FLUID CONTROL RESEARCH INSTITUTE

OIL AND GAS FLOW MEASUREMENT AND CONTROL TECHNIQUES AND STANDARDS (AUGUST 2 to OCTOBER 30, 2010)

Project Report on

Experimental Study of Performance Characteristics of Turbine Meter in Oil Medium

Prepared by:
1. Onwunyili Christian (NIG)

2. Shungu Alfred George (TAN) 3. Khadjikhanov A.A. (UZB) 4. Mustafa Falah (IRQ) Project Guide:

Mr. P. K. Suresh, FCR, India.

Abstract
Although the liquid turbine meter principle dates back many decades, the axial flow turbine meters presently employed for liquid measurement are quite new. The axial flow turbine meter was first used where accuracy of measurement was not the prime factor, reliability was of greater importance, so parts were made rugged and the rotor was designed to be more non-clogging than accurate. Through time, the turbine flow meter (TFM) has maintained its reliability and ruggedness while attaining a high degree of accuracy. The performance characteristics of TFMs depend greatly on the upstream flow condition, the fluid properties and the geometrical parameters of the rotor. Hence, a systematic study has been carried out to establish the effect of a skewed velocity profile on the lower limit of the linear range, usable range and k-factor (meter factor) of the meter. Alteration of the velocity profile has been achieved by providing a protruding gasket at some pipe diameters upstream of the meter, decreasing the upstream straight length pipe and by subjecting the meter to some atypical pressure conditions. From the result, it can be seen that the average meter factor does not change for a slightly skewed velocity profile but increases drastically for a highly skewed velocity profile. To overcome the skewed velocity profile effect, flow straightener have been provided upstream of the rotor of the TFM which effectively conditions flow within the meters linear range when operated according to standards or/and manufacturers recommendations. The official recognition of the turbine meter by the petroleum industries took place in March 1970 with the publication of API Standard 2534. The publication of this standard has opened the door to use turbine meters for custody transfer of refined products and pipeline systems, tanker, barge, and truck loading or unloading. This has brought the turbine meter to worldwide recognition as a primary register for the exchange of energy. For clarity and ease of understanding, this project report has been divided into five chapters. Whereas chapters one and two lay the background, mainly theoretical, to the subject of Turbine Meter as a flow measuring device, chapter two considers the practical aspects of the subject highlighting the various stipulations of flow measurement standards and the TFM manufacturers recommendations. Chapter four presents the methodology employed in the laboratory for calibration of the flow meter, the results and the result analysis. The report was concluded in chapter five with some recommendations. Onwunyili Chris (NIG) Shungu Alfred George (TAN) Khadjikhanov (UZB)
i

A.A.

Mustafa Falah (IRQ)

Acknowledgement Projects of this kind always receive many different kinds of assistance from many different people. Ours is no exception. We wish to acknowledge a number of these contributions by name, and also give thanks to many others as well. We owe a special debt of gratitude to FCRI, the institute that made this project possible. Special recognition is due Mr. P. K. Suresh of FCRI who in his capacity acted as our project guide and made great deal of contributions to us in the course of this work. We appreciate the assistance of M. Arun, P. Sudheesh, V. R. Gineesh, Vinod and Fahad of FCRI Oil Flow Laboratory. Finally, we appreciate the valuable contributions of all staff of FCRI and our various colleagues, who in the past 3 months shared knowledge with us, leading to the success of this study.

.ITEC 2010 Project Group

ii

CONTENTS Abstract Acknowledgement List of Figures List of Tables Chapter One

iii

i iii iv v v 1

Introduction 1.1 1.2 Why Turbine Flowmeter Aim of the Project 5 5 5 1 3

CHAPTER TWO The Theory of Turbine Meter

2.1 2.2

General description of Liquid Turbine Flow Meter Dynamics of Turbine Flow Meters

8 13 13 13 14 15 16 17 17 18 19 19 19

CHAPTER THREE Practical Considerations

3.1

Meter Performance Performance Indicators Installation Procedure Piping Consideration Strainers/Filters Flow Straighteners Signal Conditioners/Converters

3.2
3.3 3.4 3.5 3.6 3.7

CHAPTER FOUR Laboratory Test

4.1

Equipment Arrangement, Schematic & Operation Sequence

iv

4.2 4.3 4.4

Equipment Data Method of Calibration Results 28

22 22

4.4.1 Result for Test 1 4.4.2 Sample Calculations

28 30 43 44 47 55 56 57 58 58

4.4.3 Result for Test 2 4.4.4 Result for Test 3

4.4.5 Result for Test 4 4.5 4.6 4.7

Summary of Result Discussion of Results and Recommendations Suggestion for further studies

CHAPTER FIVE Conclusion

List of Figures Figure 1. Sectional View of Turbine Meter Figure 2. Signal magnetic pickup coil Figure 3. Velocity Diagram for Turbine Blade Figure 4 Typical Performance Curve of Liquid Turbine Flow meter Figure 5 Schematic Diagram of Oil Flow Laboratory
v

5 8 9 15 20

Figure 6. Set up for Calibration of TFM under normal operating conditions Figure 7. Set up for Calibration of TFM under normal operating conditions Figure 8. Caliberation Curve of TFM for Test 1 Figure 9. Caliberation Curve of TFM for Test 3 Figure 10. Caliberation Curve of TFM for Test 1 Figure 11. Caliberation Curve of TFM for Test 1: Extended to non-linear range Figure 12. Caliberation Curve of TFM for Test 1: Extended to non-linear range Figure 13. Caliberation Curve of TFM: Combined Curves

25 26 29 45 48 50 52 54

List of Tables

Table 1: Datasheet for TFM calibrated under normal operating conditions Table 2: Statistical Calculations Table 3: Calculation of Uncertainty Table 4: Calculation of Flowrates using Best Fit Equation Table 5: Calculation of Uncertainty in Best Fit Equation (F Vs Qa) Table 6: Uncertainty Budget in Flowrate Table 7: Datasheet for TFM calibrated under atypical pressure conditions Table 8: Datasheet for TFM calibrated with a gasket protruding upstream of the meter Table 9: Calculation of Uncertainty for Test 3 44 31 32 37 39 40

27

42 43

Table 10: Datasheet for TFM calibrated with upstream straight length pipe reduced to 10D. Table 11: Calculation of Uncertainty in k
vi

46

47

Table 12: Test 4 Extended to Non-Linear Range Table 13: Test 3 Extended To Non-Linear Range Table 14: Comparison of Tests Results References 58

49 51 53

vii

CHAPTER ONE Introduction

Many factors influence the quality of turbine meter flow measurement. Recognizing these factors allows a user to optimize the flow measurement performance of a turbine metering system. In order to ensure that turbine meters in the field are operating at their best, turbine metering stations must be properly designed, calibrated, installed, and maintained. This project explains the working principle of liquid turbine meters and the physical processes that affect the accuracy of turbine metering systems. Laboratory test was carried out using FCRI Oil Test Facilities and the results are presented here to explain some of these phenomena.

1.1

Why Turbine Flowmeter?

Petroleum products bought and sold on the worldwide market may be transported over thousands of miles and change ownership many times from the well head to the end user. Each time the product changes ownership, a custody transfer is completed and both buyer and seller expect their asset share to be accurately measured. The dynamic measurement provided by meters is a convenient and accurate means to measure valuable petroleum products. Selecting the right meter for the job with a high level of confidence is imperative to ensure accurate measurement at the lowest cost of ownership. Typical petroleum applications where measurement is required include: production, crude oil transportation, refined products transportation, terminal loading, fuel oil tank truck loading and unloading, aviation and lube oil blending. Each of these applications is unique and a specific type of meter may be better suited for each application. Selecting the correct meter for a specific measurement task is dependent on the following operating conditions:

System characteristics Pressure and temperature are typically specified but other characteristics such as pulsating flow from a PD pump or valve operation /location should also be considered as they may cause measurement errors for some types of meters. Product characteristics The basic product characteristics of viscosity, specific or API gravity, chemical characteristics and lubricating quality must be specified. Also, any contaminates such as particulates, air or water contained in the product must be identified and noted in an application analysis. Flow range This is the minimum and maximum flow rate over which the meter will operate. The flow range can also be expressed as the turndown range, which is the ratio of the maximum to the minimum flow rate (e.g., a flow range of 10 bph to 100 bph is a 10:1 turndown range). Viscosity Range Just as the flow range can be expressed as a turndown range, the maximum to the minimum viscosity be expressed as a turndown range. Meter Accuracy Requirements and Criteria: Accuracy requirements for the wholesale and retail trade are normally defined by the weights and measures regulations in the country or jurisdiction in which the sale is conducted. Sales within the petroleum industry that are not normally defined by weights & measures, but by a contract between the trading parties, are known as Custody Transfer transactions. A typical contract may define a specific measurement standard such as one of the American Petroleum Industry (API) Standards. Currently API recognizes four types of dynamic measuring devices Positive Displacement (PD) Meters, Turbine Meters, Coriolis Mass Flow Meters (CMFMs) and recently approved Liquid Ultrasonic Flow Meters (LUFMs).Contracts are also based on other recognized standards but all these standards have one thing in common they all strive to minimize measurement error for a specific application. Error is defined as the difference between the measured quantity and the true value of the quantity. The criteria associated with custody transfer and all accurate measurement includes:

 Repeatability the variation of meter factor under stable operating conditions, i.e., constant flow rate, temperature, pressure, and viscosity. The typical requirement is that a meter must repeat within +/- 0.05% at 95% confidence level. Linearity the variation of meter factor over a flow range at constant temperature, pressure and viscosity. Stability or Reproducibility the variation of meter factor over time. Unlike repeatability runs where conditions can be kept nearly constant, operating conditions over time may have wider variations. Therefore, it is important that the meter selected have minimum sensitivity to operational variations to achieve required accuracy. On-site Verification or proving has always been, and remains, fundamental to custody transfer measurement. It is the only sure method to determine and correct for both constant and variable systemic errors.

Turbine Flowmeters (TFMs) have been used extensively for custody transfer of refined petroleum products for over 30 years. Significant advantages associated with the use of turbine flowmeters, in lieu of other metering principles, make increased future use inevitable. The TFMs are known for their excellent repeatability, higher accuracy, wide operating range, high shock capability, low pressure loss and wide temperature and pressure limits. Probably the two main advantages of the turbine meter over conventional differential head devices are: 1. The extended and more accurate registration of flow in the low flow range of operation, which results from the registration being proportional to the velocity rather than the velocity squared. 2. The comparatively low head loss across the meter. The turbine meter has several advantages over the conventional positive displacement (P.D.) meters. The turbine can handle two to three times the flow of its equivalent size P.D. meter,
3

resulting in a cost savings. Second, the turbine, because of its inherent design, can withstand severe service such as sand, over-ranging, and salt water, with less maintenance and better life. In general, other advantages include a wide range of fluid applications, fast response and convenient readout for control purposes. Each application must be considered on its own merit, and economic consideration between the turbine and other metering is a necessity. Depending on the type of read-out and the control required, the turbine meter system compares favorably with other metering systems. However, where maximum accuracy, wide operating range, repeatability, and convenient control is needed, the turbine meter must receive the primary consideration. 1.2

Aim of the Project

The performance characteristics of TFMs depend greatly on the upstream flow condition, the fluid properties and the geometrical parameters of the rotor. In other words, meter performance may be affected by conditions such as: flow rate, viscosity of the liquid, temperature of the liquid, density of the liquid, pressure of the flowing liquid, cleanliness and lubricating qualities of the liquid, foreign material lodged in the meter or flow-conditioning element, changes in mechanical clearances or blade geometry due to wear or damage, changes in piping, valves, or valve positions that affect fluid profile or swirl and conditions of the prover. It is therefore imperative to study the performance of turbine meter subjected to different calibration conditions in order to become familiar with fundamental characteristics and conditions surrounding the liquid turbine flowmeter in order to better understand its usage. In this project titled Experimental Study of Performance Characteristics of Turbine Meter in Oil Medium, a liquid turbine meter was subjected to varying degrees of upstream disturbance under different flow ranges to determine the effects of resulting swirls and jetting on the its performance. The turbine meter was also subjected to different operating pressures to study pressure effects on its performance.

CHAPTER TWO The Theory of Turbine Meter

2.1

General description of Liquid Turbine Flow Meter

Figure 1. Sectional View of Turbine Meter

. It consists of multiple- bladed, free spinning, permeable blade rotor housed in a non magnetic stainless steel body. In operation, the rotating blades generate a frequency signal proportional to the liquid flow rate which is sensed by the magnetic pick up and transferred to the read out indicator The basic construction of the turbine flowmeter incorporates a bladed turbine rotor installed in a flow tube. The rotor is suspended axially in the direction of flow through the tube. The turbine
6

flowmeter is a transducer, which senses the momentum of the flowing stream. The bladed rotor rotates on its axis in proportion to the rate of the liquid flow through the tube. Turbine Rotation As the liquid product strikes the front edge of the rotor blades, a low-pressure area is produced between the upstream cone and the rotor hub. The blades of the turbine rotor will tend to travel toward this low-pressure area as a result of this pressure differential across the blades. The pressure differential (or pressure drop) constitutes the energy expended to produce movement of the rotor. The initial tendency of the rotor is to travel downstream in the form of axial thrust. But since the rotor is restrained from excessive downstream movement, the only resulting movement is rotation.Fluid flowing through the meter impacts an angular velocity to the turbine rotor blades, which is directly proportional to the linear velocity of the liquid. The degree of the angular velocity or number of revolutions per minute of the turbine rotor is determined by the angle of the rotor blades to the flowing stream of the approach velocity. With axial thrust forcing the turbine rotor downstream, the friction resulting from contact between the turbine rotor and the downstream cone would cause excessive wear if there were not some means of balancing the turbine rotor on its axis between the upstream and the downstream cone. Bernoulli's Principle states that when flow velocity decreases, the static pressure increases. Therefore, a high-pressure area exists at the downstream side of the turbine rotor exerting an upstream force on the rotor. As a result, the turbine rotor is hydraulically balanced on its axis. Signal Output Electrical output is generated using the principle of reluctance. A pickup coil, wrapped around a permanent magnet, is installed on the exterior of the flow tube or the meter body immediately adjacent to the perimeter of the rotor (Figure 1). The magnet is the source of the magnetic flux field that cuts through the coil. Each blade of the turbine rotor passing in close proximity to the pickup coil causes a deflection in the existing magnetic field. This change in the reluctance of the magnetic circuit generates a voltage pulse within the pickup coil. Each pulse generated represents a discrete amount of volumetric throughput. Dividing the total number of pulses generated by the specific amount of liquid product that passed through the
7

turbine flowmeter determines the K-Factor. The K-Factor, expressed in pulses per unit volume, may be used with a factoring totalizer to provide an indication of volumetric throughput directly in engineering units. The totalizer continuously divides the incoming pulses by the K-Factor (or multiplies them with the inverse of the K-Factor) to provide factored totalization. The frequency of the pulse output, or number of pulses per unit time, is directly proportional to the rotational rate of the turbine rotor. Therefore, this frequency of the pulse output is proportional to the rate of the flow. By dividing the pulse rate by the K-Factor, the volumetric throughput per unit time of the rate of flow can be determined. Frequency counters or converters are commonly used to provide instantaneous flow rate indication. Plotting the electrical signal output versus flow rate provides the characteristics profile or calibration curves for the turbine flowmeter. Electrical output is also generated using the principle of inductance. A pickup coil is installed on the exterior of the flow tube immediately adjacent to the perimeter of the turbine rotor. The magnetic source of the flux field in this type of output is either the rotor itself or small magnets installed in the rotor. In the case of the rotor, the material of construction would be nickel or some other easily magnetized flux field. The results are identical to that of the reluctance principal.

Figure 2. Signal magnetic pickup coil

2.2

Dynamics of Turbine Flow Meters

There are two approaches described in the current literature for analyzing axial turbine performance. The first approach describes the fluid driving torque in terms of momentum exchange, while the second describes it in terms of aerodynamic lift via airfoil theory. The former approach has the advantage that it readily produces analytical results describing basic operation, some of which have not appeared via airfoil analysis. The latter approach has the advantage that
9

it allows more complete descriptions using fewer approximations. However, it is mathematically intensive and leads rapidly into computer generated solutions. In the following, we have used the momentum exchange approach to highlight the basic concepts of the axial turbine flowmeter.

Blade

Figure 3. Velocity Diagram for Turbine

In a hypothetical situation, where there are no forces acting to slow down the rotor, it will rotate at a speed which exactly maintains the fluid flow velocity vector at the blade surfaces.

When one introduces the total flow rate this becomes:

10

where

is the 'ideal' rotational speed, Q is the volumetric flow rate, A is the area of the annular

flow cross section and is now the root-mean-square of the inner and outer blade radii, (R, a). Eliminating the time dimension from the left hand side quantity reduces it to the number of rotor rotations per unit fluid volume, which is essentially the flowmeter K factor specified by most manufacturers. Hence, according to Eq. (2), in the ideal situation the meter response is perfectly linear and determined only by geometry. (In some flowmeter designs the rotor blades are helically twisted to improve efficiency. This is especially true of blades with large radius ratios, (R/a). If the flow velocity profile is assumed to be flat, then the blade angle in this case may be described by tan =constant X r. This is sometimes called the 'ideal' helical blade.) In practice, there are instead a number of rotor retarding torques of varying relative magnitudes. Under steady flow the rotor assumes a speed which satisfies the following equilibrium: Fluid driving torque = rotor blade surfaces fluid drag torque + rotor hub and tip clearance fluid drag torque + rotation sensor drag torque + bearing friction retarding torque 3 Referring again to figure 3 , the difference between the actual rotor speed, r , and the ideal rotor speed, , is the rotor slip velocity due to the combined effect of all the rotor retarding torques as described in Eq. (3), and as a result of which the fluid velocity vector is deflected through an exit or swirl angle, . Denoting the radius variable by r, and equating the total rate of change of angular momentum of the fluid passing through the rotor to the retarding torque, one obtains:

which yields:

11

where is the fluid density and NT is the total retarding torque. Combining Eqs. (1) and (4) and rearranging, yields:

The trends evident in Eq. (5) reflect the characteristic decline in meter response at very low flows and why lower friction bearings and lower drag pickups tend to be used in gas versus liquid applications and small diameter meters. In most flowmeter designs, especially for liquids, the latter three of the four retarding torques described in Eq. (3) are small under normal operating conditions compared with the torque due to induced drag across the blade surfaces. As shown in figure 2, the force, F, due to this effect acts in a direction along the blade surface and has a magnitude given by:

where CD is the drag coefficient and S is the blade surface area per side. Using the expression for drag coefficient corresponding to turbulent flow , this force may be estimated by:

where Re is the flow Reynolds number based on the blade chord shown as dimension c in figure 2. Assuming is small compared with , then after integration, the magnitude of the retarding torque due to the induced drag along the blade surfaces of a rotor with n blades is found to be:

Combining Eqs. (7) and (5), and rearranging yields:

12

Eq. (8) is an approximate expression for K factor because it neglects the effects of several of the rotor retarding torques, and a number of important detailed meter design and aerodynamic factors, such as rotor solidity and flow velocity profile. Nevertheless, it reveals that linearity variations under normal, specified operating conditions are a function of certain basic geometric factors and Reynolds number. These results reflect general trends which influence design and calibration. Additionally, the marked departure from an approximate (actually via Re dependence of the fluid drag retarding torque on flow properties under turbulent flow, to other relationships under transitional and laminar flow, gives rise to major variations in the K factor versus flow rate and media properties for low flow Reynolds numbers. This is the key reason why axial turbine flowmeters are generally recommended for turbulent flow measurement

13

CHAPTER THREE Practical Considerations

3.1

Meter Performance

Meter performance is defined by how well a metering system produces, or can be made to produce, accurate quantity measurement. The overall performance of measurement by a meter depends on the condition of the meter and its accessories, the temperature and pressure corrections, the proving system, the frequency of proving, and the variations between operating and proving conditions. The inherent accuracy of a meter is often published in the manufacturers specification, and may be expressed as repeatability and/or linearity. In other words, accuracy is based on how repeatable and how linear the meter can stay within the manufacturers performance specifications. Manufacturers specifications are based on meter operation within recommended flow ranges, within a narrow range of pressures, temperatures, and fluid viscosities. For custody transfer applications, meters with the highest inherent accuracy should be used and should be proved on site. The meters should operate within the manufacturers specifications. An excellent indicator of how well a meter performs is the development of, and history of, its meter factor from proving the meter. A meter factor obtained for one set of conditions will not necessarily apply to a changed set of conditions. Meter performance curves can be developed from a set of proving results. The following conditions may affect the meter factor: a. Flow rate. b. Viscosity of the liquid. c. Temperature of the liquid.
14

d. Density of the liquid. e. Pressure of the flowing liquid. f. Cleanliness and lubricating qualities of the liquid. g. Foreign material lodged in the meter, strainer or flow-conditioning element. h. Changes in mechanical clearances or internal geometry due to wear or damage. i. Changes in piping, valves, or valve positions that affect fluid profile or swirl into a turbine meter. j. Conditions of the prover.

3.2 Performance Indicators

Accuracy The accuracy of a turbine flowmeter is derived from its output (electrical or mechanical) and is the measure of the deviation of an indicated measurement from the referenced standard. The accuracy must include the error associated with the calibration standard. In the India, the National Accreditation Board for Testing and Calibration Laboratories represents the flow standard. Linearity is the variation of the flowmeter K-factor from a nominal value of a point on a curve. Normally during calibration, a value is chosen which makes linearity fall in line with accuracy. Linearity may remain constant during meter life although the absolute accuracy level has changed. Repeatability is the ability of a turbine flowmeter to reproduce its output indefinitely under constant operating conditions at any point over its specified operating range.

15

Figure 4 Typical Perfomance Curve Of Liquid Turbine Flowmeter 3.3

Installation Procedure

Upon receipt of the turbine flowmeter a visual inspection should be performed checking for any indications of damage which may have occurred during shipment. Inspect all packing material carefully to prevent the loss of meter parts or auxiliary components which may have been packed with the shipment. The meter housing is marked by a flow direction arrow and the inlet is marked IN and the outlet is marked OUT. The meter must be installed in the piping in the correct orientation to ensure the most accurate and reliable operation. Care should be taken in the proper selection of the mating fittings. Size, type of material, and pressure rating should be the same as the flowmeter supplied. The correct gaskets and bolts should be utilized. The flowmeter may be installed horizontally or vertically for liquid service without affecting the meter calibration. When it is expected that flow will be intermittent, the meter should not be mounted at a low point in the
16

piping system. Solids which settle or congeal in the meter may affect meter performance. In order to achieve optimum electrical signal output from the flowmeter, due consideration must be given to its isolation from ambient electrical interference such as nearby motors, transformers, and solenoids.

3.4 Piping Consideration

As stated in the Principle of Flowmeter Operation, the fluid moving through the flowmeter engages the vaned rotor. Swirl present in the fluid ahead of the meter can change the effective angle of engagement and, therefore, cause a deviation from the supplied calibration (performed under controlled flow conditions). Turbine meters are constructed with flow straighteners to minimize the effects of fluid swirl and non-uniform velocity profiles is adequate for most installations. However, it is good practice to maintain a minimum straight run of pipe approximately 10 pipe diameters ahead of the inlet and 5 pipe diameters following the outlet. Proper installation of the flowmeter minimizes the negative effects of fluid swirl. Blocking and Bypass valves should be installed if it is necessary to do preventive maintenance on the flowmeter without shutting down the flow system. The Bypass valve can be opened before the Blocking valves allowing the flow to continue while removing the turbine flowmeter for service. All flow lines should be purged prior to installing the meter.To prevent possible damage to the meter, install the meter ONLY in flow lines that are clean and free of debris. Upon initial start-up of the system a spool piece should be installed in place of the flowmeter so that purging of the system can be performed to remove all particle debris which could cause damage to the meter internals. In applications where meter flushing is required after meter service, care should be taken as to not overspeed the meter, as severe meter damage may occur. Avoid over-spinning the meter. Over-spinning the meter may cause damage to the meter internals and lead to needless meter failure.
17

To maintain an accurate flow measurement it is necessary to maintain a downstream pressure sufficient to prevent flashing/cavitation. Flashing of the liquid will result in an indication of flow significantly higher than the actual flow. In order to eliminate this condition adequate downstream pressure must be maintained. The minimum required downstream pressure may be calculated using the following equation: Minimum Pressure =(2 x Pressure Drop) + (1.25 x Vapor Pressure) Downstream pressure may be maintained by a downstream valve that provides the necessary downstream pressure to prevent flashing/cavitation in the metering run. 3.5 Strainers/Filters Turbine flowmeters are designed for use in a clean fluid service. However, the service fluid may carry some particulate material which would need to be removed before reaching the flowmeter. Under these conditions a strainer/filter may be required to reduce the potential hazard of fouling or damage that may be caused by foreign matter. If a strainer/filter is required in the system, it should be located upstream of the flowmeter taking care that the proper minimum distance is kept between the strainer and flowmeter.

3.6 Flow Straighteners

Proper application of the Turbine Flowmeter requires a suitable piping section to achieve optimum accuracy. While an inlet straight pipe run of 10 pipe diameters and an outlet straight pipe run of 5 pipe diameters provide the necessary flow conditioning in general, some applications require an upstream flow straightener. This consists of a section of piping that contains a suitably dimensioned and positioned thin walled tube cluster to eliminate fluid swirl. The term swirl is used to describe the rotational velocity or tangential velocity component of fluid flow in a pipe or tube. Depending on its degree and direction, swirl will change the angle of attack between the fluid and the turbine rotor blades, causing a different rotor speed at a constant flow rate to non-swirling conditions at the same flow rate. Liquid swirl and non-uniform velocity profiles may be introduced upstream of the turbine flowmeter by variations in piping configurations or projections and
18

protrusions within the piping. Swirl may be effectively reduced or eliminated through the use of sufficient lengths of straight pipe or a combination of straight pipe and straightening vanes installed upstream of the turbine flowmeter.

3.7 Signal Conditioners/Converters

Consideration should be given to properly interface the turbine flowmeter output to the host electronics. If the system is installed in an electrically noisy area or if the distance from the turbine flowmeter to the host electronics exceeds 500 feet, a signal conditioner may be necessary. Signal Conditioners for the turbine flowmeter provide amplification, filtering, and wave shaping of the low level flowmeter pickup signal and generate a high level pulse output signal suitable for transmission to a remote host system through a noisy environment. The conditioned pulse output signal may be transmitted several thousand feet.

19

CHAPTER FOUR Laboratory Test 4.1 Equipment Arrangement, Schematic & Operation Sequence

The tests were conducted at FCRI Oil Flow Calibration Laboratory. The layout of the laboratory is shown as figure 4. It consists of a flow source (centrifugal pumps), flow conditioners, pipe lines, test section for the Meter-Under-Test installation, valves for changing the flow, collection tanks, weigh scales, and a timer. The facility is located above an oil reservoir that has a capacity of approximately 600 m3. Oil flow is produced and maintained in the system by a total of four pumps: three variable velocity pumps and one constant velocity pump. One of the variable velocity pumps is driven by 120Hp electric motor while each of the other pumps is driven by 50 Hp electric motors. Maximum of three pumps is put in use at a time. A manifold splits the flow into three separate test section pipelines and a bypass of 8 diameter, with a maximum line size of 200mm. Downstream of the manifold, each pipeline has a flow conditioner that delivers a symmetric, fully developed turbulent velocity profile to the flow meter in the test section. Flow at the test section is controlled by some sets of valves. One set is located upstream near the pump, a main valve for each pipeline and a bypass throttle valve that controls the amount of oil returned to the reservoir without passing through the meter test section. The other valves on each pipeline (located downstream of the test section) are the fine and coarse controls for setting the oil flow rate and the pressure in the test section of the OFCF. The facility makes use of static gravimetric calibration system for measurement and the calibration is performed by "Standing start and stop" method. All measurements made at FCRI are traceable to Indian National Standards.
20

The medium of flow is EXXSOL D80, with a kinematic viscosity of 2.06 Centistokes at 250C and a specific gravity of 0.78. It is also possible to change the medium of calibration to either Hydrol 100 or Hydrol 220 for flow rates up to 100 m3/h (450 GPM). The kinematic viscosity of Hydrol 100 and Hydrol 220 are 100 centistokes and 200 centistokes respectively at 40 0C. This multi viscous facility can be used to generate "Universal viscosity curves" for turbine flow meters up to 4 " NB.

21

8" RESERVOIR 3 (30 m )

8"
H

8" 12" 6"

H

DRAIN

Reservoir for 200 cSt/100 cSt oil

8"
H H

12" 3" 6"

H H

6"

LCP

HCP

HCP

HCP

BY-PASS 8"

BELLMOUTH ENTRY

4"
H H H H H H H

6"

6" 8"

TFM

6"

3"
M

6" 8"
H

6"
H

6" 8" 8" FILTER 8"

8" TFM
H

8" 10TON WEIGHING TANK

ON/OFF VALVE

8"
ON/OFF VALVE H H

4" 8" 4" (200mm) (TEST LINES) (100mm) 8"

4"
H

4" 2TON
WEIGHING TANK

4"
M

4"

4"
H

STEELYARD

STEELYARD

4"

Densitometer

- GATE VALVE;

- CONTROL VALVE;

- CHECK VALVE;

- MANUAL ACTUATOR;

- PISTON ACTUATOR;

- ROTARY MOTOR ACTUATOR;

TFM - TURBINE FLOWMETER; LCP - LOW CAPACITY PUMP (50m^3 /h,5bar);HIGH CAPACITY PUMP (200 m^3 /h, 5bar)

Figure 5

Schematic layout of Oil flow laboratory

22

4.2

Equipment Data
: : : : : : : : Turbine flowmeter 2'' NB HO-2X2-15-225-B-2M-F2SS 20307 Hoffer Flow 10 60 m3/hr EXXSOL D80 27.09.2010

Details of Flow Meter under Test FLOW ELEMENT SIZE MODEL TAG NUMBER MAKE FLOW RANGE MEDIUM OF CALIBRATION DATE OF CALIBRATION

Specification of Reference Instruments used Instrument Weighing system Timer DensitoMeter Pulse counter Temperature Pressure gauge Rang e 300 1000 1000 1000 0 20 50 10.0 kg sec kg/m ^3 Hz C bar Uncertaint y 7.60Ek 03 g 9.61E04 sec 1.00Ekg/m 01 ^3 1.40E03 Hz 2.00E01 C 5.80E02 bar
23

Calibration Due 13.11.2010 05.05.2010 07.09.2011 18.11.2010 20.05.2010 07.05.2010

Readabi lity 0.00100 0 0.00001 0 0.01000 0 0.00001 0 0.01000 0 1.00000 0

4.3 Method of Calibration

Test 1: Caliberation of 2 Hoffer Flow Turbine Flow Meter (TFM) under normal operating conditions Installation Conditions: 20D Length of straight pipe upstream of TFM, 10D Length of straight pipe downstream of TFM, with Flow Straightener. Operating Pressure: 2bar (200kPa) Test Set-up: The Turbine Flow Meter under test was installed in the standard test line of FCRI Oil Flow Laboratory using suitable reducers and expanders as shown in Fig 6. The meter was provided with suitable straight-lengths upstream and downstream to ensure uniform flow. Standing start and stop method was employed for the calibration. A universal Counter was used to determine the time of collection of oil. The upstream pressure of the meter was measured using a precision pressure gauge. The density of the working fluid was determined using an Online Densitometer. Procedure: The oil flow line was flooded and the entrapped air cleared using air bleeds. The ON/OFF valve in the system was kept open and the flow through the TFM was adjusted to the required flow rate by controlling the electrically/manually operated control valve. As soon as the flow rate was adjusted as per the requirement, the ON/OFF valve was closed. The initial mass of the weighing tank was determined as W1. Then the ON/OFF valve was opened to allow flow through the meter. After collecting a sufficient quantity of oil in the tank, at sufficient time, the ON/OFF valve was closed and final mass of the weighing tank recorded. The readings of the pulse counter, Online Densitometer, Online Temperature Indicator and Universal Counter were also recorded. This procedure was repeated for other flow rates covering the entire range of 10-60m3/hr, and tests were carried out two times for a particular flow rate.
24

NB: As per ISO 4185-1980 section 4.1, in order to eliminate the effect of residual liquid likely to have remained in the bottom of the weighing tank or adhering to the walls, a sufficient quantity of liquid was first discharged into the tank (or left at the end of draining after the preceding measurement) to reach the operational threshold of the weighing machine The results of this test is depicted in Table 1. Test 2: Calibration of 2 Hoffer Flow Turbine Flow Meter (TFM) under atypical pressure conditions. Installation Conditions: 20D Length of straight pipe upstream of TFM, 10D Length of straight pipe downstream of TFM, without Flow Straighteners. Operating Pressure: 1.5bar (150kPa) and 3.6bar (360kPa) Test Set-up: The test set-up is the same as that of Test 1. However, this test was carried out at an operating pressure of 1.5 bar and then 3.6bar instead of at 2bar which is the normal operating condition. Procedure: Same as Test 1 The results of Test 2 is depicted in Table 7.

Test 3: Calibration of 2 Hoffer Flow Turbine Flow Meter (TFM) with a gasket protruding upstream of the meter. Installation Conditions: 20D Length of straight pipe upstream of TFM, 10D Length of straight pipe downstream of TFM, Gasket protruding upstream of TFM, without Flow Straighteners. Operating Pressure: 2bar (200kPa) Test Set-up: The test set-up is the same as that of Test 1 but with a gasket made to protrude into the liquid stream upstream of the TFM. The test was carried without a Flow straightener.
25

Procedure: Same as Test 1 The result of Test 3 is depicted in Table 8.

Test 4: Calibration of 2 Hoffer Flow Turbine Flow Meter (TFM) with upstream straight length pipe reduced to 10D. Installation Conditions: 10D Length of straight pipe upstream of TFM, 10D Length of straight pipe downstream of TFM, without Flow Straighteners. Operating Pressure: 2bar (200kPa) Test Set-up: The test set-up is the same as that of Test 1. However, the length of straight pipe upstream of TFM was reduced to 10D and suitable reducers and expanders used as shown in Fig 7. The Flow straightener was also removed. Procedure: Same as Test 1 The result of Test 4 is depicted in Table 10.

26

Ref TFM - REFERENCE TURBINE FLOW METER RTD - RESISTANCE TEMP. DETECTOR CV GV ON/OFF FROM RESERVOIR R,E M - CONTROL VALVE - GATE VALVE ON/OFF VALVE

- REDUCER/EXPANDER - MANIFOLD

GOOSE NECK ASSY. R

Filter

PUMP
Ref TFM

6"

ON/OFF M 2TON WEIGHING TANK CV RTD GV 4"

>10D

>20D 4" GV RTD Turbine flow meter M

CV ON/OFF

(Under calibration)

Container 30kg weighing system

Figure 6. Set up for Calibration of TFM under normal operating conditions

27

Ref TFM - REFERENCE TURBINE FLOW METER RTD - RESISTANCE TEMP. DETECTOR CV GV - CONTROL VALVE - GATE VALVE ON/OFF VALVE FROM RESERVOIR R,E - REDUCER/EXPANDER ON/OFF M - MANIFOLD

GOOSE NECK ASSY. R

Filter

PUMP
Ref TFM

6"

ON/OFF M 2TON WEIGHING TANK CV RTD GV 4"

>10D

>10D 4" GV RTD Turbine flow meter M

CV ON/OFF

(Under calibration)

Container 30kg weighing system

Figure 7. Set up for Calibration of TFM with 10D upstream straight length pipe

4.4

Results
28

4.4.1 Result for Test 1

Table 1: Datasheet for 2 Hoffer Flow Turbine Flow Meter (TFM) calibrated under normal operating conditions. FLOW ELEMENT SIZE MODEL TAG NUMBER MAKE FLOW RANGE MEDIUM OF CALIBRATION DATE OF CALIBRATION Pu Sl. p W1 W2 No kP . a kg kg 1 2 3 4 20 0 20 0 20 0 20 0 141.0 5 117.2 5 98.45 151.6 5 767.20 721.30 1049.1 0 1105.0 5 : Turbine flowmeter 2'' : NB HO-2X2-15-225-B-2M-F2SS : 20307 : Hoffer Flow 10m^3/ : 60 hr : EXXSOL D80 : 27.09.2010 Densi T ty deg. C kg/m3 28.3 5 28.4 9 28.4 7 28.5 3 786.0 9 785.9 9 786.0 1 785.9 6
29

t sec.

Qa m^3/ hr 38.06 7 37.79 5 59.09 2 59.13 0

Va litres 797.637 769.583 1211.133 1214.714

N pulse s 3078 2 2969 6 4686 9 4701 0

K p/litre 38.59 15 38.58 72 38.69 85 38.70 05

F Hz. 408.07 405.12 635.21 635.65

75.43333 1 73.30259 3 73.78516 4 73.95566 0

5 6 7 8

20 0 20 0 20 0 20 0

106.0 0 186.7 0 113.1 0 105.3 0

916.00 821.65 607.25 603.95

122.3725 12 96.19345 6 161.0730 50 163.0947 77

29.0 5 29.0 9 29.3 3 29.3 8

785.5 9 785.5 8 785.4 0 785.3 7

30.37 4 30.29 0 14.08 1 14.03 4

1032.496 809.373 630.039 635.800 Average KFactor Linearity(%) Repeatability (%)

4001 5 3119 1 2428 6 2450 3

38.75 56 38.53 73 38.54 68 38.53 88

326.99 324.25 150.78 150.24 ppl % %

38.61 = 95 0.282 = 7 0.006 = 0

30

70.000 65.000
4.4.2 Sample Calculations
The following sample calculations are made from SI. No 5 of Table 1 Data: Pup W1 W2 t = = = = 200 106.000 916.000 122.372512

60.000

kPa kg kg s

T N

= = =

29.05 785.59 40015

deg.C kg/m3 pulses

Buoyancy Correction Factor, B =1 + = 1 + a (1/ o-1/ m) = 1+ 1.128((1/ 785.59)-(1/ 8000)) 1.001381 Flow rate: Qa= (W2-W1) X B X 3600X1000/(t X o) = (926-106) x1.001381x3600x1000/ (122.372512x785.59) = 30374.354 lph Volume in liters, Va= (W2-W1) X B X 1000 / o = (926-106) x1.001381x1000/ 785.59 =1032.496 ltrs. Frequency, F = N / t = 40015/122.37251 = 326.99 Hz K-Factor, k = N/Va = 40015 / 1032.496 p/litre

Linearity in % = (Kmax-Kmin) x 100/ (2xKavg) From Table Kmax Kmin Kavg = = = 38.756 38.537 38.620 p/litre p/litre p/litre

Linearity = (38.756-38.537) x100 / (2* 38.620)

32

= 0.283% Repeatability= n-1 x 100 / ym x n1/2 The repeatability was determined for the operating range as = 0.086 x100/ 38.6195 x 81/2 = 0.1573 % Table 2: Statistical Calculations: From Sl. No. 5 of Table 1. Pup W1 W2 time Temperature Density N K-factor Error no. of points erlp(x) Repeatability Readability of N Uncertainty in Buoyancy Readability of weigh system Uncertainty in Weigh System Uncertainty in Timer Uncertainty in Density Uncertainty in DAS Uncertainty of pulse counter 200 106.000 916.000 122.372512 29.05 785.59 40015 38.756 8 0.0449 0.157 1.000000 0.001381 0.001000 kg kPa kg kg sec deg C kg/m ^3 pulse p/litre p/litre p/litre Pulse

0.007600 kg 9.61E-04 sec kg/m 0.100 ^3 0.001400

33

Calculation of Uncertainty in k-factor If we denote Qa as x and K as y then from Table 1 we have: x 38.067 37.795 59.092 59.130 30.374 30.290 14.081 14.034 282.8634 54 y 38.591 38.587 38.698 38.700 38.756 38.537 38.547 38.539 308.956 071 xy 1469.048 1458.415 2286.752 2288.344 1177.176 1167.310 542.795 540.856 10930.69 6795 x^2 1.449E+03 1.428E+03 3.492E+03 3.496E+03 9.226E+02 9.175E+02 1.983E+02 1.970E+02 12101.035 521 (y-ABx)^2 1.34E-03 1.61E-03 1.44E-05 3.22E-05 2.31E-02 4.38E-03 2.80E-05 1.73E-04 0.03064 1

Table. 3 : Calculation of Uncertainty Sl Qa K m^3/h No. r p/litre 1 38.067 38.591 2 37.795 38.587 3 59.092 38.698 4 59.130 38.700 5 30.374 38.756 6 30.290 38.537 7 14.081 38.547 8 14.034 38.539 xm ym
34

erl(x) p/litre 0.02562 0.02555 0.04482 0.04486 0.02643 0.02647 0.04171 0.04177

35.3579 38.620 Sy Sxx 0.07146 2099.6 Standard Uncertainty of best fit, Us Expanded Uncertainty UE Xm = Ym = n= n*Xm*Ym = (xy)-n*Xm*Ym = x^2 = B= A= (y-A-Bx)^2 = n-2 = Sy = Sxx = S= .(xy) =

8 A 3.851E+01 = = 0.0449 0.1356 35.36 38.6195 8.00 10924.05 6.65 12101 3.17E-03 38.5075 3.06E-02 6.00 0.0715 2.10E+03 0.05 10930.7

B 3.167E-03 p/litre p/litre

B= (xy)-n*Xm*Ym / ( x2- n Xm2) = 6.65/[12101-(8X35.362)] = 0.00316923

A= Ym-BXm = 38.6195-(0.00316923X35.36)
35

= 38.5075 Sy= [(y-A-Bx)^2/(n-1)]^0.5 = (0.030641 /6)^0.5 = 0.071462 Sxx = x2 nXm2 = 12101-(8*35.362) = 2100 erl(x) = Sy[(1/n)+((Xi-Xm)^2/Sxx)]^0.5 = 0.0715[(1/8)+(( 30.374 -35.36)^2/2100)]^0.5 =0.02643 p/litre Standard Uncertainty of best fit, Us = Maximum value of erl(xi) for i = 1 to n

UNCERTAINTY BUDGET In order to evaluate the Expanded Uncertainty,UE, we need to calculate the partial derivaties of our calculated quantities, Q and k, (Sensitivity Coefficient) with respect to each of the sources of uncertainty so as to know their respective Uncertainty contribution. Sample calculations in this regard are as follows: Recall that Q=(1+)W/(*t)=W{1+ a (1/-1/wt)}/( p*t) , and K=N/(Q *t) Q/a=W(1/-1/wt)/(p*t) = 9.6721*10-6 Q/wt=W a (1/-1/wt)/(p*t* wt2)= 1.4850*10-10 Q/=W(-1-2(a /)- a /wt)/(p2*t)= -1.0755*10-5 Q/ W ={1+ a (1/-1/wt)}/(p*t)= 1.0416*10-5 Q/ t = W {1+ a (1/-1/wt)}/(p*t2)= -6.8942*10-5
36

Q/ Q = 1 K/ N=1/(Q*t)= 2.69*10-7 K/ K = 1 The result is tabulated as follows :

37

UNCERTAINTY BUDGET Probabil ity Distribu tion Estimat Limi Divis Type A es * Xi Unit ts or or B kg/m 4.74 Type B, 1.128 3 E-04 2.000 Normal kg/m 4.00 Type B, 3 8000 E-01 2.000 Normal kg/m oil Readability W(Mass) Readability time Readability 785.590 0.001
3

Source of uncertainty air wt

Standar d Partial uncertai derivat nty ive 2.37E-04 dQ/dair 2.00E-01 dQ/dwt 7.86E-02 dQ/doil 4.08E-04 dQ/doil 5.37E-03 dQ/dW 4.08E-04 dQ/dW 4.81E-04 dQ/dt 2.89E-06 dQ/dt

1.57 Type B, E-01 2.000 Normal Type B, 5.00 Rtglr E-04 1.225 7.60 Type B, E-03 1.414 Normal 5.00 Type B, E-04 1.225 Rtglr 9.61 Type B, E-04 2.000 Normal 5.00 1.732 Type B, E-06 Rtglr

kg/m
3

810.00 kg. 0.0010 kg 122.372 5 sec 1.00E- sec 05

Sensiti vity Coeffici ent 9.6721E06 1.4850E10 1.0755E05 1.0755E05 1.0416E05 1.0416E05 6.8942E05 6.8942E-

Uncerta nty contrib tion 2.2923 0 2.9701 1

-8.4487 0

-4.3898 0 5.5982 0 4.2514 0

-3.3127 0 -1.9901 1

05 Q Flowrate Pulse Readability K-Factor m^3 0.00844 /s 30374.3 lph 54 Puls 40015 es Puls 1.0000 e p/litr 38.756 e Type B, Normal 5.60 Type B, 2.000 E-01 Normal 5.00 Type B, 2.000 E-01 Rtglr Type A 1.00E+0 0 dK/dN dK/dN 2.69E-07 2.69E-07 1.00E+0 0

5.08E-02 2.80E-01 2.50E-01 4.49E-02

8.4739 0 5.0844 0 7.5358 0 6.7259 0 4.4864 0

Combined standard uncertainty Effective Degree of freedom (Calculated Value) From the student's distribution table, for confidence level of 95.45 % and above said effective degrees of freedom, the coverage factor, k Expanded uncertainty UE

0.067 = 8 p/litre = 31

= 2 0.13 = 56

p/litre

39

Table. 4 : Calculation of Flowrate using Best Fit Equation

Sl No.

F Hz

Qa (lph) y =92.922x+90.493

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

225.00 325.00 425.00 525.00 625.00 725.00 825.00 925.00 1025.00 1125.00 1225.00 1325.00 1425.00 1525.00 1625.00 1725.00 1825.00

20997.943 30290.143 39582.343 48874.543 58166.743 67458.943 76751.143 86043.343 95335.543 104627.743 113919.943 123212.143 132504.343 141796.543 151088.743 160380.943 169673.143

Sl No. 1 2 3 4 5 6 7 8 xm 379.5381 Sy 63.27244

F Hz. 408.069 405.115 635.209 635.651 326.993 324.253 150.776 150.238 ym 35357.932 Sxx 243157.7 = =

Qa lph 38066.651 37795.351 59091.550 59129.626 30374.354 30290.429 14081.433 14034.059 n 8 A 90.4929 39.754 100.180

erl(x) lph 22.66776 22.60964 39.70708 39.75396 23.36412 23.46801 36.90569 36.96067

B 92.9220 lph lph

Standard Uncertainty of best fit, Us Expanded Uncertainty UE

Table. 6 : Uncertainty
41

Budget in Flowrate UNCERTAINTY BUDGET Source of uncertai nty air wt oil Readabili ty W(Mass) Readabili ty time Readabili ty Probabili ty Standar Sensiti Distributi d Partial vity Divis on Type uncertai derivat Coeffici or A or B nty ive ent Type B, 9.6721E2.000 Normal 2.37E-04 dQ/dair 06 Type B, 1.4850E2.000 Normal 2.00E-01 dQ/dwt 10 Type B, 1.0755E2.000 Normal 7.86E-02 dQ/doil 05 Type B, 1.0755ERtglr 1.225 4.08E-04 dQ/doil 05 Type B, 1.0416E1.414 Normal 5.37E-03 dQ/dW 05 Type B, 1.0416E1.225 Rtglr 4.08E-04 dQ/dW 05 Type B, 6.8942E2.000 Normal 4.81E-04 dQ/dt 05 1.732 Type B, 2.89E-06 dQ/dt Rtglr 6.8942E42

Estimat es * Xi 1.128 8000 785.590 0.001

Unit kg/m
3

kg/m
3

Limits 4.74E04 4.00E01 1.57E01 5.00E04 7.60E03 5.00E04 9.61E04 5.00E06

Uncertai nty contribu tion 2.2923E09 2.9701E11 -8.4487E07 -4.3898E09 5.5982E08 4.2514E09 -3.3127E08 -1.9901E10

Degre of Freedo m

kg/m
3

kg/m
3

810.00 kg. 0.00 kg 122.3725 1 sec 1.00E-05 sec

05 m^3 8.44E-03 /s 30374.35 lph Flowrate 36 Combined standard uncertainty Q Effective Degree of freedom (Calculated Value) From the student's distribution table, for confidence level of 95.45 and above said effective degrees of freedom, the coverage factor, k Expanded uncertainty UE = 2.520 = 100.180 lph 3.98E+0 1 = 39.754 = 6 1.00E+0 0 lph 8.4739E07 3.9754E+ 01 6

Type A

43

4.7.1

Result for Test 2

Table 7: Datasheet for 2 Hoffer Flow Turbine Flow Meter (TFM) calibrated under atypical pressure conditions. Sl. No . 1 2 3 4 5 6 7 8 9 Pu p kP a 15 0 15 0 15 0 15 0 15 0 36 0 36 0 36 0 36 W1 kg 169.1 5 175.2 5 214.5 0 187.1 5 171.6 5 149.3 0 161.0 0 170.3 5 146.0 W2 kg 1046.9 5 832.85 1017.3 5 877.30 1034.0 0 929.15 986.60 1025.0 0 1032.5 t sec. 136.5722 72 107.3662 49 136.6558 32 81.11366 9 101.3135 37 131.9833 76 149.9440 71 96.76963 2 99.93087 T deg. C 29.5 5 29.3 5 29.6 2 29.3 4 29.4 0 29.9 2 30.0 7 30.2 3 30.3 Densi ty kg/m3 785.2 5 785.3 9 785.2 1 785.3 9 785.3 6 785.0 2 784.9 1 784.8 0 784.7 Qa m^3/ hr 29.50 7 28.11 3 26.97 3 39.05 4 39.07 1 27.13 4 25.28 8 40.56 9 40.75 Va litres 1119.404 838.447 1023.877 879.949 1099.548 994.786 1053.293 1090.507 1131.219 N pulse s 4498 9 3465 2 4189 6 3452 3 4311 0 4073 2 4328 0 4292 9 4443 K p/litre 40.19 01 41.32 88 40.91 90 39.23 30 39.20 70 40.94 55 41.09 02 39.36 61 39.28 F Hz. 329.42 322.75 306.58 425.61 425.51 308.61 288.64 443.62 444.69

2 Average KFactor Linearity(%) Repeatability (%)

33 ppl % %

40.17 = 37 2.640 = 7 1.090 = 9

4.4.4

Result for Test 3

Table 8: Datasheet for 2 Hoffer Flow Turbine Flow Meter (TFM) calibrated with a gasket protruding upstream of the meter. Sl. No . 1 2 Pu p W1 kP a kg 20 159.3 0 0 20 126.5 0 0 W2 kg 1062.8 0 1066.3 5 t sec. 69.92589 7 72.75303 4 T deg. C 29.8 3 29.8 4 Densi ty kg/m3 784.9 7 784.9 7
45

Qa m^3/ hr 59.33 9 59.32 8

Va litres 1152.589 1198.960

N pulse s 4441 5 4620 3

p/litre Hz. 38.53 50 635.17 38.53 59 635.07

3 4 5 6 7 8 9 10

20 0 20 0 20 0 20 0 20 0 20 0 20 0 20 0

101.8 5 137.8 5 152.4 5 135.8 0 140.5 0 81.70 147.3 0 194.9 5

1047.0 0 1057.1 5 1077.4 0 1141.5 5 1057.9 5 961.45 985.75 793.60

140.3084 21 135.3089 21 106.2689 50 114.3945 90 105.3912 63 195.1172 45 183.3352 18 161.9595 68

29.8 6 29.8 9 29.9 3 29.9 2 29.9 3 29.9 8 29.9 9 30.1 5

784.9 6 784.9 3 784.9 1 784.9 1 784.9 0 784.8 7 784.8 7 784.7 5

30.93 7 31.20 3 39.97 5 40.38 0 39.98 2 20.70 9 21.00 6 16.98 0

1205.737 1172.805 1180.043 1283.127 1170.489 1122.434 1069.741 763.908 Average KFactor Linearity(%) Repeatability (%)

4639 3 4515 5 4547 8 4889 7 4490 6 4317 3 4110 5 2936 9

38.47 69 38.50 17 38.53 93 38.10 77 38.36 52 38.46 37 38.42 52 38.44 57 38.43 = 96 0.561 = 4 0.150 = 2

330.65 333.72 427.95 427.44 426.09 221.27 224.21 181.34 ppl % %

Table. 9 : Calculation of Uncertainty for

46

Test 3 Sl No. 1 2 3 4 5 6 7 8 9 10 xm 35.9839 Sy 0.13570 Qa m^3/h r 59.339 59.328 30.937 31.203 39.975 40.380 39.982 20.709 21.006 16.980 ym 38.440 Sxx 2008.8 = = K p/litre 38.535 38.536 38.477 38.502 38.539 38.108 38.365 38.464 38.425 38.446 n 10 A 3.840E+01 0.0827 0.2152 B 1.176E-03 p/litre p/litre erl(x) p/litre 0.08271 0.08268 0.04555 0.04529 0.04458 0.04493 0.04459 0.06309 0.06243 0.07178

Standard Uncertainty of best fit, Us Expanded Uncertainty UE

47

70.000 65.000
4.4.5 Result for Test 4
48

Table 10: Datasheet for 2 Hoffer Flow Turbine Flow Meter (TFM) calibrated with upstream straight length pipe reduced to 10D. Sl. No . 1 2 3 4 5 6 7 8 9 10 13 Pu p kP a 20 0 20 0 20 0 20 0 20 0 20 0 20 0 20 0 20 0 20 0 20 0 W1 kg 120.9 5 128.9 0 138.8 0 119.7 0 70.20 160.4 0 128.9 5 140.1 0 158.3 0 365.8 5 156.6 0 W2 kg 1078.5 0 1023.2 0 1069.9 0 1036.6 0 932.80 1026.7 0 1026.5 5 1020.9 5 624.20 813.60 505.35 t sec. 73.50145 9 68.78930 3 84.79217 7 83.51434 6 102.8457 45 103.3226 92 135.5605 66 133.5461 94 99.97732 2 95.36025 8 107.4358 06 T deg. C 28.6 1 28.6 3 28.7 3 28.7 8 28.9 4 28.9 9 29.0 7 29.1 3 29.3 3 29.3 9 29.5 8 Densi ty kg/m3 785.9 4 785.9 1 785.8 4 785.8 0 785.6 8 785.6 5 785.5 9 785.5 5 785.4 0 785.3 7 785.2 3
49

Qa m^3/ hr 59.75 6 59.63 4 50.37 4 50.36 8 38.48 4 38.47 2 30.38 5 30.26 9 21.39 0 21.55 2 14.90 3

Va litres 1220.033 1139.488 1186.483 1168.448 1099.419 1104.177 1144.159 1122.865 594.020 570.901 444.751

N pulse s 4717 8 4410 1 4582 4 4511 5 4249 7 4257 7 4404 8 4321 3 2290 3 2199 7 1714 2

K p/litre 38.66 95 38.70 25 38.62 17 38.61 11 38.65 41 38.56 00 38.49 82 38.48 46 38.55 59 38.53 03 38.54 29

F Hz. 641.86 641.10 540.43 540.21 413.21 412.08 324.93 323.58 229.08 230.67 159.56

14

20 208.5 0 0

513.45

99.32243 9

29.6 2

785.2 0

14.09 6

388.909 Average KFactor Linearity(%) Repeatability (%)

1498 5

38.53 09 38.58 = 01 = 0.282 0.032 = 9

150.87 ppl % %

50

Table. 11 : Calculation of Uncertainty in k Sl No. 1 2 3 4 5 6 7 8 9 10 13 14 xm 35.8068 Sy 0.04419 Qa m^3/h r 59.756 59.634 50.374 50.368 38.484 38.472 30.385 30.269 21.390 21.552 14.903 14.096 ym 38.580 Sxx 2959.2 = K p/litre 38.669 38.702 38.622 38.611 38.654 38.560 38.498 38.485 38.556 38.530 38.543 38.531 n 12 A 3.846E+01 0.0233 erl(x) p/litre 0.02326 0.02318 0.01740 0.01740 0.01294 0.01294 0.01349 0.01353 0.01732 0.01723 0.02124 0.02177

B 3.424E-03 p/litre

Standard Uncertainty of best fit, Us

Expanded Uncertainty UE

0.1370

p/litre

52

70.000 65.000
Table 12: TEST 4 EXTENDED TO NON-LINEAR RANGE
53

Sl. No . 1 2 3 4 5 6 7 8 9 10 11 12 13

Pup kPa 200 200 200 200

W1 kg 120.9 5 128.9 0 138.8 0 119.7 0

W2 kg 1078.5 0 1023.2 0 1069.9 0 1036.6 0 932.80 1026.7 0 1026.5 5 1020.9 5 624.20 813.60 400.40 502.50 505.35

t sec. 73.50145 9 68.78930 3 84.79217 7 83.51434 6 102.8457 45 103.3226 92 135.5605 66 133.5461 94 99.97732 2 95.36025 8 226.1572 74 186.5694 62 107.4358

T deg. C 28.6 1 28.6 3 28.7 3 28.7 8 28.9 4 28.9 9 29.0 7 29.1 3 29.3 3 29.3 9 29.4 0 29.4 7 29.5

Dens ity kg/m3

Qa m^3/ hr

Va litres 1220.0 33 1139.4 88 1186.4 83 1168.4 48 1099.4 19 1104.1 77 1144.1 59 1122.8 65 594.02 0 570.90 1 357.84 6 285.82 3 444.75

N pulse s 4717 8 4410 1 4582 4 4511 5 4249 7 4257 7 4404 8 4321 3 2290 3 2199 7 1484 9 1259 0 1714

K p/litre 38.66 95 38.70 25 38.62 17 38.61 11 38.65 41 38.56 00 38.49 82 38.48 46 38.55 59 38.53 03 41.49 56 44.04 83 38.54

F Hz. 641.86 641.10 540.43 540.21 413.21 412.08 324.93 323.58 229.08 230.67 65.66 67.48 159.56

200 70.20 160.4 200 0 128.9 200 5 140.1 200 0 158.3 200 0 365.8 200 5 119.7 200 5 278.3 200 5 200 156.6

785.9 59.75 4 6 785.9 59.63 1 4 785.8 50.37 4 4 785.8 50.36 0 8 785.6 38.48 8 4 785.6 38.47 5 2 785.5 30.38 9 5 785.5 30.26 5 9 785.4 21.39 0 0 785.3 21.55 7 2 785.3 6 5.696 785.3 1 5.515 785.2 14.90
54

14

0 208.5 200 0 513.45

06 99.32243 9

8 29.6 2

3 785.2 0

3 14.09 6

1 2 388.90 1498 9 5

29 38.53 09 150.87

55

CALIBRAT 50.000

Table. 13: TEST 3 :EXTENDED TO NON-LINEAR RANGE

56

Sl. No . 1 2 3 4 5 6 8 9 10 11 12 13 14

Pup kPa 200 200 200 200 200 200 200

W1 kg 159.3 0 126.5 0 101.8 5 137.8 5 152.4 5 135.8 0 140.5 0

W2 kg 1062.8 0 1066.3 5 1047.0 0 1057.1 5 1077.4 0 1141.5 5 1057.9 5

t sec. 69.92589 7 72.75303 4 140.3084 21 135.3089 21 106.2689 50 114.3945 90 105.3912 63 195.1172 45 183.3352 18 416.3198 11 161.9595 68 199.1799 76 215.3579

T deg. C 29.8 3 29.8 4 29.8 6 29.8 9 29.9 3 29.9 2 29.9 3 29.9 8 29.9 9 29.9 1 30.1 5 29.9 6 29.9

Dens ity kg/m3

Qa m^3/ hr

Va litres 1152.5 89 1198.9 60 1205.7 37 1172.8 05 1180.0 43 1283.1 27 1170.4 89 1122.4 34 1069.7 41 564.40 3 763.90 8 379.27 7 530.36

N pulse s 4441 5 4620 3 4639 3 4515 5 4547 8 4889 7 4490 6 4317 3 4110 5 7715 8 2936 9 2341 1 2433

K p/litre 38.53 50 38.53 59 38.47 69 38.50 17 38.53 93 38.10 77 38.36 52 38.46 37 38.42 52 136.7 07 38.44 57 61.72 53 45.87

F Hz. 635.17 635.07 330.65 333.72 427.95 427.44 426.09 221.27 224.21 185.33 181.34 117.54 112.99

200 81.70 961.45 147.3 200 0 985.75 200 0.00 194.9 200 5 107.9 200 5 200 251.0 442.40 793.60 405.20 666.70

784.9 59.33 7 9 784.9 59.32 7 8 784.9 30.93 6 7 784.9 31.20 3 3 784.9 39.97 1 5 784.9 40.38 1 0 784.9 39.98 0 2 784.8 20.70 7 9 784.8 21.00 7 6 784.9 2 4.881 784.7 16.98 5 0 784.8 1 6.855 784.8 8.866
57

40

96

58

CALIBRAT

60.000
59

4.5

Summary of Result

Table 14: Comparison of Tests Results

TEST

KAvera ge (ppl)

K @ Linear Qnom ity (ppl) %

Repeatabi Standard Expanded lity Uncertainty Uncertaint of best fit, y UE % Us

1 2.i [@ 1.5 bar] 2.i [@ 3.6 bar] 3 4

38.619 38.62 5 40.173 40.17 7 37 40.175 40.17 6 56 38.439 38.39 6 81 38.580 38.54 1 92

0.282 7 2.640 6 2.249 0 0.561 4 0.282

0.0060 1.0719

0.0449 N/A

0.1356 N/A

1.0909

N/A

N/A

0.1502 0.0329

0.0827 0.0233

0.2152 0.1370

60

61

4.6

Discussion of Results and Recommendations

Careful analysis of the tests results reveals the following Within the linear range under different operating conditions, the meter factor curves were fairly linear and repeatable, especially for Test 1 which was conducted according to standard/manufacturers recommendations At the low end of the flow rate range the meter factor curves become less linear and less repeatable than they are at the medium and higher rates If a plot of meter factor versus flow rate has been developed for a particular liquid, and other variables are constant, a meter factor may be selected from the plot for flow rates within the meters operating/linear range However, for greatest accuracy, the meter should be reproved at the new operating flow rate The average k-factor, and also the k-factor at the nominal flow rate, for each of the tests (test 2, test 3 and test 4) differs from that of test 1 The deviation in k-factor values, and repeatability, (from that of test 1) was more pronounced in test 2 (conducted under atypical pressure conditions)
The average meter factor does not change significantly for a slightly skewed velocity profile

but increases drastically for a highly skewed velocity profile

The relative volume of the liquid will change as a result of its compressibility. (The physical dimensions of the meter will also change as a result of the expansion or contraction of its housing under pressure.)
62

 Volumetric corrections for the pressure effects on liquids referenced to appropriate pressure is therefore necessary The deviation in k-factor due to the protruding gasket upstream of the TFM, and that caused by reduction of upstream straight length pipe, were not as pronounced as would have been expected owing to Hoffer Flow TFM which has an in-built flow conditioner The Hoffer Flow meter with in-built flow

conditioners

4.7

Suggestion for further studies

63

Extend test to larger and smaller size Turbine Flow Meters: e.g for 1 NB and 4 NB

 Repeat test at least five times for each flow rate for better assessment of repeatability Extend all tests beyond the linear range for better understanding of TFM behaviour at the non-linear range Conduct with a TFM make without in-built Flow Straighteners

Extend test to study the response of TFM under conditions of varying temperature, viscosity, density, and changes in piping, valves, or valve positions that affect fluid profile or swirl.

CHAPTER FIVE Conclusion

In view of the foregoing observation:
The overall performance of measurement by a meter depends on the condition of the meter

and its accessories, the operating pressure, the calibration system and the variations between operating and calibration The potential for error increases in proportion to the difference between the calibration and operating conditions

Calibration and usage of Turbine Flow Meter should be according to standards and manufacturers specifications

A meter factor obtained for one set of conditions will not necessarily apply to a changed set of conditions
64

 For greatest accuracy, the meter should be proved at the operating conditions

Meter factors shall be determined by proving the meter under conditions of rate, viscosity, temperature, density, and pressure similar to those that exist during intended operation. References API MPMS 5.3 (2005): Measurement of liquid hydrocarbons by turbine meters. API MPMS 12.2 (2005): Calculation of liquid petroleum quantities measured by TFM or Displacement meters. ISO 4185-1980: Measurement of liquid flow in closed conduit-weighing methods. ISO 5168-2005: Measurement of fluid flow-procedures for evaluation of uncertainties.

65