEDINGS OF THE THIRTY-FIFTH TURBOMACHINERY SYMPOSIUM 2006

VERIFICATION OF CONTROL SYSTEM USING DYNAMIC SIMULATOR

Table of Contents 30 PROCEEDINGS OF THE THIRTY-FIFTH TURBOMACHINERY SYMPOSIUM 2006

ACTUAL CONTROL PERFORMANCE

Fx G/T GTCC IGV L/S LSS MV Pd PIC PLC Ps PV Qd SEL SV Td Ts UIC

Table of Contents 32 PROCEEDINGS OF THE THIRTY-FIFTH TURBOMACHINERY SYMPOSIUM 2006

THE ROLE OF ONLINE AERODYNAMIC PERFORMANCE ANALYSIS

and M. Theodore Gresh

MAINTENANCE AND OPERATIONAL IMPROVEMENTS

Table of Contents 56 PROCEEDINGS OF THE THIRTY-FIFTH TURBOMACHINERY SYMPOSIUM 2006

Table 2. Output Data for GB1701 Methanation Compressor.

THE ROLE OF ONLINE AERODYNAMIC PERFORMANCE ANALYSIS

Table of Contents 58 PROCEEDINGS OF THE THIRTY-FIFTH TURBOMACHINERY SYMPOSIUM 2006

30,000 hp axial air compressor blade failure resulting in a 50

THE ROLE OF ONLINE AERODYNAMIC PERFORMANCE ANALYSIS

Table of Contents 60 PROCEEDINGS OF THE THIRTY-FIFTH TURBOMACHINERY SYMPOSIUM 2006

Figure 13. Goodman Diagram for Failed Axial Blade.

Figure 15. Synthetic Natural Gas Compressor Showing Drop in Efficiency.

THE ROLE OF ONLINE AERODYNAMIC PERFORMANCE ANALYSIS

Knowing the machine performance immediately significantly aids

the process of troubleshooting a machine problem and minimizing downtime/loss production.

Table of Contents 62 PROCEEDINGS OF THE THIRTY-FIFTH TURBOMACHINERY SYMPOSIUM 2006

THE CONSEQUENCES OF COMPRESSOR OPERATION IN OVERLOAD

and Jay M. Koch

Table of Contents 64 PROCEEDINGS OF THE THIRTY-FIFTH TURBOMACHINERY SYMPOSIUM 2006

Figure 1. Definition of Overload.

Table of Contents THE CONSEQUENCES OF COMPRESSOR OPERATION IN OVERLOAD 65

CONDITIONS LEADING TO OVERLOAD OPERATION

requirements for an existing compressor or misapplication of a new machine),

Table of Contents 66 PROCEEDINGS OF THE THIRTY-FIFTH TURBOMACHINERY SYMPOSIUM 2006

POTENTIAL CONSEQUENCES OF OVERLOAD OPERATION

Figure 5. Impeller Incidence.

Table of Contents THE CONSEQUENCES OF COMPRESSOR OPERATION IN OVERLOAD 67

Figure 7. CFD Results Showing Pressure Field Around LSD Vanes.