Documente Academic
Documente Profesional
Documente Cultură
Central Railway
February 2006
Rambll Denmark A/S Bredevej 2 DK-2830 Virum Denmark Phone +45 4598 6000 www.ramboll.com
Liability
Ramboll Denmark and its sub consultants assume no warranty with regard to accuracy, completeness, or usefulness of the information contained in this publication, and specifically no liability with regard to the product described in the publication, or to the use, or usefulness of the product for specific purposes.
Table of contents
1. 1.1 1.2 1.3 1.4 1.5 2. 2.1 2.1.1 2.1.2 2.1.3 2.1.4 2.1.5 2.1.6 2.1.7 2.2 2.3 2.4 2.5 2.6 2.7 3. 3.1 3.2 3.3 3.4 3.5 3.6 3.7 4. 4.1 4.2 4.3 4.4 5. 5.1 5.2 5.3 5.4 5.5 5.6 5.7
Introduction Non Destructive Testing Non Destructive Evaluation Scope of the present manual Purpose Applications NDT-methods Extended principal inspection overview Primary Planning Requisition List of bridges Planning travel route Information retrieval Check of equipment Making appointments Safety considerations Detailed planning of tests Execution of visual inspection Execution of tests Assessment of damage cause and extent General considerations regarding future maintenance activities Reporting Special inspection - overview Primary Planning Detailed planning of tests Execution of tests Assessment of damage cause and extent Setting up of relevant repair strategies Economic analysis of the strategies Reporting Planning of inspections using NDT-measurements Visual inspection Areas requiring investigation Homogeneous areas Evaluation of test results Types of damage General structural damage Damage due to water Damage on surface treatment systems Damage on concrete structures Damage on steel structures Damage at masonry structures Damage at wooden structures
1-1 1-1 1-1 1-2 1-3 1-4 2-1 2-1 2-1 2-1 2-2 2-2 2-2 2-2 2-3 2-3 2-3 2-4 2-4 2-4 2-4 3-1 3-1 3-1 3-1 3-1 3-2 3-2 3-2 4-1 4-2 4-2 4-2 4-3 5-1 5-1 5-1 5-1 5-1 5-2 5-2 5-2
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
5.8 5.8.1 5.8.2 6. 6.1 6.2 6.2.1 6.2.2 6.2.3 6.2.4 6.2.5 6.2.6 6.2.7 6.2.8 6.2.9 6.2.10 6.2.11 6.2.12 6.2.13 6.2.14 6.3 6.3.1 6.3.2 6.3.3 6.3.4 6.3.5 6.3.6 6.3.7 6.3.8 6.3.9 6.3.10 6.3.11 6.3.12 6.3.13 6.3.14 6.4 6.4.1 6.5 6.5.1 6.5.2 6.5.3 6.6 6.6.1 6.6.2 6.6.3 6.7 6.7.1 6.7.2 6.7.3
Registration of damage Concrete structures Steel structures Damage mechanisms General Structural Deficiencies I. Structural cracks (load induced cracks) in concrete members II. Excessive/unintended deflections and movements III. Fracture/crushing IV. Structural Problems, Steel Components Structural Cracks in Concrete Members Flexural Cracks Longitudinal cracks Shear Cracks Bending and Shear Torsion Bearing Cracks Splitting Cracks Structural Cracks, Examples Structural Problems, Steel Components Non-Structural Cracks in Concrete Shrinkage cracks (due to drying) Thermal Cracks (due to hydration) Cracks due to plastic shrinkage Cracks due to plastic settlement Initiation of Corrosion Carbonation Chlorides Carbonation and Chlorides Propagation of Corrosion Corrosion products and corrosion rate Local/general corrosion Bridge Deck Pier column Wing walls / retaining walls Alkali-aggregate Reactions Crack Pattern Chemical Attack of Concrete and Masonry Acid Attack Sulphate Attack Seawater Attack Erosion / Scour Aggradation / degradation General scour Local scour Corrosion of steel structures Electrochemical corrosion Chink Corrosion Galvanic Corrosion
5-2 5-2 5-4 6-1 6-1 6-1 6-1 6-1 6-2 6-2 6-2 6-4 6-5 6-5 6-7 6-8 6-8 6-10 6-12 6-23 6-26 6-26 6-28 6-28 6-29 6-31 6-32 6-34 6-38 6-39 6-40 6-42 6-43 6-44 6-45 6-46 6-47 6-49 6-50 6-51 6-53 6-54 6-54 6-55 6-55 6-56 6-56 6-60 6-60
II
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
6.7.4 6.7.5 6.7.6 6.8 6.9 7. 7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.8 7.9 7.10 7.11 7.12 7.13 7.14 7.15 7.16 7.17 7.18 7.18.1 7.18.2 7.18.3 7.18.4 7.18.5 7.18.6 7.19 7.20 7.21 7.22 7.22.1 7.22.2 7.22.3 7.22.4 7.22.5 7.22.6 7.22.7 7.22.8 7.22.9 7.22.10 7.23 7.23.1 7.23.2 7.23.3 7.23.4
Stress Corrosion Corrosion and Fatigue Atmospheric Corrosion Ageing of Steel Erosion of masonry structures NDT-methods General Visual inspection Crack measuring gauge Crack detection microscope Boroscope Half-cell potential Corrosion rate meter Cover meter Measurements Spraying indicators (pH) Impact-Echo equipment Impulse Response equipment Capo-Test (concrete strength) Pull-off/Bond-Test Schmidt hammer Ground penetration radar Chloride content Coring equipment Evaluation of concrete cores Macro analysis on cores and plane sections Crack detection on impregnated plane sections Micro analysis on thin sections Air void analyse on plane section Moisture analysis Residual Reactivity Test Acoustic emission monitoring Structural testing system Structural scan equipment Ultrasonic testing Definition of ultrasound Through transmission technique The pulse echo technique Definitions and general terms Refraction and reflection of ultrasonic waves Probes Examination of rolled, cast and forged objects Examination of welds Determination of defect size References Ultrasonic thickness gauge Introduction Thickness measurements of steel plates Special equipment Thickness measurements of hot steel plates
6-60 6-60 6-61 6-62 6-63 7-1 7-1 7-2 7-3 7-3 7-4 7-5 7-8 7-9 7-10 7-10 7-12 7-14 7-16 7-17 7-17 7-19 7-20 7-22 7-23 7-24 7-25 7-28 7-28 7-29 7-31 7-33 7-34 7-36 7-36 7-36 7-37 7-40 7-42 7-44 7-50 7-58 7-65 7-71 7-74 7-74 7-75 7-77 7-78
III
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.23.5 7.24 7.24.1 7.24.2 7.24.3 7.24.4 7.24.5 7.24.6 7.24.7 7.24.8 7.25 7.25.1 7.25.2 7.25.3 7.25.4 7.25.5 7.25.6 7.25.7 7.25.8 7.25.9 7.25.10 7.25.11 7.25.12 7.25.13 7.25.14 7.25.15 7.25.16 7.25.17 7.25.18 7.25.19 7.26 7.26.1 7.26.2 7.26.3 7.26.4 7.26.5 7.26.6 7.26.7 7.26.8 7.26.9 7.26.10 7.26.11 7.26.12 7.26.13 7.26.14 7.26.15 7.26.16 7.26.17 7.26.18
Thickness measurements of other materials than steel 7-79 Coating Thickness Measurement 7-81 Magnetic Film Thickness Gages 7-81 Eddy Current 7-83 Ultrasonic 7-84 Micrometer 7-84 Destructive Tests 7-85 Gravimetric 7-85 Thickness Measurements in Practice 7-86 Thickness Standards 7-87 Dye penetrant 7-88 Introduction and History of Penetrant Testing 7-88 Improving Detection 7-90 Basic Processing of a Dye Penetrant Testing 7-91 Common Uses of Dye Penetrant Inspection 7-93 Advantages and Disadvantages of Dye Penetrant Testing 7-94 Dye Penetrant Testing Materials 7-96 Penetrants 7-98 Emulsifiers 7-105 Developers 7-106 Preparation of Part 7-109 Selection of a Penetrant Technique 7-110 Penetrant Application and Dwell Time 7-113 Penetrant Removal Process 7-115 Use and Selection of a Developer 7-119 Quality Control 7-123 System Performance Check 7-129 Nature of the Defect 7-130 Health & Safety Precautions in Dye Penetrant Inspection 7-131 References and Resources 7-133 Magnetic Particle Flow Test 7-135 Introduction to Magnetic Particle Inspection (MPI) 7-135 Basic Principles 7-136 History of Magnetic Particle Inspection 7-137 Magnetism 7-138 Magnetic Materials 7-139 Magnetic Domains 7-140 Magnetic Field Characteristics 7-141 Electromagnetic Fields 7-143 Magnetic Field Produced by a Coil 7-144 Quantifying Magnetic Properties (Magnetic Field Strength, Flux Density, Total Flux and Magnetization) 7-145 The Hysteresis Loop and Magnetic Properties 7-146 Permeability 7-148 Magnetic Field Orientation and Flaw Detectability 7-149 Magnetization of Ferromagnetic Materials 7-151 Magnetizing Current 7-153 Longitudinal Magnetic Fields, Distribution and Intensity 7-155 Circular Magnetic Fields, Distribution and Intensity 7-157 Demagnetization 7-161
IV
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.26.19 7.26.20 7.26.21 7.26.22 7.26.23 7.26.24 7.26.25 7.26.26 7.26.27 7.26.28 7.26.29 7.26.30 7.26.31 7.26.32 7.26.33 7.26.34 7.26.35 7.27 7.27.1 7.27.2 7.27.3 7.27.4 7.27.5 7.27.6 7.27.7 7.28 7.29 7.30 7.31 8. 8.1 8.2 8.2.1 8.2.2 8.2.3 8.2.4 8.2.5 8.2.6 8.2.7 8.2.8 9. 9.1 9.2 9.2.1 9.2.2 9.2.3 9.2.4 9.2.5
Measuring Magnetic Fields Portable Magnetizing Equipment Stationary Magnetizing Equipment Multidirectional Magnetizing Equipment Lights Magnetic Field Indicators Magnetic Particles Suspension Liquids Testing Practices Inspection using Magnetic Rubber Continuous and Residual Magnetization Techniques Field Direction and Intensity Particle Concentration and Condition Lighting Eye Consideration Examples of Visible Dry Indications Examples of Fluorescent Wet Indications Strain gauging Measurement Principle Gauge Construction Applications Structural design Fitness for purpose Testing and documentation Mechanical Strain Gauge Electromagnetic Testing (ET) or Eddy Current Testing Radiography (RT) Sonic Methods Accelerometers Economic analysis General Present Value Method Repair Strategies Service Life Time Frame Time of Repair Residual Value Discount Rate Sensitivity Analysis Optimum solution special inspection Reporting of Extended Principal Inspection General Text Section Cover Page Front Page Summary Motivation of the extended principal inspection Background documents
7-162 7-164 7-168 7-169 7-170 7-173 7-176 7-178 7-179 7-181 7-181 7-183 7-186 7-187 7-189 7-189 7-192 7-195 7-195 7-195 7-196 7-199 7-200 7-201 7-203 7-204 7-205 7-206 7-207 8-1 8-1 8-2 8-3 8-4 8-5 8-5 8-6 8-6 8-6 8-7 9-1 9-1 9-1 9-1 9-1 9-2 9-2 9-2
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
9.2.6 9.2.7 9.2.8 9.3 9.3.1 9.3.2 9.3.3 9.3.4 9.3.5 9.3.6 10. 10.1 10.2 10.2.1 10.2.2 10.2.3 10.2.4 10.2.5 10.2.6 10.2.7 10.2.8 10.2.9 10.3 10.3.1 10.3.2 10.3.3 10.3.4 10.3.5 10.3.6 10.3.7 11.
Registrations Evaluation of registrations General considerations regarding future maintenance activities Appendices A: Background Material B: Selected Drawings C: Visual Inspection D: NDT-method no. 1 E - ?: NDT-method no. 2 - ? F.. Other Reporting of Special Inspection General Text Section Cover Page Front Page Summary Motivation of the special inspection Background documents Registrations Evaluation of registrations Repair strategies Recommendation of follow-up activities Appendices A: Background Material B: Selected Drawings C: Visual Inspection D: NDT-method No. 1 E - ?: NDT-method No. 2 - ? F: Economic analysis G.. Other References
9-2 9-2 9-3 9-3 9-3 9-4 9-4 9-4 9-5 9-5 10-1 10-1 10-1 10-1 10-1 10-2 10-2 10-2 10-2 10-2 10-3 10-4 10-4 10-4 10-4 10-4 10-5 10-5 10-5 10-5 11-1
VI
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Appendices A: Handout of slides from classroom training in NDT-methods A1: Introduction to the Classroom Training in NDT- and UWI A2: General Introduction to Deterioration Mechanism A3: General Introduction to Systematic Operation and Maintenance A4: Special Inspection A5: Structural Assessment - Case A6: Crack Measuring Gauge and Crack Detection Microscope A7: Boroscope A8: Half Cell Potential Measurements A9: Corrosion Rate Meter A10: Covermeter A11: Spraying Indicators A12: Impact-Echo A13: Impulse Response (sMASH) A14: CAPO-test A15: Pull off / Bond test A16: Schmidt Hammer A17: Ground Penetration Radar A18: Chloride Content A19: Coring Equipment A20: Evaluation of Concrete Cores A21: Acoustic Emission Monitoring A22: Structural Testing System A23: Structural Scan Equipment A24: Introduction to Non Destructive Testing of Steel Structures A25: Ultrasonic Testing A26: Ultrasonic Thickness Gauge A27: Magnetic Thickness Gauge A28: Dye Penetrant Inspection System A29: Magnetic Particle Testing A30: Strain Gauging A31: Introduction to Rehabilitation of Concrete, Steel and Masonry Bridges A32: Introduction to Laboratory Tests of Steel B: Template for Extended Principal Inspection Report C: Template for Special Inspection Report
VII
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
1.
Introduction
The present manual is addressed to Central Railway and covers the subjects of Non Destructive Testing (NDT) and inspections including NDT. The manual has been prepared as part of a pilot project in the area of Non Destructive Testing of railway bridges on Central Railway. The manual describes the basic issues regarding planning, execution and reporting of bridge inspections including NDT-investigations. Selected types of damage and damage mechanisms are also described in this manual as an extensive knowledge of the possible damage mechanisms and signs of damage are very important for selecting the right NDT-method to apply in each individual case. Selected NDT-methods to be used on concrete, steel and masonry bridges are described in the manual. The methods described in the present manual are primarily the NDT-methods included in the pilot project program.
1.1
1.2
1-1 1-1/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
1.3
The above mentioned evaluations are determined in both the extended principal inspection as well as in the special inspection. When performing a special inspection the expected development in time of damage is determined in addition to the type, extent and cause. The information from the NDT-investigations provides the basis for decisions concerning the selection of the optimum repair strategy. Extended principal inspections and special inspections include field measurements and field testing. For this purpose, special equipment and tools such as NDTmethods are required. The report from the inspections is an important part of the necessary background material for the planning and specification of repair and rehabilitation works. Performing inspections using NDT-equipment involves making evaluations and taking decisions on-site, based on knowledge and experience. As a consequence, it is not possible to prescribe a full step-by-step set of instructions in performing an extended principal inspection or a special inspection and in using
1-2 1-2/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
NDT-methods. The descriptions in this manual must only be regarded as guidelines, hints and examples.
1.4
Purpose
The purpose of this manual is to give guidelines for carrying out inspections using NDT-methods and to describe some of the most common damage mechanisms. For each of the NDT-methods included in the pilot project descriptions are given with regards to the following subjects: Theory Technical Method Description Applications and Limitations Test Planning and Execution of Field Tests Interpretation and Reporting of Results
As appendices to this manual the handouts of the presentations from the classroom training are enclosed. All the above mentioned subjects are described in the handouts. Chapters on reporting of extended principal inspections and special inspections are also included as well as a chapter of economic analysis using the present value method.
1-3 1-3/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
1.5
Applications NDT-methods
The manual covers descriptions of the NDT-methods mentioned in Table 1-1, which are the methods included in this pilot project. The table also includes methods and procedures used for calibration of various NDT measurements and includes methods for analysing material samples. In addition to the test methods in Table 1-1, few additional test methods are described in section 7. NDT-method Crack measuring gauge Crack detection microscope Boroscope Half cell potential measurements Corrosion rate meter Cover meter Spraying indicators (pH) Impact-Echo equipment Impulse response equipment CAPO test Pull-off/Bond test Schmidt Hammer Ground Penetration Radar Chloride content Coring equipment Evaluation of concrete cores Acoustic emission monitoring Structural testing system Structural scan equipment Ultrasonic Thickness gauge Ultrasonic testing Magnetic thickness gauge Dye penetrant Magnetic particle testing Strain gauging Used for structures made of: Concrete, steel and masonry Concrete, steel and masonry Concrete, steel and masonry Concrete Concrete Concrete Concrete Concrete Concrete Concrete Concrete Concrete and masonry Concrete and masonry Concrete Concrete and masonry Concrete Steel Concrete, steel and masonry Concrete, steel and masonry Steel Steel Steel Steel Steel Steel
Table 1-1: Test methods included in this pilot project. 1-4 1-4/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
2.
2.1
Primary Planning
2.1.1 Requisition In the requisition of the extended principal inspection some information is required which the inspector needs in order to make a proper planning, i.e. the proper selection of the required NDT-methods. The owner of the bridge must state the reason for initiation of the extended principal inspection and which components to include in the NDT-testing. Future expected changes to the road and bridge usage, such as load assumptions or possible road widenings must be stated as well. 'As built drawings' and all relevant reports including the inventory report should be enclosed with the requisition. Before execution of the inspection, the owner of the bridge and the company carrying out the inspection should agree on a time schedule and a budget. 2.1.2 List of bridges The primary planning often covers a series of extended principal inspections involving the same NDT-inspections performed by the same inspection team. In general, the bridges with the most serious damage should be inspected first, but other factors may influence the ranking list, such as the traffic volume or a convenient travel route to remote bridges. A series of bridges of the same design and with the same type of damage should preferably be inspected together.
2-1 2-1/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
For railway bridges an important issue in planning the inspections is the need for shutting down the power lines and to gain access to working on or close to the rails. To minimize the traffic interference extended principal inspections might be carried out on several bridges on the same railway line at the same time. During the planning phase it is important to contact the local railway authority as soon as possible to schedule the work on or close to the rails. The need for traffic blocks, traffic regulations and safety arrangement for railway traffic must be clarified in cooperation with the local railway authority. Typically the work has to be carried out in intervals with the lowest traffic intensity if possible. 2.1.3 Planning travel route
2.1.4 Information retrieval The major part of the primary planning consists of retrieving information on the bridge(s) in question. This information includes: All relevant previous reports concerning the bridge(s) (including inventory). "As built" drawings. Any previous calculations done on the bridge e.g. structural capacity assessments, damage evaluations and so forth. Information about power cables or other utilities buried or located within the bridge that might be damaged during the NDT-inspection (particularly when breaking up roadway surface) if no precautions are taken. Information on major events during maintenance. Check of equipment
2.1.5
What kinds of tests must be done on each bridge? Which equipment in the inspection vehicle must be used? Check that the equipment (and necessary consumables) is present and is functioning. Which equipment (e.g. traffic signs, ladders, scaffolding) and personnel should be provided by the local Technical Unit?
2.1.6
Making appointments
With the Bridge Engineer (concerning closure of railway line, traffic signs, inspection, equipment, possible reestablishment after break-ups etc.). With other authorities e.g. the road owner if the bridge is crossing a road. About accommodations for inspection personnel. 2-2 2-2/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
2.1.7 Safety considerations During the planning it is important to consider the safety factors. The inspector must always contact the bridge owner and other relevant authorities to gain knowledge of the safety rules submitted by the authorities. Some authorities may for instance require all inspectors to attend a specific safety course. When working on or nearby the railway line some standard precautions have to be taken as for instance: Contact the railway authority to get information regarding demands for minimum distances to the railway line of scaffolding, material, etc. Make sure that the railway authority knows in what exact position of the railway line you are working on. Make sure that the power cables are shut off. Do not use any metal equipment close to the power lines. Make sure that there is always someone watching for trains to warn you if a train should pass.
2.2
2.3
2-3 2-3/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
2.4
Execution of tests
The tests are carried out according to the test plan and following the 'directions for use' of the test equipment. See also chapter 7 for more detailed description of the NDT-methods included in this pilot project. It is important to follow the test plan and not draw premature conclusions from the first test results obtained.
2.5
2.6
2.7
Reporting
In order to facilitate comparison of extended principal inspection reports, and in order not to forget important aspects of the inspection, the reporting is made using a fixed table of contents: Summary. Motivation of (reason for) the extended principal inspection. Background documents (list of the background material that has been available for the inspection). Registrations (extent and location of tests, and a summary of the results). Evaluation of registrations (condition rating of the bridge components, what do the NDT-registrations indicate regarding cause and extent of damage). General considerations regarding future maintenance activities.
The inspection report will include those relevant of the following appendices: Background materials (inventory and principal inspection printouts with possible revisions). Selected drawings (relevant extracts from 'as built' drawings). Visual inspection. 2-4 2-4/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The content of each section in the extended principal inspection report is described in more details in chapter 9.
2-5 2-5/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
3.
3.1
Primary Planning
The primary planning of the special inspection is similar to the one described for the extended principal inspection see section 2.1.
3.2
3.3
Execution of tests
The detailed planning of tests included in the special inspection is similar to the one described for the extended principal inspection see section 2.4.
3.4
3-1 3-1/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
3.5
A repair strategy does not only include a choice of repair method, it also describes the optimum year of repair, the traffic situation before, during and after the repair, the estimated repair cost and possible maintenance costs before and after the repair.
3.6
3.7
Reporting
In order to facilitate comparison of special inspection reports, and in order not to forget important aspects of the inspection, the reporting is made using a fixed table of contents: Summary Motivation of (reason for) the special inspection Background documents (list of the background material that has been available for the inspection) Registrations (extent and location of tests, and a summary of the results) Evaluation of registrations (what do the registrations indicate regarding cause and extent of damage, including the risk of further deterioration)
3-2 3-2/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Relevant repair strategies with cost estimates (client's estimate) and economic analyses Recommendation of follow-up activities (such as: further investigations, monitoring, repair works, doing nothing).
The inspection report will include those relevant of the following appendices: Background materials (inventory and principal inspection printouts with possible revisions). Selected drawings (relevant extracts from 'as built' drawings). Visual inspection. NDT-method no 1. NDT-method no 2. etc. Economic analysis.
The content of each section in the special inspection report is described in more detail in chapter 10.
3-3 3-3/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
4.
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
4.1
Visual inspection
A detailed visual inspection is always carried out as the first activity in the field. The visual inspection must include the entire structure. On paper sketches the cracks (length, width, direction) are marked as well as areas with spalling, rust stains, disintegration and other relevant observations. The result of the visual inspection is part of the basis for determining the NDT-methods to be used, and the extent of the tests. If the visual inspection reveals significant damage to other components than those for which a special inspection was required, the inspector must consult the owner of the bridge in order to revise the requisition. If the inspection engineer is not the same person as the rehabilitation engineer, the test plan must be discussed with the rehabilitation engineer before the inspection is carried out.
4.2
As a rule-of-thumb: In determining the areas requiring investigation, just as much attention should be paid to the areas without visible signs of deterioration as to areas with visible signs (especially in case of corrosion). Evaluation of the condition in areas with no visible signs of deterioration is more difficult, but very important, when estimating the remaining service life of the bridge and the optimum repair strategy.
4.3
Homogeneous areas
On the basis of the visual inspection and prior knowledge the structure may be divided into homogeneous areas. A homogenous area is defined as an area where there the present level of deterioration and parameters affecting the deterioration of the structure exhibits only a random variation. Consider for example a bridge pier in saline water. The chloride surface concentration will be large in the tidal and splash zones. The chloride surface concentration will decrease with increasing distance from the mean water level. In this case it makes no sense to compare results from different piers if the tests are not performed at the same distance from the mean water level. To overcome this problem the piers may be divided into homogenous areas. Tests originating from the same homogenous
4-2 4-2/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
area may be treated as coming from the same population e.g. in conjunction with a statistical analysis. The division of the structure into homogenous areas also depends on the material parameters. If for example two different concrete compositions have been used for two piers in saline water results of chloride measurements from these two different structures cannot be treated as a whole even though the measurements have been performed at the same distance from the mean water level.
4.4
4-3 4-3/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
5.
Types of damage
This section includes a short description of the different types of damage observed for the different causes of damage and for the different types of materials. The bridge owner is advised to compile and update a database of the most typical damage observed at the principal inspections. The content of this database will depend on the type of bridges, the climate, etc.
5.1
5.2
5.3
5.4
5-1 5-1/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Weathering honeycombing
5.5
5.6
5.7
5.8
Registration of damage
When inspecting the inspectors should pay particular attention to the following components: 5.8.1 Concrete structures Reinforced concrete girders may present a crack pattern as shown in Figure 5-1. This is not necessarily dangerous however if the cracks are due to overloading then a structural assessment should be made. The crack widths can be measured using a crack measuring gauge see section 7.1. Reinforced cantilevered structures show a crack pattern, see Figure 5-2, slightly modified. However the same principles apply. The influence of cracks on the bearing capacity may be harmless at the time of inspection but some cracks may initiate corrosion that later may be critical. Fine cracks in reinforced structures may be harmless unless the structure is exposed
5-2 5-2/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
to very aggressive environment, e.g. positioned in the splash zone of saline water. Fine cracks in pre-stressed structures are more critical. Bearings often need a close inspection. The stresses at the bearings are high, therefore there is a danger of concrete crushing, in particular if the bearings are misplaced or badly designed. Common reinforced concrete structures will not fail without an early warning such as coarse cracks and visible deflections. Pre-stressed concrete structures are much more sensitive to damage and corrosion of cables or failure of anchorage may lead to sudden failures of a structure.
5-3 5-3/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
5.8.2 Steel structures Connections in steel structures are very likely to collect dust and water at corners or poorly drained surfaces. Therefore the risk of corrosion is high. The inspector should pay attention to such areas that it might be necessary to clean to see if there is corrosion. If a welding in high-stress areas of steel structures is not executed correctly, fatigue cracks may occur at the edge of the welding. A close inspection is necessary to find such cracks. If the inspector detects more cracks at the same type of weld particular attention should be paid to extend the random samples to cover larger samples of components with the same positions. Connections in steel structures exposed to repetitive loads may fail in fatigue without any other warning than very fine cracks. Therefore, potentially dangerous details of steel structures should be pinpointed in advance of the inspection in order to give these details a closer inspection.
Surface treatment: The quality of the surface treatment relies in general on the quality of manual application of the paint. Poor climate conditions during application and hardening and also the manual performance failures, e.g. an edge that is forgotten, a blocked spray nozzle, erroneous distance or angle for spraying, etc. Therefore the position and extent of damage will very often mirror the cause of the damage. Particular attention should be drawn to inspect areas where the access for application has been poor or where the number of construction damage due to temporary attachments have been higher than at normal sections.
5-4 5-4/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
6.
6.1
Damage mechanisms
General
This chapter contains a description of the most commonly occurring damage mechanisms encountered. The intention is that the description should be sufficient as a guideline to recognise the damage when it occurs on a bridge, and to evaluate how significant it is. However, the descriptions in this relatively short manual cannot be exhaustive, and it is essential that investigations are carried out by experienced engineers with a thorough knowledge of bridges and damage mechanisms, and a good amount of common sense.
6.2
Structural Deficiencies
Structural deficiencies may be a danger to the structural safety. Therefore, identifying such problems is very important. Structural deficiencies can be divided into the following four types, which can be distinguished by their appearance: 6.2.1 I. Structural cracks (load induced cracks) in concrete members Structural cracks can be recognised as cracks with well defined orientation and with specific crack patterns related to each type of internal forces (bending, shear). Structural cracks might be a sign of a structural deficiency. For reinforced concrete (RC) structures, cracks in most cases are not serious. RC is allowed to crack. The crack width and spacing will indicate whether there is something wrong or not, taking the specific type of load and type of reinforcement into consideration. Coarse cracks are an indication of over-load and/or under-design. For pre-stressed concrete, cracking is an indication of a potentially serious problem (overload, inadequate initial load bearing capacity, or plain design error). Normally no cracks should be visible.
6.2.2 II. Excessive/unintended deflections and movements Examples: Settling of foundation (possible causes: poor soil condition, scour). 6-1 6-1/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Deflection of girders (possible causes: low stiffness, creep, poor design or improper formwork).
Horizontal movements of retaining walls and wing walls (possible causes: low stiffness, creep, compaction of back fill, soil condition, under-design).
Bearings out of position (possible causes: wrong positioning, unforeseen movements, shrinkage, creep, deterioration and temperature).
6.2.3 III. Fracture/crushing Examples: Local crushing at supports/bearings (possible causes: honeycombs, wrong positioning of bearings and/or reinforcement, overload, inadequate initial load bearing capacity). Columns (possible causes: impact during flooding). Superstructure (possible causes: impact from vehicle (vertical clearance)). Local crushing at expansion joints (possible causes: inadequate joint system, wear, movements restrained).
6.2.4 IV. Structural Problems, Steel Components Examples: Fatigue cracks at welded connections. Brittleness due to ageing and cold brittleness. Buckling of compression members. Eccentricities in welded connections.
In the following sections, selected examples from the four groups of structural deficiencies are shown.
6.2.5 Structural Cracks in Concrete Members Types of Structural Cracks From a structural point of view, it is important to distinguish structural cracks from the non-structural cracks caused by concrete shrinkage etc. 6-2 6-2/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Pure Tension All cracking in concrete members is caused by tensile stresses (concrete has a low tensile strength, but high compressive strength). Therefore it is obvious to consider pure tension as the basic case.
In a reinforced prismatic concrete beam subjected to pure tension, cracks formed will cross the whole cross section. An increase/jump in the steel stress will suddenly arise at the crack, when the cracks are formed. This affects the bond between the concrete and the bar in a certain zone (slip distance l0) around the cracked section, so that no shear stresses can be transferred, see Figure 6-1. Between the cracks, the steel stresses will be lower due to the effect of the surrounding uncracked concrete (this is called tension stiffening).
The crack width has its minimum at the rebar and increases with the distance from the bar, see Figure 6-1 and Figure 6-2. This variation is one of the reasons for the statistical scattering of measured crack widths, which for an example is greater for slabs with larger rebar spacing than for beams with close rebar spacing.
6-3 6-3/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
6.2.6 Flexural Cracks For a beam subjected to bending, two different types of cracks will occur, see Figure 6-3. The first cracks to be formed emerge from the face in tension and extend to the neutral axis. They are called bending cracks. When the bending moment is increased, new cracks will emerge from the face in tension to just beyond the main bars. These cracks are tension cracks. In a heavily reinforced beam with a depth (height) more than 0.4 m, these usually closer spaced tension cracks tend to join the bending cracks in the web forming a fork like crack pattern. The crack width of the bending cracks in the web above the main bars can be very high, if there is not sufficient longitudinal reinforcement in the web between the main bars and the compressive flange.
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
It must be noted that the bending cracks indicate the position of the neutral axis. This means that it is possible to compare this position with that found from calculations. Since the position of the neutral axis depends on the amount of reinforcement, measurements of the actual position can be used as a check of the reinforcement. In the same way, the measured crack widths indicate the stress level in the bars, bearing in mind that normally the crack width measured neither corresponds to the maximum load nor the dead load, but lies somewhere in between. Note that a systematic crack pattern in the surfacing on top of the bridge may be closely related to structural crack in the superstructure (for example bending cracks caused by a negative bending moment above an intermediate support for a continuous bridge).
6.2.7 Longitudinal cracks Longitudinal cracks may be formed in girders as a consequence of the stresses in the main bars giving local compression stresses in the concrete around the bars: The tension strain in a deformed main bar produces inclined compressive stresses between the concrete and the ribs of the bar. These stresses tend to split the crosssection transversely. See Figure 6-4. This kind of longitudinal cracks may occur in cases of high stresses in deformed main bars or in cases of anchorage failure at the end of a reinforcement bar (at the cross section where the number of bars change). 6.2.8 Shear Cracks In beams and slabs subjected to shear (and bending), inclined shear cracks will occur in areas at the supports, see Figure 6-5.
6-5 6-5/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Close to a simple support, the angle between the inclined shear cracks and the beam axis will be approximately 45 degrees (maximum shear, bending moment low). Often, some of the usually fine tension cracks from the bending moment tend to join the shear cracks, see Figure 6-5. In the area between mid-span and the support, the bending cracks will be more or less inclined by the shear force depending on the ratio between bending and shear, refer Figure 6-5 and Figure 6-6.
6-6 6-6/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
For T-girders, the stringer force in the compression flange will change to a tensile force above the support. This means, that possible tension cracks tend to join the shear cracks resulting in more vertical cracks just above simple supports, see Figure 6-5. The ordinary inclined shear cracks in a girder web may in some cases be connected to horizontal shear cracks, see Figure 6-7.
Shear cracks cross the total web thickness of a girder. But differently from tension cracks, shear cracks have limited strain development due to the main longitudinal rebars and the compression zone. This means, that even if the yield stress may be reached in the stirrups when the cracks are formed, a new state of equilibrium is established between stringer forces, stirrups and concrete struts, which leads to lower stresses in the stirrups after cracking. Therefore, the crack width can be rather high if the web is wide and/or the shear forces are high. 6.2.9 Bending and Shear Bending and shear cracks in girders: Shear cracks may not only appear at the support. Many older bridges (typically slabgirder bridges with two girders) have shear problems in the mid-span zone also, which are closely connected to low bending capacity in this area combined with a low shear capacity as well (few stirrups) in this area. The start of a typical "shear-bending" failure is characterised by a steep shear crack connected to a horizontal crack along the bridge deck and a horizontal crack just above the main reinforcement, forming an S-curve, refer Figure 6-8. 6-7 6-7/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The next step may be a fatigue failure in a stirrup or in one of the main bars (if there is a local splicing in the critical zone). The final step is a complete "shear-bending" failure. This type of cracks may be seen in girders in the mid-span zone as well as in the areas close to the supports. This type of cracks is serious and should immediately initiate an assessment of the load carrying capacity, which may lead to possible weight restrictions until a rehabilitation and strengthening project is prepared in order to prevent a real failure in order to save money in the repair phase. 6.2.10 Torsion Torsion causes inclined cracks similar to the shear cracks, but different from the ordinary shear cracks they are spiral and are crossing the whole depth (not only the area between the stringers) of all faces of prismatic members, refer Figure 6-9.
6.2.11 Bearing Cracks Bearing cracks are defined as cracks, which occur in connection with bearings. Some typical examples of bearing cracks are shown in Figure 6-10.
6-8 6-8/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
In (a), the elastomeric pad is placed too close to the end of the girder and a kind of a splitting crack is formed between the cover and the bent main bars. Generally, bent main bars in large dimensions give rise to bearing cracks, because the large bar size requires larger bending diameters. This means, that the edge of the end of the beam is not sufficiently reinforced, especially if horizontal forces due to e.g. temperature may occur too. This case is shown in (b). A special variant of case (b) is shown in (c). Wrong concreting or un-removed polystyrene causes friction between girder and cap beam, which results in cracks behind the bearing. In (d), a bearing crack in a cap beam is shown. This type of bearing usually occurs in connection with slab bridges with only asphaltic paper as bearing. The cracking is caused by friction due to horizontal forces (temperature) and the angle of rotation, which tend to move the reaction force to the outer edge. Many problems are connected to the concrete block rocker bearing type see Figure 6-11. Often the lead is pressed out and bearings out of position may cause either spalling or/and splitting or/and crushing of the concrete block itself, the bearing zone of the girder or/and the top of the pier. This increases the risk of unintended settlements.
6-9 6-9/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Honeycombs in the bottom of the girders due to concentration of reinforcement bars above the bearings increases the risk of local crushing and thereby unintended settlements. Narrow supports increase the risk of local spalling, local crushing, and especially of loosing the support during an earthquake. 6.2.12 Splitting Cracks Splitting cracks are related to highly concentrated loads, for an example at bearings and in the anchorage zones for the pre-stressing cables. Usually two types of splitting are considered. The first type of splitting is located very close/just below the acting concentrated force, which tries to split the concrete section locally. This is normally prevented by a "fretwork" of reinforcing bars. The second type is caused by the necessary distribution of the concentrated force to the whole cross-section, which normally will take place over a certain distance depending on the geometrical conditions, see Figure 6-12.
6-10 6-10/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
6-11 6-11/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
6.2.13 Structural Cracks, Examples In the following pages, some further examples are shown. Slabs
Columns and Piers Eccentricities between piles in pile bents - out of plane - may introduce considerable bending moments and under extreme circumstances lead to collapse.
6-12 6-12/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 6-14: Eccentricity in pile bent. Compression failure in a column will have the same appearance as a compression failure in a test cylinder in the laboratory.
6-13 6-13/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Corbels A special type of corbel is considered in Figure 6-19 and Figure 6-20 which is a projecting beam with reduced depth often seen in bridges with suspended spans. For this type, attention should be paid to the very important "lifting" reinforcement, which transfers the load to the top of the beam. In Figure 6-21, a beam with a corbel is shown.
6-14 6-14/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
6-15 6-15/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Diaphragms Special problems are related to the connection between the diaphragms and the main girders. Especially the connection between an outer main girder and a diaphragm. The problem is related to a very weak anchorage of the main bars of the diaphragm in the very often thin web of the main girder. The problem appears as spalling and cracking at the connection, and influences the load transfer between the diaphragm and the girders (lateral distribution and load transfer from the bridge deck slab to girders).
In case of more than two main girders, large bending cracks are seen in the areas close to the intermediate main girders. These cracks indicate a high stress level and that the bending reinforcement is not sufficient compared to the actual load effects. Bridge Deck Panels Quite often, deck panels with small depths are heavily cracked and potholes are developed if measures are not taken. The reasons for this damage mechanism are first
6-16 6-16/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
of all heavy loads, which lead to high tensile stresses in the reinforcement. But also too little thickness of the deck (some times less than 20 cm with 4 crossing layers of bars and varying accuracy of the position of the bars) and poor concrete quality are contributing to the problem. In thin slab panels the steel stresses are very sensitive to the thickness of the actual concrete cover. The problem starts as map cracking with cracks in both directions, which normally follow the bars because the rebars act as crack "guidance", refer Figure 6-23, section A-A and B-B. As mentioned previously, cracking is followed by loss of bond between rebars and concrete on a certain distance from the crack. When the high loadings are repeated several times, this slip distance will increase. And a mechanism which may be called "bond fatigue" will take place, which eventually leads to the separation of the concrete cover. When the concrete cover is lost, a punching shear failure may develop and finally a big pothole may develop, where only the bars remain. However, it will often be possible to detect the problem in an early stage, as cracks will show in the bridge surface above the girders.
6-17 6-17/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
6-18 6-18/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
6-19 6-19/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 6-26: Deflections caused by large span lengths in RC concrete (creep) and improper formwork.
Retaining walls
Figure 6-27: Deflections caused by low stiffness (creep), compaction or soil conditions.
6-20 6-20/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Bearings
Figure 6-29: Crushing caused by honeycombs, wring type of bearing or poor workmanship.
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Vertical clearance
6-22 6-22/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
6.2.14 Structural Problems, Steel Components Truss Bridges The most common damage to truss bridges with interior passage are impacts to the lateral braces from trucks either because of too low vertical clearance or because the total height of the vehicle exceeds the allowed height. The consequences may vary from almost harmless if only secondary braces are damaged to very harmful/collapse if the primary components are damaged. Even though only the braces are hit, it may introduce problems for the main components. If for an example the braces are connected to vertical compression members, the impact may create a permanent deformation of the compression members making these components more or less useless (and introducing a redistribution of the load effects through the structure), refer to Figure 6-33.
6-23 6-23/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 6-33: Impact damage to truss bridges Stability problems (buckling) in compression members - especially vertical - may also occur due to overload introducing a similar redistribution of the load effects as mentioned above. In both cases, a load capacity evaluation should be carried out to evaluate the risk of a possible collapse. Special attention should be paid to the connections/nodes between the different components (tension/compression components and horizontal beams components), since they are disposed for corrosion due to accumulation of dust and humidity. Plate Girders (including plate girders in steel-concrete composite structures) A common type of damage to plate girders is impact from vehicles with heights exceeding the actual vertical clearance. Girders may be seriously affected by such an impact, which means that the load effects for some of the remaining girders will be increased. This is not an acceptable situation and remedial actions should be carried out.
6-24 6-24/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 6-34: Fatigue problems in welded steel members. Many plate girders are welded. It is normal to find stress peaks at welds. The stress level depends of the type of weld and the quality of the welding. If the stress range from the live load is considerable at a poor welding, there may be a risk of fatigue cracks. Therefore welded details, which may be critical in fatigue should be inspected carefully. This may for example be at a welding in the bottom flange of a girder, refer Figure 6-34. Or at a welded connection between the horizontal wind truss and the bottom of the girders. Such details may at the same time also be critical to corrosion. Arch Bridges Attention should be paid to the hanger connections to the main girders and the arch. Look for corrosion of the hanger cables, which normally have a high stress level and therefore may be exposed to stress corrosion with the consequence of a brittle behaviour. Look also for other connections for an example between the longitudinal girders and the transverse beams.
6-25 6-25/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
6.3
6.3.1 Shrinkage cracks (due to drying) The appearance and development of shrinkage cracks depend on the geometry, the size of the member and possible restraints. The crack orientation is normally well defined and depends on the geometric conditions (e.g. restraints caused by other parts of the structure). Drying shrinkage cracks pass through the whole cross section. Normally, these cracks are harmless from a structural point of view. But they may be harmful with regard to durability. Some examples are shown below.
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
6-27 6-27/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
6.3.2 Thermal Cracks (due to hydration) The appearance of thermal cracks caused by thermal stresses due to temperature differences in the hardening concrete is very similar to the appearance of ordinary shrinkage cracks (geometric conditions). However, different from ordinary shrinkage cracks and structural cracks, thermal cracks are young cracks (developed in the young concrete). This means that the cracks will follow the surface of the coarse aggregates and stones, and not go through them. 6.3.3 Cracks due to plastic shrinkage Plastic shrinkage cracks are caused by rapid drying of the concrete surface (low humidity, wind, high temperature) in its plastic state (e.g. caused by improper curing). Similarly to the temperature cracks, the cracks will follow the surface of the stones, not go through them. The cracks are normally wide and shallow and may form a definite pattern. In cases of a wide surface (concrete wearing course or similar), a state of hydrostatic tension will arise (no shear stresses). If there is no crack "guidance", the crack will be formed "at random" and the appearance will be a net crack pattern, in most cases with hexagonal meshes. Typical crack patterns caused by plastic shrinkage are shown in Figure 6-40-Figure 6-42. Plastic shrinkage cracks are normally harmless from a structural point of view (although wide cracks may influence the load carrying capacity and the behaviour under service load), but may be harmful to durability.
6-28 6-28/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
6.3.4 Cracks due to plastic settlement Normally, these cracks are due to a high concrete slump when cast. The appearance and position of these cracks is normally above the reinforcement at the surface, refer Figure 6-43, or at changes in the cross section, refer Figure 6-44. Plastic settlements are also seen in slabs with voids. Normally, the cracks are harmless from a structural point of view, but may be very harmful with regard to durability. This is because the bars are not sufficiently protected against environmental effects (the cracks often reach the bars).
6-29 6-29/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
6-30 6-30/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
6.3.5 Initiation of Corrosion In the highly alkaline environment in concrete (pH-value close to 13) the anodic steel surface becomes coated with a very thin grey passive layer of ferric oxide. The ferric oxide is stable over a wide range of potentials and so acts as a protective coating. Thus, reinforcement is protected against corrosion when embedded in a concrete of a good quality and with a sufficient cover. But, the protection against corrosion is not everlasting. The surroundings will always affect the concrete and finally lead to a breakdown of the passive layer. The breakdown of the passive layer may be caused by free chlorides at the reinforcement or by carbonation of the concrete cover. These mechanisms are described in the following sections. However, corrosion depends on moisture content and the availability of oxygen and therefore of the rate at which oxygen diffuses through the concrete. The period during which the passive layer breaks down is normally called the period of initiation. The duration of the initiation period depends on: The thickness of the concrete cover; the thinner the cover, the shorter is the period of initiation. The quality of the concrete cover (primarily water/cement ratio dependent); i.e. the initiation period decreases when the concrete quality gets poorer (the water/cement ratio increases). In special cases (honeycombs, "cold joints", too small cover), poor workmanship can lead to corrosion immediately after casting. The aggressiveness of the environment, the temperature and the humidity. The kind of mechanism causing deterioration. Chloride penetration is by far the most aggressive mechanism, leading to a much shorter initiation period than the mechanism of carbonation (chloride ions facilitate the corrosion process). During the period of initiation there is no actual corrosion going on. The protection of the reinforcement is being broken down with no visible signs of deterioration, neither on the surface of the concrete nor on the reinforcement. Therefore: The risk of future corrosion damage can only be assessed by performing special investigations. Four steps of corrosion of steel in concrete may be defined as: 1. 2. The passive state. Pitting corrosion. 6-31 6-31/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
3. 4.
These relate to corrosion as an electrochemical process which requires a potential difference between two connected electrodes in an electrolyte. In concrete the electrodes may be neighbouring points on the same reinforcing bar, or separate bars or groups of bars which have a potential difference between them. 6.3.6 Carbonation Carbonation is caused by the carbon dioxide (CO2) in the air. The CO2 reacts with the calcium hydroxide, Ca(OH)2, in the cement paste eventually leading to a critical decrease of the alkalinity. The pH-value decreases to less than 9, which normally is insufficient to protect the reinforcement against corrosion. The reaction in which calcium hydroxide is converted to calcium carbonate is as follows: Ca(OH)2 + H2CO3 CaCO3 + 2H2O 3CaO2SiO23H2O + 3CO2 3CaCO32SiO23H2O These reactions consist of the following elementary steps: 1. The diffusion of atmospheric CO2 in the gaseous phase of the concrete pores. 2. The dissolution of solid Ca(OH)2 from cement gel into the pore water and the diffusion of dissolved Ca(OH)2 from regions of highly alkalinity to those of low. 3. The reaction of dissolved CO2 with dissolved Ca(OH)2 in the pore water. 4. The reaction of dissolved CO2 with CSH. The effective diffusivity, De, CO2, of CO2 in concrete is given by the following empirical expression:
De ,CO2
RH = 2.1 10 p1 100
6
2.2
Where p is the porosity of the hardened cement paste and RH is the ambient relative humidity. The speed of the reaction will depend on the rate of removal of water formed. In other words, carbonation depends on a drying atmosphere and are impeded in the presence of water. On the other hand dry CO2 does not react with dry Ca(OH)2 so the presence of moisture is essential to the carbonation process. The optimum moisture content for carbonation is intermediate between 40 and 70 % relative humidity. An increase in temperature with 10 oC will approximately double the speed of the reaction. Carbonation of concrete results in increased strength and reduced permeability, possible because water released by carbonation aids the process of hydration and CaCO3 reduces the voids within the cement paste.
6-32 6-32/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The carbonation depth in concrete, x in mm, is given by the following empirical equation:
Figure 6-45: Depth of carbonation in relation to humidity. Example: Compressive strength: 20 MN/m2 Age of structure: 25 years Relative humidity: 65 % RH
1 K = 72 - 0.126 = 7.03 20
6-33 6-33/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
K must be multiplied by 0.95 (Figure 6-45, RH = 65%). An estimate of the carbonation depth will be:
x= K T K=x / T c= K T1
2 2 T c c =T T1= = c K x x 2
where: x c K T T1 = = = = = the actual depth of carbonation (mm, measured on site) cover (mm) constant age of the concrete (years) period of initiation (years)
Example: Carbonation depth, measured: Concrete cover, measured: Age of the concrete:
35 T 1 = 20 39(years) 25
6.3.7
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Chloride induced reinforcement corrosion is in many areas considered the main durability problem for reinforced concrete structures. The amount of chloride necessary to initiate reinforcement corrosion (the critical chloride concentration or the threshold value) depends, among others, on the composition of the concrete and the moisture content of the concrete, see Figure 6-46.
Figure 6-46: Critical chloride content as a function of relative humidity. Chlorides in the concrete may originate from various sources: the mix water, the aggregates, admixtures curing water, surrounding soil (from which chlorides are washed out in wet periods), de-icing salt (in cold areas), in coastal areas from the seawater (reaching the concrete directly or air-borne in windy periods). In general, most of the chlorides contained in fresh concrete ("initial chlorides", i.e. chlorides from mix water, aggregates and some from the curing water) will be chemically bonded during the hardening of the concrete. Bonded chlorides are not regarded as harmful while remaining bonded, as they are not breaking down the passive layer on the steel. Threshold Value 6-35 6-35/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The threshold value is the chloride value where the chloride concentration is so large that corrosion occurs in the solution can be approximated as follows:
C Cl = 0.61 C OH
Where CCl and COH is the concentrations of hydroxide and chloride ions in equivalents per litre, respectively. The hydroxide concentration can be calculated as follows:
C OH
c ( Na ) c (K ) + 23 39 100 = P
Where c is the cement content (kg/m3), (Na), (K) is the weight share Na and K respectively in cement, and P is the porosity of the concrete in % by volume. Normally, there are strict requirements to the maximum chloride content of fresh concrete. Further more sufficient protective properties of the cover (denseness and cover thickness) should be selected considering the environmental exposure. If the desired protective properties of the cover cannot be obtained, additional protective means need to be applied. Diffusion of chlorides into concrete Ingress of chlorides may take place by: Capillary suction or permeation of water containing chloride. Diffusion of chloride ions in the pore liquid.
Based on Ficks 2nd law diffusion of chlorides into concrete can roughly be described by the following two equations:
x = Kx T
(1) (2)
x C x = C s (C s C i ) erf 2 D T
Where: x = Kx = T = Cx = Cs = = Ci Erf = D = depth below concrete surface (mm) constant age of concrete (years) chloride content at depth x chloride content at the concrete surface initial chloride content Error function diffusion coefficient (mm2)
6-36 6-36/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The chloride diffusion coefficient of ordinary Portland cement concrete (w/c > 0.4) in a Danish environment may be obtained form the following equation: D = 2000 (w/c 0.35) [mm2/year]
The Error function takes the following values. x 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 erf(x) 0 0.122 0.223 0.329 0.428 0.521 0.604 0.678 x 0.8 0.9 1.0 1.2 1.4 1.6 2.0 2.4 erf(x) 0.742 0.797 0.843 0.910 0.952 0.976 0.995 0.999
Based on measurements of the chloride profiles in the concrete, the different parameters can be calculated giving a prediction of the time until corrosion occurs. The calculations will be similar to the calculations regarding carbonation. A typical chloride content profile looks as follows:
Figure 6-47: Chloride content profile. In this example, the initial chloride content has been approximately 0.02 %. 6-37 6-37/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
A prediction of the service life depends on the knowledge of a critical chloride level. Each structure is supposed to have its own critical limits, because the limit is dependent on various factors such as humidity and concrete quality. As a preliminary assumption, 0.05 % of dry concrete weight can be used as the critical limit in normal concrete. For piers in seawater a value of 0.10 % of dry concrete in normally used. 6.3.8 Carbonation and Chlorides For concrete with a high initial content of chlorides, the chemically bonded chlorides in front of the carbonation front will be freed. Since carbonated concrete provides almost no resistance against chloride penetration, the chlorides will be break through the carbonation front (diffusion). Therefore, the chloride content will increase constantly behind the carbonation front when the carbonation front is moving into the concrete, leading to corrosion when the critical limit of chloride content is reached at the reinforcement. A typical chloride content profile in a concrete with a carbonation depth of approximately 30 mm is as follows:
Figure 6-48: Chloride content profile with carbonation. If this mechanism takes place together with alternating wetting and drying in a chloride contaminated environment, the corrosion process can run very fast (e.g. in tidal waters). 6-38 6-38/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
A simple method for evaluating the chloride penetration from outside is, in this case, to ignore the carbonated concrete layer. The thickness of the carbonated layer can be calculated from the previous mentioned equations.
6.3.9 Propagation of Corrosion When the chloride content at the reinforcement level reaches the critical limit or the carbonation front reaches the reinforcement, the passive layer is broken down and the corrosion process starts. A corrosion process is an electrochemical process, where a current runs between corroding areas (the anodes, where the passive layer has been broken down) and non-corroding areas (the cathodes, where the passive layer is complete). The current runs due to the theoretical fact of the behaviour of metals in liquids and concrete, see Figure 6-49. If two metals with different electrochemical potential are electrically connected, corrosion is likely to take place on the metal with the lowest potential. Therefore, the risk of corrosion can be evaluated by measuring the potentials (HCPmeasurements, see Chapter 7.6).
Figure 6-49: The electrochemical series. When the passive layer breaks down locally, the area changes its place in the electrochemical series becoming more negative. This creates a potential difference, which generates the corrosion current. Where the passive layer is broken, the following "anodic reaction" takes place. Anodic reaction: Fe -> Fe++ + 2e6-39 6-39/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
If water and oxygen is present at the steel surface, the "cathodic reaction" takes place. Cathodic reaction: O2 + 2H2O + 4e- -> 4OH-
The anode and the cathode may be far apart, as long as there is an electric connection between them. The anodic process produces electrons, and the cathodic process consumes electrons. If there is an electric (through the reinforcement) and electrolytic (through the moist concrete) connection between the anode and cathode, an electric current will flow. The corrosion process is illustrated in Figure 6-50.
Figure 6-50: The corrosion process. 6.3.10 Corrosion products and corrosion rate The Fe++ reacts with oxygen, OH- and water, forming corrosion products. The type of corrosion products produced primarily depends upon the available amount of oxygen and water. If there is little available oxygen, the first products will be white Fe(OH)2. These white products may be transformed into black FeO and water. This type of corrosion product is typical for local, chloride initiated corrosion. 6-40 6-40/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
If oxygen is added to FeO, black Fe3O4 may be formed. Fe3O4 is not expansive, so there are not necessarily any exterior signs of corrosion. If water and oxygen become available to the white Fe(OH)2, the products turn via a green intermediate stage into brown Fe(OH)3. If additional water is available, expansive yellow/red/brown Fe(OH)3,nH2O (rust) is formed. If oxygen is plentiful, the expansive yellow/red/brown Fe(OH)3,nH2O (rust) is formed without any intermediate stages. This type of corrosion product is typical of carbonation initiated corrosion in porous concrete. The development of the corrosion attack and the velocity of the process primarily depend upon: The temperature. The ratio between corroding and non-corroding areas. The moisture content.
As for most chemical reactions the corrosion rate increases with increasing temperature. If the area of the anode is small compared to the area of the cathode (Aa/Ac << 1) the corrosion rate will be high, because the corrosion takes place in a small area (local corrosion). The process of local corrosion normally leads to corrosion products, which are black, non expansive and liquid-like. This means that local corrosion cannot be expected to give visible signs of corrosion on the surface. Local corrosion is hidden corrosion where serious attacks can be developed without visible signs, increasing the risk of unexpected collapse. In case of chloride attack, the chloride ions facilitate the formation of Fe++, thus increasing the corrosion rate. As chloride initiated corrosion normally starts as local corrosion, a considerable reduction of cross section of a steel bar may take place within a short time and without visible signs on the concrete surface. A ratio between anode area and cathode area, Aa/Ac approx. 1, gives what is called general corrosion. This type of corrosion leads to the well-known yellow/red/ brown corrosion product, which quickly gives rise to visible signs of corrosion (spalling of the concrete, cracks, rust stains). The corrosion rate depends on the moisture content and the electrolytic contact between anode and cathode. The RH % below refers to the moisture content in the pores of the concrete at the anode area (the electric resistance at the anode is decisive for the corrosion current, and the resistance drops with increasing moisture con-
6-41 6-41/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
tent. In outdoor concrete there will normally be sufficient water for the cathode process). The moisture content in the concrete in average over a year can be different from the RH % in the air, especially in case of the presence of chlorides in the concrete. At RH % less than 80, the corrosion rate is negligible. Normally the corrosion rate drops at RH above 95% and reaches 0 at RH = 100%. See. But if there is electric contact to areas with less moisture content, a cathode may be formed in this area and the corrosion rate will increase.
This may be the case when a column is partly submerged in saturated soil. Corrosion may take place at an anode area below ground (RH=100%), because the cathode process takes place above ground at a lower humidity. The two types of corrosion can normally be located at different places in the structure.
6.3.11 Local/general corrosion Local Corrosion (giving no visible signs on the surface) is typically located: at restrained cross-sections (max. negative bending moment), e.g. brackets and cantilever decks, 6-42 6-42/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
at construction joints and cracks in concrete with high content of chloride and moisture, in concrete with a high initial content of chloride.
General Corrosion (giving visible signs at the surface) is typically located: in concrete of poor quality, especially in the cover concrete; in concrete with insufficient thickness of the cover; in concrete, where alternating wetting and drying takes place, such as horizontal surfaces (especially soffits) and vertical surfaces above sea- or ground level. In practice, most structures (except structures submerged in water or saturated soil) will be exposed to both wetting and drying. In the early phase, corrosion will normally develop as local corrosion. In the later phase, general corrosion follows with evident signs of corrosion at the surface. In structures with corrosion caused by chlorides, the corrosion (i.e. the reduction of cross section) will normally be critical in the later phase. Three typical cases of corrosion (bridge deck, column and wing wall) are described in the following.
6.3.12 1)
Bridge Deck
Areas with cracks caused by negative bending moment: The corrosion will develop as local corrosion with a high corrosion rate and consequently reduction of cross section.
2)
Areas in which the corrosion in the first phase will develop as local corrosion and in later phases will lead to more general corrosion followed by spalling of the concrete. 6-43 6-43/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The development of corrosion depends on the moisture content and the duration of wet and dry periods. In decks with no overlay/wearing course in hot climate and with only occasional water influence (alternating wetting and drying), the corrosion will quickly develop as general corrosion with expanding corrosion products. The time until corrosion attack becomes visible on the surface is normally shorter but the damage to the reinforcement less than in decks with a constantly high water content (e.g. in case of an overlay which is not waterproofed, allowing water to penetrate but not to evaporate again). In decks with an overlay, which is not waterproofed, the corrosion conditions are at optimum. In this case, visible signs of corrosion normally occur at a very late phase when the overlay cracks (or when the overlay for other reasons is replaced). A high rate of corrosion can be seen, even if there is a lack of the necessary oxygen at the upper layer. This is because the oxygen-consuming cathodic reaction can take place on the lower reinforcement layer. 3) Normally, the risk of corrosion in the soffit is limited except from areas with water leakages, especially with chloride contaminated water. If larger areas suffer from water (and chloride) influence, the mechanism of damage will be equal to case 2) giving rise to spalling of the cover.
6.3.13 1)
Pier column
Areas with the risk of local corrosion especially in constantly wet and chloride contaminated concrete. In concrete exposed to alternating wetting and drying, the corrosion product will expand causing cracks and spalling of the cover.
2)
General corrosion, mainly caused by carbonation. The corrosion will develop slowly, but visible damage will appear at an early stage.
6-44 6-44/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
3)
Areas at risk of local corrosion. The lack of oxygen does not prevent corrosion as the cathodic process will take place at the steel areas above ground level. The corrosion rate can be very high.
In general, most of the chlorides contained in fresh concrete ("initial chlorides", i.e. chlorides from mix water, aggregates and some from the curing water) will be chemically bonded during the hardening of the concrete. Bonded chlorides are not regarded as harmful whilst remaining bonded, as they are not breaking down the passive layer on the steel. Diffusion of chlorides into concrete can roughly be described by the following two equations:
x= K x T
(1)
C x = C s - ( C s - C i ) erf 0.5 x / TD
(2)
6.3.14 Wing walls / retaining walls Wing walls may suffer from several types of corrosion attack:
6-45 6-45/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
1)
At ground level and below, the conditions are similar to those of a column (example 1).
2)
At the top, there is a risk of very high chloride content originating from surface water running from the overpassing road and the slopes (through chloride contaminated soil) to the top and the upper part of the back side of the wall. Due to alternating wetting and drying, visible signs of corrosion of the reinforcement are seen as vertical splitting at the top. This splitting causes easier access to water, which increases the rate of the corrosion process, etc.
3)
On the backside of the wall, there may be a constantly high moisture and chloride content level. Even if there is a lack of oxygen, severe corrosion may occur. (The situation is similar to that of a column footing).
4)
At the front of the wall, the chloride content may be very high, because saline water penetrating from the back side evaporates, leaving chlorides in the concrete.
5)
At all surfaces exposed to the air, carbonation initiated general corrosion may occur.
Note Any investigation of corrosion problems on out-door concrete structures has to include an evaluation of: The actual corrosion level and -activity The future risk of corrosion
Both in areas with evident corrosion and in areas without visible signs of corrosion (especially in areas where corrosion normally is expected to take place), the investigation should be carried out as described in Chapter 4.2.
6.4
Alkali-aggregate Reactions
In humid concrete, alkalis from the cement (or from the surroundings) may react with siliceous aggregate. The process may briefly be described as follows: Alkali ions (Na+ and K+) from cement, mix water or from the surroundings (on bridges, typically seawater or de-icing chemicals) lead to an increase of the concentration of OH- in the water in the pores of the concrete. OH- dissolves amorphous silicon (SiO2) which may be present e.g. in flint aggregate, producing a gel.
6-46 6-46/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
This gel absorbs possible water from the surroundings and expands. Under certain circumstances the expansion pressure leads to cracks in the concrete, forming a crack pattern on the concrete surface. If the reactions take place in coarse aggregate near the concrete surface, the result may be 'pop-outs'. See Figure 6-55. 6.4.1 Crack Pattern In general, concrete attacked by alkali silica reactions (or some other deterioration processes) will expand and crack in 'the easiest way'. This means that: The coarsest cracks will form parallel to the section with the weakest reinforcement. Weaknesses in the design of the reinforcement will be disclosed. E.g. very coarse cracks may appear if a cross section is not (or very weakly) reinforced. Normally, a system of internal cracks parallel to the concrete surface is formed. This kind of cracks may be detected by drilling out cores, or by more sophisticated techniques, such as the 'Impact-Echo' method (see Chapter 7 regarding test methods). The interaction between internal tension stresses from the alkali-aggregate reactions and from the load induced forces reveals the path of the internal forces in the structure: The cracks will follow the direction of the internal stresses, particularly clear at concentrated loads. See Figure 6-12. The conclusion on the above damage pattern is that serious carrying capacity problems may occur if the design of the reinforcement is insufficient. (E.g. un-reinforced sections or too small overlap at splices). If the reinforcement design is good (closed stirrups, sufficient overlaps, U-stirrups at beam ends, transverse reinforcement at anchorage zones etc.) alkali-aggregate reactions do not reduce the carrying capacity considerably.
6-47 6-47/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 6-55: Alkali silica reactions. The gel may form drops emerging from the cracks. The gel is soluble in water, therefore on outdoor structures it is washed away by rain. But on indoor structures it may be seen. Alkali-aggregate reactions can only take place if the concrete contains alkalis, calcium hydroxide, silica and water see Figure 6-56. It has to be noted that even if the concrete does not contain enough alkalis the reactions may take place anyway if alkalis are supplied from the surroundings.
Ca(OH)2 AAR
Water (H2O)
tions.
To prevent damage, one or more of the four components must be shut out or limited to an acceptable level. On existing structures with alkali-aggregate problems, the only practicable way is to keep water away from the concrete, either by putting up a shield, or by applying a surface protection (waterproofing, but permitting diffusion of vaporised water).
6-48 6-48/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
It is not possible to establish exact threshold values, but the following may serve as a guide regarding the humidity: If the relative humidity in the pores of the concrete is about 80-95% RH, the gel is solid, and the expansion pressure may cause cracks and pop outs. If the humidity is higher, the gel is liquid. It will fill pores and cracks and it may penetrate to the surface, but it will not cause serious damage. If the humidity is lower than about 80%, the reactions are not expected to take place. On new structures (and on rehabilitation works on existing structures), alkaliaggregate reactions must be prevented by using aggregate which does not react with alkalis, and/or using cement with a low alkali content and at the same time protect the structure from alkalis from external sources. If no knowledge is available regarding the reactivity of the aggregates, it must be determined by tests. It is very important in preventing alkali-aggregate reactions that a great effort is made to avoid initial cracks in the concrete (see Chapter 6.3). The alkali-aggregate reactions may take place for a considerable time (years) before visible signs of damage occur. In time, the internal stresses have grown to a magnitude that causes expansion and cracking of the concrete. The cracking and expansion go on for some time (years) after which they die out typically caused by lack of alkalis or reactive material. To evaluate the risk of future development of damage due to alkali silica reactions a residual reactivity test can be performed see section 7.18. The deterioration rate may rise considerably if frost/thaw action takes place at the same time, because freezing water creates internal stresses in the cracks initiated by the alkali silica reactions. The only reliable way to identify alkali-aggregate reactions with certainty is by means of specific microscope techniques in the laboratory (microscope inspection of thin slices of a drilled out concrete core) see section 7.18.
6.5
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
6.5.1 Acid Attack Pollutant gases are increasing drastically in the atmosphere. These air contaminants are usually found in gas form (SO3, HNO3, HCl, O3 and organic acids), in particle form (H2SO4, NH4SO4, (NH4)2SO4) or dissolved in water as droplets (CH+, NH4+, HSO3-, SO42-, NO2-, NO3-). They are fairly reactive and may cause degradation on concrete and masonry. Measurements in Gothenburg in Sweden have shown that pH-value varied between 3.0 4.5. Degradation of concrete may occur as the result of the dissolution of the cement hydration products causing a more porous weaker matrix as a result of internal expansion associated with gypsum formation and its subsequent reactions. Pure water can dissolve the calcium products (calcium hydroxide) in concrete. Theoretically the hydrolysis of cement paste can continue until most of the calcium hydroxide has been leached away. This exposes other cementitious constituents to chemical decomposition. The process leaves behind silica and alumina gels with little or no strength. The hydrated calcium silicate, which is alkaline, is neutralised by contact with hydrochloric acid water: 2HCl + Ca(OH)2 CaCl2 + 2H2O Calcium chloride made by the reaction of hydrated calcium silicate and hydrochloric acid is very soluble to water. All acids attack concrete, and the cement paste is degraded generally as in the example with hydrochloric acid: 2HX + Ca(OH)2 CaX2 + 2H2O where X is the acid residue. Exposure to NOX, while also lowering pH, results in the formation of calcium nitrate hydrates. The calcium nitrate hydrates volumes significantly exceed those of the solid reactants; however, these nitrates are highly soluble and readily leached. As a consequence, the principal durability issue in this regard is the development of a more porous matrix, in turn, more susceptible to other forms of attack. Nitric acid will react with the calcium hydroxide and form calcium nitrate: Ca(OH)2 + 2HNO3 + 2H2O Ca(NO3)24H2O Exposure to SO2, either alone or in combination with elevated levels of CO2, is likely to lead to severe disruption. pH is lowered and the formation of gypsum results in expansion in cement. In addition the subsequent reactions of gypsum with the hydration products of cement result in additional expansion.
6-50 6-50/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Sulphuric acid will attack calcium hydroxide with the formation of gypsum Ca(OH)2 + H2SO4 CaSO42H2O Water containing free CO2 attacks concrete and produces Ca(HCO3)2 which is very soluble. Without considerations of stoichiometry the following may be described: H2O + CO2 + Ca(OH)2 Ca(HCO3)2 Or attack on carbonated calcium hydroxide H2O + CO2 + CaCO3 Ca(HCO3)2 Acid Carbonic Sulphuric Nitric Reactant Ca(OH)2 (OH)2 Ca(OH)2 Product CaCO3 CaSO42H2O Ca(NO3)24H2O Volume change (%) 11 123 274
Table 6-1: The changes in the volumes of the solid phases associated with the attack of calcium hydroxide and calcium silicate hydrate by nitric, sulphuric or carbonic acid. Rainwater contains dissolved carbon dioxide forming a very mild acid which dissolves calcium carbonate by producing soluble bicarbonate. Lime mortars will eventually be destroyed by percolating rainwater because calcium carbonate is their main binding agent. 6.5.2 Sulphate Attack Water containing sulphates attacks the concrete in different ways. When sulphate ions come in contact with the hydration products, a chemical reaction starts. In most cases, this reaction results in volume expansion. 1. High concentrations of sulphate ions (SO4- -): SO4- - + Ca(OH)2 + 2H2O -> CaSO42H2O (gypsum) + 2OH- + volume expansion 2. Lower SO4- - concentrations: Calcium Aluminate Hydrate + CaSO42H2O -> 3CaOAl2O3CaSO432H2O (ettringite) + volume expansion 3. Most serious: Magnesium and Ammonium Sulphate: MgSO4 reacts with CaO3Al2O3's hydrate products 6-51 6-51/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
and MgSO4 + Ca(OH)2 -> CaSO4 + Mg(OH)2 + volume expansion In case 1 and 2, hydrated calcium silicates will be transformed into gypsum and ettringite, which means that the concrete will lose its strength. Further, the cement paste will expand, which causes cracking and disintegration of the concrete surface (scaling). Risk of Sulphate Attack The risk of sulphate attack depends on the following parameters: The content of sulphate in the surroundings (seawater, ground water, sewage, soil). The aggressiveness of water and soil can be divided into three groups: o o o Moderate: water with less than 300 mg SO3/l or soil with less than 0.2 % SO3. Aggressive: water with 300 - 1000 mg SO3/l or soil with 0.2 - 0.5 % SO3. Very aggressive: water with more than 1000 mg SO3 or soil with more than 0.5 % SO3.
The moisture content of the concrete. The reaction needs the presence of water. The type of cement. Especially the C3A content of the cement. Sulphate contaminated aggregates. The permeability and/or the ability of capillary suction (e.g. from the buried part of a column to the free part). The sulphate concentration will constantly increase in the evaporation zone leading to a rapid deterioration. Capillary suction may also lead to a general sulphate attack on the concrete in the whole cross section.
A special case of both sulphate attack and corrosion is seen on partly submerged structures, especially structures with large dimensions. Partly submerged structures normally have the biggest problems in the splash zone. However, if the cover below sea level disintegrates due to sulphate attack, one big anode is formed under the sea level. The cathode will be the part above sea level. In structures with large dimensions, a large amount of current flows, leading to severe corrosion attack on the submerged part.
6-52 6-52/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
In masonry sulphate attacks are the most common problem and is due to the reaction between sulphate ions in water solution and the tricalcium aluminate (C3A) phase in mortars to form calcium sulphoaluminate or ettringite. The commoner sulphates are the sodium, potassium and magnesium salts, which are all freely soluble, and calcium sulphate, which is less soluble but will leach in persistently wet conditions. The sulphates may be present in groundwater and can effect masonry below the waterproofing and masonry in contact with the ground such as retaining walls, bridges and tunnels. Sulphates are also present in some types of clay bricks and will be transported to the mortar in wet conditions. Sulphates do not attack pure lime mortars as there is no calcium aluminate present but may have some effect on hydraulic lime mortars. 6.5.3 Seawater Attack Concretes exposed to marine environment may deteriorate as a result of the combined effects of chemical action of sea water constituents on cement hydration products, alkali silica expansion, crystallisation of salts in concrete, corrosion of embedded steel and physical erosion due to wave action and floating objects. Direct chemical attack comes from the magnesium salts in seawater. The concentrations may be low but they are sufficient to produce calcium chloride, gypsum (both soluble) and ettringite. In the following the action of magnesium and calcium salts are given. Action of sulphate: MgSO4 + Ca(OH)2 CaSO4 + Mg(OH)2 Where calcium sulphate can be soluble or solid and may act in a secondary reaction and produce ettringite: CaSO4 + C3A + 32H2O C3A3CaSO432H2O Action of chloride, MgCl2: MgCl2 + Ca(OH)2 CaCl2 + Mg(OH)2 The action of calcium chloride can result in producing chloroaluminate, ettringite and thaumasite with large expansions as result: Chloroaluminate: CaCl2 + C3A + 10H2O C3ACaCl210H2O 6-53 6-53/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Interestingly, in spite of the high sulphate content of seawater, and even with high C3A Portland cement and large amounts of ettringite from sulphate attack, little expansion is usually present. The deterioration is usually characterised by erosion or loss of solid constituents form the mass. It has been suggested that the ettringite expansion is suppressed in environments where OH- ions have essential been replaced by Cl- ions. Concrete between the tide marks, subjected to alternating wetting and drying, is severely attacked, while permanently immersed concrete is attacked less. The actual progress of attack by seawater varies, and is slowed down by the blocking of the pores in the concrete through deposition of magnesium hydroxide. In warmer environments the attack is more rapid.
6.6
Erosion / Scour
Scour is the erosive action of running water, excavating and carrying away material from the bed and banks of waterways. Scour is one of the most frequent causes of bridge failures, mainly because it may develop to a very large extent within a short time. If the level of the riverbed has changed significantly in general or around piers/abutments there is always reason to carry out closer investigations. Note that there may very well be problems, even if the erosion has not reached the level of the underside of the foundation. In many cases the load carrying capacity of a direct foundation is dependent on the pressure (the weight) from the surrounding soil. And particularly pile foundation depends on the surrounding soil. Scour problems may be divided into three groups: Aggradation / degradation General scour Local scour
6.6.1 Aggradation / degradation Aggradation and degradation are long term changes in the level of the riverbed. Aggradation is deposition of material, elevating the riverbed, while degradation is the lowering of the bed caused by erosion.
6-54 6-54/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Aggradation reduces the cross section of the waterway. This will cause the surface of the water to rise, and during a flood, the superstructure may be affected by streaming water and debris, causing unintended horizontal forces to the bridge. Degradation may lead to undermining of the foundation, eventually leading to failure of piers / abutments and thus of the whole bridge. 6.6.2 General scour General scour is characterised by the removal (erosion) of material from the whole width of the waterway. Generally, it is caused by increased water speed. There is no strict distinction between general scour and degradation, but in general, degradation is a slow erosion of material over a long period of time (years), while general scour may take place over a shorter period. General scour often occurs because of a contraction of the flow of water. It may be a result of the construction of the bridge, as piers, abutments and embankment slopes reduce the cross section of the waterway channel. However, it may also be caused by obstructions or other changes in the waterway, upstream or downstream. Another possible cause of general scour is mining in the riverbed, i.e. excavation of sand and gravel. 6.6.3 Local scour Local scour is scour that only affects a minor part of the width of the waterway. Generally, it occurs where obstructions (natural or artificial) change the flow of water, creating accelerations and vortex systems. The occurrence of local scour very much depends on the design of the obstructions to the water flow (Piers, abutments). If a scour protection only covers part of the riverbed, local scour may occur at the edges of the protection. Particularly if the protection is of a solid type like concrete or asphalt. Open, flexible types of protection (wire mattresses, riprap) are less vulnerable to local scour. Scour is prevented by: Minimising the reduction of the waterway cross section caused by the bridge. Not making regulations of the stream (upstream or downstream) that cause increasing water speed at the bridge. Giving structures in the stream (piers and abutments, scour protection) designs that minimise the formation of vortexes. Protecting riverbed and slopes. 6-55 6-55/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
6.7
6-56 6-56/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Oxygen acts as electron acceptor by creation of hydroxyl ions with the free electron and water. The formed OH- reacts with the free Me++ and will create rust products. For instance if we look at corrosion of iron in water: The iron reacts with the following process: Fe Fe++ + 2e- (anode process) From the atmosphere the water uptake oxygen and together with the free electron is created hydroxyl ions: 2H2O + O2 + 4e- 4OH- (cathode process) Fe++ from the anode process and OH- from the cathode process creates Fe(OH)2. This product is not stabile in oxidise atmosphere; it is oxidised further to the readbrown Fe(OH)3 which is able to uptake water and create rust Fe(OH)3nH2O. It is the last created product that result in the typical read-brown colour of the rust. Iron is stable as long as the air relative humidity is below 65 %. Over 65 % will the water film, which is on the surface, be so thick that it is able to act as an electrolyte.
6-57 6-57/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Water Drop
Steel
Figure 6-57: As example on the rust creation we will look on a water drop on a steel surface. The water drop will result in visible rust after a few hours. The rust will be formed ring shaped around blank steel. This due to the following: The water drop will dissolve oxygen for the atmosphere, however in the marginal zone where the water layer is thin the oxygen reach the steel very fast and act as an electron acceptor according to the cathode process. Inhomogeneities in the steel surface result in a certain area to be come anode and dissolve Fe++ while the remaining e- runs though the steel and leaves it near the marginal zone where it is up taken by the acceptor. The circuit is working and the steel corrodes anodic in the centre and a rust ring of iron hydroxide is created around the anode area. The rust ring will grow to a rust hillock that will cover the blank anode. Be aware that oxygen is necessary for the process to continue but no corrosion takes place where the oxygen is supplied.
6-58 6-58/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Electrochemical series Thermodynamic series Metal/Cation Normal potential in volt at 25 C against hydrogen electrode -2,34 -1,67 -0,76 -0,74 -0,44 -0,40 -0,25 -0,14 -0,13 0,00 +0,34 +0,52 +0,80 +1,12 +1,68
Metal
Potential in volt at 20-25 C against hydrogen electrode -1,4 -0,8 -0,8 -0,5 -0,5 -0,5 -0,4 -0,3 -0,1 -0,1 -0,1 0,0 0,00 0,0 -0,1 +0,3 +0,1 +0,4 +0,4 +0,5
Anodic
Mg/Mg++ Al/Al
++
Magnesium Zinc Al alloys Cadmium Steel, cast iron Stainless steel, active Copper Tin Lead H/H+ Ni-Al-Bronze Stainless steel, passive Silver Platinum Graphite
Table 6-2:
6-59 6-59/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
6.7.2 Chink Corrosion A normal situation for electrochemical corrosion is where oxygen concentration cells create cathode areas on free steel surfaces while the corresponding anode areas with the dangerous corrosion is hidden in pores, cracks and connections. The attack is often called chink corrosion or crack corrosion.
Water
Figure 6-58: The principle in chink corrosion is as follows: Two plates overlap each other, and there is water in the overlaps (in the chink). The free water surface uptake oxygen and the metal surfaces in the marginal zones become cathodes. In the chink the oxygen have difficulties to penetrate and the metal surfaces become anodes and is corroded. This corrosion is not visible from the outside and dangerous. 6.7.3 Galvanic Corrosion If two different metals are in electrical contact in a moist environment galvanic corrosion is created. The more anodic metal releases ions and is corroded. An important factor in evaluation of the danger of this type corrosion is the ratio between anode and cathode area. If the anode area is small compared to the cathode area the current density over the anode will be large and the corrosion will be severe. The galvanic series is shown in Table 6-2. 6.7.4 Stress Corrosion Stress corrosion occurs in corrosive environment when the steel is exposed to tension stresses. Often the attacks are not visible and thereby very dangerous. The stresses may be introduced when the steel melt is solidified, or due to cold deformation of the steel, or due to outer static forces. The corrosion takes place where the stresses and thereby the energy level is highest. 6.7.5 Corrosion and Fatigue If the steel is exposed to alternating stresses there are a risk of fatigue fracture of the steel. This risk is increased when corrosion occurs together with the alternating stresses. 6-60 6-60/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
6.7.6 Atmospheric Corrosion Atmospheric corrosion is corrosion on un-protected steel surfaces exposed to the atmosphere. The air humidity is normally always higher than 65 % relative humidity that is the limit for adsorption of connected water film and thereby electrochemical corrosion. The water films is only a few molecule layers thick, however when there are small amounts of different salts on the surface, the water film becomes a powerful electrolyte. Some salts, for instance calcium hydroxide is very hygroscope and creates a water film already at 30 % relative humidity. Different factors influence the risk and velocity of the corrosion: Temperature The corrosion velocity is doubled for every 10 oC increase in temperature. Air pollution Both NaCl and other salts from the sea and the content of the SO2 play an important role. The sulphur dioxide originates the volcanic activity and from burning of fossils fuels like coal and oil. The sulphur dioxide creates H2SO3 which is oxidized to sulphur acid (H2SO4) which increases the corrosion velocity. In industrial environments the creation of soot is high. The soot contains sulphur and carbon and due to the hygroscopic properties the soot will be changed to sulphur acid. Between carbon and steel the sulphur acid is a strong electrolyte and a powerful corrosion cell is created. In marine environments the large amount of salts in the air may cause stronger corrosion than in the inner parts of the country. The salts originate from the sea where fog and moisture is airborn.
6-61 6-61/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Corrosion class 0
In-door environment
Out-door environment
Rooms with relative humidity lower than 60 %, dehumidified rooms Non heated well ventilated rooms without condense, steel buildings with natural ventilation Heated rooms with relative humidity more than 60 % without condense
Insignificant
Medium
Non polluted land atmosphere and similar environment with low sulphate acid-base and chloride pollution Polluted atmosphere, sulphate and other pollution from industry occurs Not polluted marine atmosphere
Large
Very large
6.8
Ageing of Steel
Impact on steel at very low temperatures may result in fracture without any large deformations as seen at normal temperatures. The brittle fracture form may also be seen on very old steels at normal temperatures. This phenomenon is called brittle fracture due to ageing of the steel. Steels ability to uptake dynamic loading is measured in Charpy s impact-notch sensitivity measuring test. The tendency to brittle fracture is evaluated by carrying out Charpy tests over a specified temperature interval. The ability is characterised by the transition temperature. It normally vary from 85 oC for the best steel types to +55 o C for the poorest steel types. However it should be mentioned that the transition
6-62 6-62/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
temperature is not similar to the lowest operation temperature of a structure constructed with the said steel. The operation temperature depends on the loading, thickness of the members and the notch sensitivity in the structural connections. The steel may be aged and hereby the transition temperature is changed. This is due to dissolved nitrogen in the steel immediately after the rolling which is precipitated after some time as very small nitride crystals that make the steel brittle. The change in the transition temperature is approximately 15 oC for the best steels and 80 oC for the poorest steels. Steel with a low transition temperature have normally lesser tendency to ageing. Impact ductility according to Charpy-V test Brittle Transition Ductile
Ageing
Transition temperature
Figure 6-59: Example on evaluation of a steels ability to brittle fracture. At normal temperatures the steel is ductile. Around 0 oC the impact ductility drops to very low values. This property is described by the transition temperature which may be defined as the temperature where the impact ductility is reduced by 27 Nm.
6.9
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Salt crystallization damage Salt crystallization damage is a analogous process to frost attack and is due to expansive crystallization of hydrated salts in the pore structure. Salt crystallization damage often occurs in warm conditions where the rapid drying of water is causing the salts to crystallize out below the surface. Abrasion Abrasion by particles in wind and water mostly acts together with other processes. The appearance will normally be of loss of surface and change of colour and texture.
6-64 6-64/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.
7.1
NDT-methods
General
The philosophy of a special inspection is to combine a superior visual assessment of a structure with appropriate test methods to obtain sufficient information on the condition of the structure. Location and selection of representative samples are important for giving accurate conclusions for the entire structural component considered. The extent of tests must be sufficient for determining the right repair strategy, and for giving a good estimate of the total area requiring repair. The personnel performing the special bridge inspection and using NDT-methods should be experienced and competent in three ways. They should be knowledgeable on the subjects stated below: How to carry out the available methods of testing in practice, including how to operate the equipment. How to select the right type of test method and the right test locations for different types of damage. How to interpret the results of the measurements.
The test methods can be divided into three categories: The non-destructive survey methods, which are suitable for mapping damage on large areas of the structure. The detailed non-destructive and destructive sampling and measurements on small areas. The laboratory analysis, which when applied on the samples, provides very detailed and precise information about a specific location. Normally, the combination of all three categories leads to very reliable conclusions on mechanisms of deterioration (deterioration types), causes and the extent of damage. However in most cases, the two first categories of test are satisfactory to conclude upon. Recording of registrations must refer to a unique numbering system for the structure. The numbering system must be indicated on a sketch (normally a plan view). One way to select a numbering system is to use the compass directions, e.g. column N1, E1 etc. The road destinations in the two directions should appear from the plan.
7-1 7-1/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.2
Visual inspection
Visual inspection involves using an inspector's eyes to look for defects. The inspector may also use special tools such as magnifying glasses, mirrors, endoscopes, or borescopes to gain access and more closely inspect the subject area. Visual examiners follow procedures that range from simple to very complex. Accessories for visual inspection
Mirrors Magnifying glasses
Borescope
Endoscope
7-2 7-2/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.3
Some cracks are measured continuously at specified intervals in order to monitor a possible crack growth. When the crack width is measured using a calliper it is necessary to establish permanent measuring points on both sides of the crack. At each measurement the distance between these two points is measured. Naturally, the line connecting the two points should be perpendicular to the crack. By each measurement the crack width (the distance between the two points) is measured together with the air temperature in order to remove temperature dependencies from the measurements. Normally, a series of measurements spanning several years must be made in order to determine whether the crack is expanding. If a large number of measurements have been performed the temperature-induced changes of the crack width may be filtered out. The results are reported in terms of a graph showing the crack width as a function of time.
7.4
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
A crack detection microscope is used to locate cracks and to measure the width of cracks. Some crack detection microscopes may also be used to measure the depth of the crack. The crack detection microscope is placed over the crack line and the focus is adjusted to get a sharp picture. Using the built-in scale the crack width is measured. Using a crack detection microscope the crack width may be determined with an accuracy of 0,01 mm. Based on the findings using the crack detection microscope and a visual inspection a map of the detected cracks may be produced.
7.5
Boroscope
This method can be used for both concrete, steel and masonry structures. The method is most commonly used on concrete and masonry structures. A boroscope is used to look inside inaccessible or small voids. For example, if cable ducts are not injected, it is possible to inspect the strands by means of an endoscope through a contact drilling (here a drilled hole from the surface to the cable duct). For steel structures the method is usually used for investigation of closed profiles to gain information regarding the condition of the interior surfaces of the closed profiles. For masonry structures the boroscope can be used to gain information of the depth of the outer layer of bricks or natural stones and it can provide information of the filling material in between the arches. It may also be used to examine the mortar between bricks or natural stone. The boroscope equipment includes a lighting source and a fibre optic cable to transfer the light to the boroscope. A system of lenses enables the boroscope to be used as a monocular. A camera or video camera can also be mounted on the boroscope for photo documentation. Generally speaking, the method is appropriate and may also be used for inspections of structural components such as expansion joints, honeycombs and cracks/slots. The many variations and features which can be obtained for borescopes make them an almost universal tool for internal inspections. These include a wide range of lengths and diameters, solid tubular or flexible bodies, lenses for forward, sideways or retro viewing, still and video camera attachments, and mains or battery power supplies. 7-4 7-4/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.6
Half-cell potential
This method can be used for concrete structures. The potential difference between a standard half-cell (normally a copper/copper sulphate reference electrode) placed on the surface of the concrete and the reinforcement underneath may be measured using the principle shown in the figure below.
Voltmeter
+
Reference electrode
Reinforcement
Concrete
The potential difference is associated with the rate of corrosion of the reinforcement. The purpose of potential measurements is to map the electrochemical potentials in order to locate areas with risk of corrosion see section 6.3.9. In the field the following steps have to be followed: 1) Exposure of a rebar for the electrical connection. Normally go for: Stirrups The most convenient areas:
The cover often varies, find the areas with the smallest cover on bridge decks. With an asphaltic overlay, the connection can be made to rebars in the edge beams.
2)
Check the circuit of the reinforcement: On columns, make a contact to the reinforcement of another column to check the circuit by using the multimeter. The resistance must be zero to have the circuit required. If that is the case, use the same connection during the required series of measurements. 7-5 7-5/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
If the resistance is not zero, first of all check the connection to the rebar. If this connection is good, the internal connection of the reinforcement is not sufficient and it is necessary to make contact to every column. Look for joints in the bridge deck and check columns on both sides of the joints. 3) On decks, use the above described procedure.
Make a measuring grid (columns and rows) on each part to be measured, noting the following: When making survey measurements on large areas, a mesh size of 500 x 500 mm may be chosen. Prior to making the grid, survey measurements at (more or less) random locations may help locating the areas to be mapped. When making measurements in areas where corrosion is likely to occur (selected as a result of survey measurements, experience or other test types), the mesh size should be 250 x 250 mm or less. The grid size, location and orientation must be marked on sketches of the structure.
4)
Check the stability of the potential measurements: wet a single measuring point place the electrode and note the potential and time wait until the potential is stable. NOTE the potential and time. This time difference is the necessary time required between wetting and measuring. In very dry concrete, it is normally necessary to wet continuously for a longer period. This means that one person is constantly wetting the structure in front of the person doing the measuring.
5)
Start the Measuring It should be noted that the potential measurement itself does not lead to a final assessment of the condition. Supplementary testing has to be carried out. As a first guide to an evaluation of the reliability of the measured potential values, the measurements are normally divided into groups. Immediately after completing the measurements, the results are printed out and then evaluated 7-6 7-6/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
according to a scale based on experience, e.g. (when using a copper/copper sulphate reference electrode): Group 1: potential > -200 mV: 90 % probability of no corrosion Group 2: -200 mV > potential > -350 mV: An increasing probability of corrosion Group 3: -350 mV > potential 90 % probability of corrosion The probabilities of corrosion given above are also given in ASTM C876. It is important to notice that the probability of corrosion also depends on many other factors such as: The oxygen concentration Moisture content Carbonation Chloride concentration Temperature Use of corrosion inhibitors Concrete resistance Coatings and sealers Cathodic protection systems
All the above mentioned factors must be taken into account when assessing the probability of corrosion. Hence, the results of half-cell measurements must always be calibrated on the basis of break-ups. Make break-ups to confirm the first evaluation and to evaluate the reduction of cross-sections. Note that the potential measurement is meant only for the detection of areas with corrosion activity. The reduction in cross sections cannot be assessed by half-cell measurements. Break-ups must be carried out for each group. As a rule of thumb, the breakups are placed in the most negative areas in each group. Breaks-up should therefore be performed in groups 1, 2 and 3. Start making a break-up in group 1 and group 2. If the rebars are without corrosion in group 2 then no break-up is necessary in group 1. When the connection between the potential values and the actual corrosion condition has been established through break-ups, the half-cell measurements can be used to assess the size of the damaged areas as a basis for rehabilitation design.
7-7 7-7/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
If the first evaluation does not confirm the results (if e.g. severe corrosion is found in group 2) limits must be changed for the three groups accordingly. These new limits must be confirmed by new break-ups.
Potential [mV]
4.0 3.0 2.0 1.0 0.3 0.0 12
100-150 50-100 0-50 -50-0 -100--50 -150--100 -200--150 -250--200 -300--250 -350--300 -400--350 -450--400 -500--450 -550--500 -600--550 -650--600 -700--650
Height [m]
Hjd [m]
10
11
Core, corrosion has been initiated Core, surface corrosion Core, no corrosion
The results of the half-cell measurements may be reported in terms of a surface graph as shown above. In the surface graph all relevant measurements of the chloride concentration, observations from break-ups and cores may also be given.
7.7
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The tests are conducted and reported in the same manner as the half-cell measurements. Unfortunately, the results of the corrosion rate measurements may vary with an order of magnitude depending on the instrument used for the measurements. Furthermore, the corrosion rate exhibits a much larger variation with factors such as humidity and temperature than the potential. Hence, the results of the corrosion rate measurements should be used with caution and should always be compared with half-cell measurements and calibrated on the basis of break-ups.
7.8
The method is generally suitable. Tests have shown that the inaccuracy increases from 5-10% at approx. 35 mm depth to approx. 15-25% at 60-70 mm depth. The findings may be reported in terms of the maximum and minimum values as well as the mean and standard deviation of the cover. The findings may also be presented as a surface graph.
7-9 7-9/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.9
The indicator normally used is 'phenolphthalein' (1% solution in alcohol) - After application, the colour of the alkaline concrete surfaces will immediately turn red-violet indicating a pH > 9.5, and the carbonated surfaces will remain colourless.
7.10
Impact-Echo equipment
This method can be used for concrete structures. Impact-echo equipment introduces a short-duration stress pulse into the considered member by a mechanical impact.
The impact introduces three types of waves: P-waves (compressional wave) S-wave (shear wave) R-wave (surface wave) The P-wave will be reflected when it reaches a surface or a material with another acoustic impedance. The successive arrivals of the P-waves to the surface are registered by a displacement transducer. On the basis of a spectral analysis of the reflected P-waves, the frequency of the reflected wave is determined. The thickness of the material or the depth to the defect, d, may be determined by:
7-10 7-10/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
d=
v 2f
where v denotes the wave speed and f the frequency of the reflected wave. The principle behind the Impact-Echo method is shown in the figure below. It is seen that the frequency of the measured response is higher when a void is present than when no void is present. This is due to the fact that the wave reflected from the void reaches the transducer faster than the wave reflected from the bottom of the test specimen.
Impact Impact
Transducer
Void
Amplitude
Amplitude
Frequency
Frequency
The wave speed may be determined by testing a specimen with known thickness containing no defects. Alternatively the wave speed may be measured on the surface using two transducers. The equipment may also be used to measure the depth of a crack. Using the setup indicated below, the crack depth may be determined on the basis of a measurement of the time it takes the P-wave to reach the two transducers.
Transducer
In general the equipment may be used to determine: Thickness of members Presence and depth of cracks, voids and honeycombing Missing bond between concrete and asphalt overlay/waterproofing Injection quality of cable ducts 7-11 7-11/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The measurements are performed by producing an impact on the concrete surface e.g. with steel balls (impactors) with diameters ranging between 2 to 15 mm. The size of the steel ball should be selected on the basis of the thickness of the considered test specimen. For each measurement the spectrum of the measured response is shown on a laptop PC connected to the transducer. On the basis of the response spectrum the operator estimates whether defects are present. The results obtainable from using the Impact-Echo equipment to a great extent depend on the experience of the operator. The method is fast and an experienced operator is able to test relatively large areas (e.g. a bridge deck with the dimensions 12 x 50 m) for defects within a working day. The measurements should always be calibrated on the basis of independent tests. Usually the inspection of concrete structures is calibrated on the basis of cores, break-ups or a visual inspection using a boroscope. The results of the measurements are used to report the general condition of the considered component as well as a detailed mapping of the detected defects.
7.11
7-12 7-12/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Mobility Hammer
Impact
Transducer
Frequency
For each measurement the resulting mobility graph is shown. On the basis of the mobility graph the following parameters are determined: Average mobility: The average mobility is shown as the green line in the figure above. The average mobility depends on the thickness of the material. If the thickness is reduced the average mobility is increased. This implies that laminated concrete has a higher average mobility than non-laminated concrete. Stiffness: The stiffness is determined as the inverse of the inclination of the part of the mobility graph below 80 Mz, the red line in the figure above. The stiffness depends on the stiffness of the material, the thickness of the material and it depends on how the component is supported. Based on a comparison of the stiffness at a number of different locations potential weak areas may be located. Mobility slope: The presence of honeycombs in the concrete will reduce the damping of the signal. This implies that the mobility graph will be increasing within the considered frequency range, see figure below.
Voids index: The voids index is defined as the ratio between the initial maximum of the mobility and the average mobility. If the component is laminated the initial maximum of the mobility will be con7-13 7-13/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
siderably higher than the average mobility. If the voids index is higher than 2 4 it indicates a potentially weak area, see figure below.
The impulse response method is a fast method which may be used to screen a relatively large area within a short period of time. The equipment delivers surface graphs of the measured parameters. In the figure below a surface graph of the average mobility of a bridge deck is shown.
The results of the impulse response testing shall always be calibrated on the basis of e.g. cores, break-ups or a visual inspection using a boroscope. The locations of these tests are selected on the basis of the surface graphs of the measured parameters.
7.12
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
To ensure the correct geometry of the fracture, a special bolt and a circular counterpressure device is used. The bolt is extended in a recess milled at the bottom of a 25 mm hole in the concrete and pulled out by means of a small hydraulic jack. Reading the hydraulic pressure gives the pullout force, from which the compression strength of the concrete can be found by means of a calibration chart. Special attention must be paid to: Selection of representative areas of the damage in question. Milling of the recess with sharp edges (by keeping the miller at a right angle to the surface all the time). Greasing the insert before inserting. Assembling and tightening the various parts of the expansion bolt and jack in the right sequence. Fastening the insert without rotating it. If the fracture surface is not conical, the measurement is not valid and a new test must be made. The test measures the strength in a very small area. The presence of coarse aggregates or minor deficiencies in the concrete at the test location may affect the measured strength. To compensate for this, at least three tests are made. It is relatively time-consuming to perform CAPO-tests. It is necessary to repair the concrete surface after the tests have been made. However, it is less expensive to perform CAPO-tests than cutting cores and determine the strength in the laboratory. 7-15 7-15/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
If the purpose of the tests is to determine the characteristic value of the compressive strength of the concrete at least five tests must be made. However, the rules for determining the characteristic values from CAPO-test is determined my national codes and guidelines. Alternatively, some prior information or some information from other test methods must be used in order to determine the characteristic value of the compressive strength of the concrete.
7.13
Pull-off/Bond-Test
This method can be used for concrete structures. Normally, the method is primarily used for testing the bond strength of a newly cast concrete layer to the existing concrete. Further it can be used for testing the tensile strength of the existing concrete surface when certain strength is required before the new concrete layer and/or waterproofing can be placed. Finally, it can be used for testing the bond of a membrane to a concrete surface. In the bondtest a disc is glued on a prepared surface. The disc is pulled off after a partial core has been cut around the disc. On the basis of the pull-off force the tensile strength of the material may be determined, see the figure below.
When testing the strength of a newly cast concrete layer to the existing concrete one of the following failure types may be observed: Failure in the substrate Failure in the adhesion layer Failure in the overlay Failure in the substrate is the preferred, as it proves the adhesion strength of the overlay to be higher than the tensile strength of the substrate. The results are usually reported in terms of acceptance or non-acceptance of the strength of the considered material. The decision is made on the basis of a decision
7-16 7-16/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
rule formulated prior to the tests. If a sufficient number of tests are performed the results may be used to estimate the characteristic value of the tensile strength of the concrete.
7.14
Schmidt hammer
This method can be used for concrete and masonry structures. It is most commonly used for concrete structures. The Schmidt hammer is used for testing the strength of hardened concrete. The device consists of a spring loaded steel mass that is automatically released against a plunger when the hammer is pressed against a concrete surface. Part of the energy is absorbed by the concrete through plastic deformation and part of the energy causes a rebound of the hammer. The rebound of the hammer depends on the hardness and thereby the strength of the concrete. In order to estimate the strength of concrete at least 20 measurements should be made. The measurements shall be performed at locations where the concrete surface is smooth. The distance between the individual measurements should be at least 0,5 1,0 m. All measurement shall naturally be performed within a homogenous area. The actual measurements are made by pressing the Schmidt hammer with the plunger extended slowly against the concrete surface until the hammer is released. At the moment of impact the hammer must be held perpendicular to the surface. The Schmidt hammer should not be used to measure the strength of weak concrete, fractured concrete and concrete with an uneven surface. The results obtained using a Schmidt hammer are not as accurate as CAPO-test or strength testing of concrete cores drilled from the structure. The method is best suited for scanning large areas in order to divide the structure into homogenous areas, i.e. areas with different values (levels) of the concrete strength. The compressive strength of the concrete in the poorest areas may then be estimated on the basis of a more accurate method such as e.g. CAPO-tests.
7.15
7-17 7-17/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The investigation depth depends on the selected frequency of the ground penetration radar. For investigations of bridge decks a frequency of 1,5 MHz may be selected. The investigation depth is then 500 mm. Ground penetration radar may be used to locate reinforcement in concrete structures. In particular, the ground penetration radar may be used to locate pre-stressed reinforcement, which due to the large cover cannot be located using an ordinary cover meter. The ground penetration radar may also be used to identify areas with high humidity as well as voids. The distance from the surface to the given defect may be measured with an accuracy of 10 15 %. Investigations of bridge decks may be performed using a truck-mounted ground penetration radar. The investigation of a bridge deck may be conducted at a speed of 5 km/h. Hence, the method is highly efficient. The results of the test must be interpreted by a trained professional. Further, the results must be calibrated on the basis of independent tests such as cores, break-ups and a visual inspection.
The results of the inspection may e.g. be reported in terms of 3D-graphs showing the location of areas containing a given defect (e.g. high humidity) or the location of e.g. pre-stresed reinforcement as shown above.
7-18 7-18/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.16
Chloride content
This method can be used for concrete structures. The chloride content in concrete may be determined on the basis of: Cores Dust samples Cores are obtained as described in the section Coring equipment. The diameter of the cores should be at least 75 mm. Dust samples are obtained using a power drill. The power drill should preferably be mounted with a unit for automatic collection of the concrete dust. The dust sample should weigh at least 15 g. The dust samples are usually obtained at different depths at the same location usually in steps varying between 10 20 mm (in depth). The number of holes necessary to obtain 15 g dust is shown below as a function of the diameter or the drill. The holes should be located within a circle with a diameter of 75 mm.
It is recommended to measure the chloride content at the following depths from the surface: 0 10 mm 10 20 mm 20 30 mm 30 50 mm 50 70 mm The chloride concentration may be determined by the Rapid Chloride Test (RCT) or by Volhard titration. The Rapid Chloride Test, RCT, is a fast method of determining the acid soluble amount of chlorides of concrete in-situ. Pulverised concrete obtained by hammer drilling of hardened concrete or from a concrete core is mixed with a chloride extraction liquid and shaken for 5 minutes. The amount of acid soluble chlorides - expressed as weight percent of concrete weight -
7-19 7-19/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
is determined directly by means of a calibrated chloride sensitive electrode connected to the RCT-electrometer. Volhard titration must be conducted in accordance with a given code. Both methods measures bonded as well as free chlorides. Using the dust samples from different depths - the chloride profile is determined by testing each depth interval. Examining the profile, the probable source of the chlorides and mechanism of penetration can be detected (curing water, saline soil, seawater, freed chlorides from aggregates, air-borne chlorides etc.). On the basis of the chloride profile and a mathematical model of chloride ingress (e.g. Ficks second law see section 6.3.7) the time to initiation of corrosion may be determined. The critical chloride concentration for initiation of corrosion must be known. This value may e.g. be estimated on the basis of chloride measurements performed at break-ups where corrosion has been initiated. The results may e.g. be reported in terms of charts showing the chloride concentration as a function of the distance from the surface. The results may also be shown on the surface graphs of the registered half-cell potentials.
7.17
Coring equipment
This method can be used for concrete and masonry structures. A qualitative assessment of the concrete quality may be obtained by scrutiny of drilled-out cores. The right place to take the cores depends on the structure geometry, the condition of the concrete or masonry, and what information is required in order to determine the type and extent of damage. Prior to the drilling out of cores the condition of the concrete or masonry has usually been investigated on the basis of a visual inspection or some NDT-measurement such as e.g. half-cell potential, Impact-Echo or impulse response (for concrete structures). In areas where the previous investigations with a high degree of accuracy have shown that the structure is either damaged or undamaged only few cores should be drilled out. The majority of the cores should be drilled out at locations where the results of the previous measurements are inconclusive. The cores will then provide a basis for an interpretation of the results of the NDT-measurements in these areas, thereby assuring that the degree of deterioration of the structure is estimated with the highest possible degree of accuracy. 7-20 7-20/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The number of cores depends on the size of the considered area. Normally, about 2 4 concrete cores are drilled out within each area investigated by a given NDTmethod. Avoid cutting reinforcement bars. To ensure this, locate the reinforcement by means of the cover meter before drilling. Cores are usually drilled out using a portable electric concrete core drill as shown below.
Usually, the diameter of the concrete cores is 100 mm. Once the core has been drilled out a photograph of the cores is taken and the location where the core was taken is registered. The location should be registered in terms of the grid used for the NDT-measurements (HCP-measurements, ImpactEcho or Impulse Response), see the figure below. Also registration of the hole left in the structure (where the core has been taken out) is to be made. Especially signs of cracking are to be registered.
7-21 7-21/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Once the surface of the core has dried out the core is wrapped in saran wrap and put into an air-tight plastic bag. The concrete cores provide very accurate information about the quality of the structure from which the cores were taken. However, it is time-consuming to drill out cores. Furthermore, a core leaves a defect in the structure from which it was taken.
7.18
7-22 7-22/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.18.1 Macro analysis on cores and plane sections Close macroscopic inspection of concrete cores (possibly using a magnifying glass or a stereo microscope) can give information about the concrete mix (the aggregate type, aggregate content, encapsulated air voids), and it may uncover internal deficiencies such as cracks and inhomogeneities. Casting defects and the condition of joints can be determined by the macroscopic evaluation as well. In particular the core can tell how deep cracks reach into the concrete, thus giving an indication of the cause of the cracks. If reinforcement is included in the cores the condition of the reinforcement can also be evaluated. An example of a concrete core is shown in Figure 7-1.
Figure 7-1: Concrete core for macroscopic evaluation. The results from the macroscopic evaluations are usually registered by filling in a standard form. The registrations are always supplemented by one or more photos of the core. By making a fresh cut in the concrete core the carbonation depth can be determined by using phenolphthalein see also section 7.9.
7-23 7-23/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-2: Measurement of carbonation depth by use of phenolphthalein. The red part of the core is not carbonated. 7.18.2 Crack detection on impregnated plane sections The crack pattern can give valuable information regarding the cause of damage. To evaluate the crack pattern impregnation of a plane section of the concrete core is a great tool. The impregnation of the plane section is performed in two steps as illustrated in Figure 7-3. a) b)
Figure 7-3: Illustration of the two steps in impregnation of a plane section for detection of cracks. a) Vacuum-impregnation of full core with fluorescent epoxy resin. Cracks, voids and porous paste connected to the core surface will be filled with fluorescent epoxy resin. b) Impregnation of plane section with fluorescent epoxy resin. Cracks, voids and porous paste near the cut surface will be filled with fluorescent epoxy resin. By use of ultra-violet light on the impregnated plane section all cracks, voids and porous paste near the cut surface will be shown clearly. An example of a fluorescent impregnated plane section is shown in Figure 7-4. The crack pattern is very clear and the extent and distribution of cracks can be determined. 7-24 7-24/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-4: Fluorescent impregnated plane section from a bridge deck under ultra violet light. The cause of damage could be AAR or freeze-thaw. 7.18.3 Micro analysis on thin sections More information may be obtained from the cores by performing 'thin section petrography' which is a technique using a microscope in combination with various optical filters and epoxy resin impregnation to investigate very thin slices of the concrete. This technique requires sophisticated laboratory equipment and extensive experience. A thin section is a 20-micron thick slice of concrete, which has been impregnated with a fluorescent epoxy resin. The thin section is typically 35 mm x 45 mm in size. The semi-transparency of the concrete slice allows the examination of the concrete by transmitted light microscopy. The impregnation with the fluorescent epoxy resin makes it possible to determine the water-cement ratio and the homogeneity of the cement paste. Further more the air voids, cracks (including micro cracks) and porous materials are clearly shown in the thin section.
7-25 7-25/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-5: To the left a thin section is shown. To the right a thin section is examined in a microscope. When performing a thin section analyse it is possible to determine the following parameters: concrete composition cement type and content aggregate type and mineralogy w/c-ratio air void content and void structure defects (cracks and inhomogeneities) aggressive environment (e.g. acid) moisture conditions and effects signs of deterioration (e.g. AAR) strength level initial defects (casting, curing etc.)
7-26 7-26/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Air void
Sand
Figure 7-6: Part of thin section. To the left the thin section is shown in ordinary light with parallel polarizers and to the right the thin section is shown in fluorescent light. To the right the homogeneity of the cement paste is shown by the colour intensity the darker colour the more dense cement paste (low w/c). Signs of deterioration can also be identified in a thin section. In Figure 7-7 an illustration of alkali silica reaction is shown.
Figure 7-7: Sand aggregate of reactive porous flint with interior and exterior cracking. The results of the thin section analyse is very precise when performed by an experienced engineer or geologist. It is however very important to keep in mind that the results from the microscopic analyse is only valid for the part of the structure represented by the thin section. Thus, selection of the position for the thin section is very important remembering that the thin section is only 35 mm x 45 mm large. Typically the thin section is placed so it includes one or more cracks if any. Also intact areas of the concrete should be included in the thin section. It might be necessary to make two thin sections of one core to represent the concrete of the entire core. 7-27 7-27/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.18.4 Air void analyse on plane section By preparing a plane section of the concrete core, the air content and air void distribution can be determined.
Figure 7-8: Plane section prepared for determination of the air content and distribution. All air voids are white and all paste and aggregate are grey or black. 7.18.5 Moisture analysis By slicing the concrete core into several slices a moisture profile through the core can be determined. If the moisture profile is to be determined it is very important that the core is sealed in an air tight bag right after drilling out the core. The core must then be stored cold (e.g. in a refrigerator) untill the measurements starts.
Figure 7-9: Slicing the concrete core makes it possible to determine the moisture profile. The mass of each of the concrete slices is measured (m0) and the concrete slices are then stored in water. The initial mass is used to determine the actual water content of the concrete. The mass is measured regularly and the measurements continue till the mass is constant (mcap) this constant mass is used to determine the degree of capillary saturation. By placing the concrete slices in a pressure camber more water
7-28 7-28/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
can be pressed in to the concrete and by measuring the mass again (mpressure), the degree of pressure saturation can be determined. Finally the concrete slices are stored at 105 oC until the mass is constant (mdry) this provides the dry mass of the concrete. The formulas for determining the actual water content (U), the degree of capillary saturation (Scap) and the degree of pressure saturation (Spresure) are given by:
U=
m0 mdry mdry
S cap =
S pressure =
2.0
4.0
6.0
8.0
40-60 60-80 80-100 100-120 120-140 140-160 160-180 180-200 200-220 220-230
Kerne 2
Figure 7-10:
7.18.6 Residual Reactivity Test If the registrations from the macroscopic and microscopic evaluations indicate alkaliaggregate reactions (AAR) as the cause of damage a residual reactivity test can be preformed on one or more concrete cores. The purpose of the test is to evaluated the potential risk of development of AAR damage and to estimate the residual potential for further reactions under the following conditions: unlimited access for moisture (the test specimen is wrapped in a wet towel and kept in a plastic container by the humidity of 100%) unlimited access for moisture and sodium chloride (the test specimen is kept in a container filled with concentrated NaCl-solvent)
7-29 7-29/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The test is performed by storing two specimens cut out of the concrete core in the conditions mentioned above. The size of the test specimens could be app. 4x4x15 cm. To accelerate the chemical reactions the specimens are stored at 50 oC. By regular measurements of the expansion of the test specimens the development of AAR can be evaluated in the case of unlimited access for moisture respectively unlimited access for moisture and sodium chloride (alkalis). The time of storing and there by the time of performing the test depends on the type of reactive aggregate. If the expansion exceed 1 0/00 harmful cracking of the structure could occur in the future. In Figure 7-11 an example of test results from residual reactivity tests is shown.
Expansion 0/00
Specimens stored in sodium chloride Specimens stored at 100% RH Storing time in weeks at 50oC in sodium chloride solution and at 100% relative humidity
Figure 7-11: Example of test results for residual reactivity tests. In the example shown in Figure 7-11 there is a small risk of future harmful cracking due to AAR if alkalis are provided from the surroundings.
7-30 7-30/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.19
7-31 7-31/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The system provides a long-term continuous monitoring of flaws and early warning of e.g. crack growth. Further, the method may be used to locate flaws at locations where it is difficult or impossible to conduct a measurement. Acoustic emission monitoring is costly. The method is usually only used if prior inspections and analyses have shown that there is a high risk of severe deterioration of the bridge. In such cases the acoustic emission monitoring may be applied in order to monitor the progress of the deterioration. This provides a basis for decisionmaking concerning the repair of the structure.
7-32 7-32/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.20
Design of system The design of the system consists of selecting the proper sensors depending on the required accuracy, sampling frequency etc. Further, the designer must select the system for data acquisition, cabling, communication, user interface and operation. 7-33 7-33/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
4.
Installation, commissioning and hand-over Installation of the system should be performed by a hired professional with experience in the installation of structural monitoring systems.
The monitoring system provides a tool for gathering continuous information on critical items. This information may e.g. be used in order to determine the optimal time for initiation of a given repair strategy. The system may be designed such that an alarm is triggered if a given event occurs. The alarm may e.g. be triggered by a rise in humidity or potential. The major advantage of a monitoring system is that it may provide a large quantity of information and thereby allows for a very accurate assessment of the condition of the structure and a very accurate planning of repair and maintenance of the considered structure. Unfortunately, the use of structural monitoring is often expensive.
7.21
The system may be used for: Mapping of reinforcement Studying inhomogeneities in concrete Inspection of pre-stressed cables and cable ducts In order to use the method both sides of the considered component must be accessible. Further, the method is slow. The major advantage of the method is the high degree of accuracy. It is e.g. possible to detect a 20 mm porosity in 1000 mm thick concrete. 7-34 7-34/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The method provides a direct and precise illustration of the inner structure of the considered component. Further, the method is one of the best NDT-methods with respect to the amount of information, quality and reliability. On the other hand, the equipment is very heavy, it is very expensive and the use of the equipment requires special safety arrangements due to the radiation dangers.
7-35 7-35/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.22
Ultrasonic testing
This method can be used for steel structures. 7.22.1 Definition of ultrasound Mechanical vibrations of different kinds can travel through solids due to their elastic properties. A good example is a spring, which is tightened at one end. The other end is able to expand up and down. If it makes enough oscillations per second, you will be able to hear a sound. This is due to the fact that the air also starts vibrating as compression waves. The human ear can hear these compression waves, if the frequency is higher than the lowest audible range, which is about 12 oscillations/sec. The faster the spring oscillates the higher the sound. Over a certain number of oscillations, we are not able to hear anything. We have then reached the upper audible level, which is about 20.000 oscillations per second. Sound waves with a higher frequency are called ultrasound waves. After changing to the use of ultrasound the method became useful in a greater scale. Ultrasonic waves gives due to their higher frequency and smaller wave length a much better possibility of finding defects and determine their size and their position. The vibrations are normally generated by the use of a piezoelectric crystal, which can be excited by an electrical pulse. See section 7.22.6 for more information about transducers. We are going through the two most common test methods, the through transmission technique and the pulse echo technique. 7.22.2 Through transmission technique When using this technique you have a transmitter on one side and a receiver on the other side of the object to be tested. The transmitter sends out ultrasonic waves either as continuous oscillations or as short pulses, each consisting of a few oscillations. In the last case the pulses are send out with an interval, which is long compared to the duration of the pulse itself. The wave travels through the object and is then received by the receiver. The signal from the receiver shows the sound energy, which has travelled from the transmitter to the receiver. If the sound beam hits a discontinuity in the object, the received sound energy will be less. The signal from the receiver will then be smaller. This signal can be registered and used in different ways. For example the signal can automatically activate an alarm, if the sound beam hits a defect over a certain size, by which the received sound energy goes down under an equivalent fixed level. The signal can also be registered with ordinary ultrasonic equipment, which contains an oscilloscope. It is seen as vertical reflections of the signal on the screen at a distance to the right of the deflection on the left side of the screen. This reflection to the left is called the initial pulse. 7-36 7-36/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The ultrasonic equipment's way of working will be discussed later. If there is a defect between the transmitter and the receiver, it will prevent a larger or smaller part of the sound beam from reaching the receiver, which will weaken the signal. This is seen on the screen as a smaller deflection as seen on Figure 7-12.
A B
Figure 7-12: Principles of through transmission technique. 7.22.3 The pulse echo technique
This is the most common used technique. The principle in this technique is almost the same as used in an echo sounder. A transmitter sends out a short pulse consisting of a few oscillations into the object to be tested. The sound wave travels through the object with a constant speed, the sound velocity, which is always the same in the same material, regardless of the frequency. If the object is without defects, the pulse continues until it hits the back wall of the object, from where it is reflected like light beam from a mirror. The pulse then travels back through the object - still with the same velocity - and is received by a re-
7-37 7-37/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
ceiver. As the pulse travels with a constant speed, the time the sound pulse has travelled from the transmitter till it returns is equivalent to twice the thickness of the object. After a while a new pulse is send out, which travels exactly like the first one. In order to measure the very short time from sending out one pulse till it is received again, the ultrasonic equipment is provided with an oscilloscope or a digital display. An electron beam makes a bright spot to travel horizontally across the screen with a constant velocity from left to right. The movement begins at the same time as the pulse is send out from the crystal. The initial pulse gives a vertical deflection on the left side of the screen. After that the bright spot continues to the right with a speed that can vary from about
1 200
A B F
A F F
Figure 7-13:
If you adjust the velocity of the bright spot on the screen, to be the same as the sound velocity in the test piece, it will travel to the right on the screen with the same speed as the pulse travels inside the object. When it returns to the receiver, the bright spot has travelled a distance, which is twice the thickness of the test piece. The moment the pulse hits the receiver, it sends an electrical signal to the ultrasonic equipment. On the screen it is seen as a brief, vertical reflection of the bright spot. This is called a bottom echo. The distance on the screen between the initial pulse and the bottom echo is in this case twice the thickness of object.
7-38 7-38/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Changing the speed of the bright spot on the screen, the distance between the initial pulse and the bottom echo can be adjusted. You can change it in such a way that the thickness of the steel object between approximately 2 mm and 10 m can be read off on the screen. If the sound wave hits a reflecting surface during its way through the object for example a crack, a part of the sound will reflect back and will be seen as a vertical reflection before the bottom echo. This deflection is called the defect echo. By its position on the screen you can determine the distance from the surface of object quite accurate, see Figure 7-13. The height and shape of the flaw echoes might give you some information about the size and type of the defect. The sending out of a pulse and the movement on the screen is repeated many times a second. The single instant pictures appear on the screen as constantly shining lines, which only moves when the probe is moved across the surface of the object. However, there is a distance between the pulses, which allows the first to die out before a new pulse is send out. In most ultrasonic equipment it means that a pulse can move backwards and forwards in a 10 m long steel bar, before a new pulse is send out. Figure 7-14 and Figure 7-15 show some common ultrasonic equipment.
Figure 7-14:
7-39 7-39/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-15:
USK 7 D.
7.22.4 Definitions and general terms Sound beams are mechanical vibrations of every single particle within an object. If you imagine the object split up into many small particles, which are mutual connected with elastic power, you will generate a wave motion by getting one or several particles out of balance for example by giving them a shock. Because of the elastic power the neighbour particles will after a while get out of balance too. The shock will spread in the object as a wave motion. It is characteristic for a wave motion that a transport of energy is taking place, but it does not result in a transport of substance. Each particle oscillates with larger or smaller oscillation around their equilibrium, but keeps their position in relation to the other particles. It can be shown that the oscillations are sinus oscillations. Wave motions can appear in many different shapes. In the following we will mention some important definitions and general terms concerning ultrasonic oscillations and the most important types of oscillations. Frequency (number of oscillations) The frequency is the number of oscillations per second. One oscillation is a movement from a mean position to a maximum through a mean position to a maximum and back to a mean position again. The term for a cycle is Hertz or c/s. And the time used for one oscillation is called a period. 7-40 7-40/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
In ultrasonic testing frequencies between 1-6 MHz (megahertz, mega = million) is often used. For example is the time for one single oscillation at 1 MHz equivalent to
1 1.000.000
sec.
The sound direction is the direction of the wave propagation. It does not have to merge with the direction of the particle movement. The particles can move in the direction of propagation or at right angles to the direction of propagation. Wavelength The wavelength is the distance measured in the sound direction from one particle to the next particle in the same mode. The wavelength is inversely proportional to the frequency that is to high frequencies you will have small wavelengths and conversely. The sound velocity If you call the wavelength and frequency f this equation applies for a wave motion f x = V = constant The constant is the velocity of sound in the material and not of the single particle itself. It tells you how many wavelengths, that is how long a distance, the wave propagates per second. The velocity is a quality of the object and for a certain object, the velocity of sound is constant for all frequencies and wavelengths.
Types of oscillations Sound can propagate under various forms: Longitudinal waves Longitudinal waves are characterised by the fact that the particle motion is in the direction of propagation of the sound Longitudinal waves can propagate in solids as well as in gases and liquids. The audible sound is for example longitudinal waves in air. Transverse waves Transverse waves are characterised by the fact that the particle motion is at right angles to the direction of propagation. The elastic forces, which make the particles oscillate, are displacement forces. These forces are not found in liquids and gases, so transverse waves can only be transmitted in solids. 7-41 7-41/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Pure longitudinal- or transverse waves can only be generated inside an object if the extent is great compared to the wavelength. 7.22.5 Refraction and reflection of ultrasonic waves Refraction If a sound beam hits an interface between two different materials in an oblique direction, a part of the wave will be refracted in the surface and continue into the other material in a diverging direction. Another part of the incident wave will be reflected from the interface. This refraction and reflection happens due to the laws of optics for the lights refraction and reflection. See Figure 7-16.
V1
Material 1 Material 2
V 2
Figure 7-16:
Reflection at right angles If a sound beam hits an interface at right angles a part of the sound will travel through the interface and continue in the same direction, but with another velocity. No refraction happens in this case of course. The other part of the sound will be reflected back from the interface at a right angle. At the interface between steel and air the reflections coefficient is almost -1 that means that the incident sound wave is completely reflected and that there is no sound wave transmitted into the air. By ultrasonic testing of materials for internal defects the reflected signal will always come from an interface of this type. Most discontinuities in a material will have an interface towards the part without a defect, which has the character of an air gap. Such an interface is a great reflector at air cracks with a thickness down to mm. 7-42 7-42/311
1 10.000
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Reflection at an oblique angle of incidence If a sound wave hits an interface with an angle of incidence i, a part of the wave will be reflected with the angle of reflection r, as already illustrated in Figure 7-16. As for a light beam reflecting in a mirror the same rule applies that the angle of incidence and the angle of reflection are equal, that is i = r. Besides this similarity between the laws of light and those of sound waves, there is an important difference due to the fact that sound beams in solids can be either transverse or longitudinal and change from one form to the other under certain conditions.
Longitudinal wave
Transversal wave
Longitudinal wave
Figure 7-17:
When a longitudinal wave in a solid material hits a plane interface with an angle of incidence i, a part of the beam will be reflected as a longitudinal wave with an angle of reflection equal to i. Furthermore a part of the beam will change into a transverse wave, which is reflected with a smaller angle of reflection (Ut) due to the smaller velocity for transverse waves (Figure 7-17). Steel is one of the most common used materials in construction and therefore the material most often tested with ultrasound. Below the condition at an interface between steel and air is further discussed. As mentioned earlier such an interface will reflect a sound beam completely which means that it is totally reflected. Longitudinal waves in steel propagate with the velocity VL 5900 m/sec. and transverse waves with the velocity VT 3230 m/sec. Decibel On most equipment you have a control, with which you can regulate the gain. This control is divided and adjusted in decibel (dB). In this way it is possible to measure
7-43 7-43/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
and compare the height of different echoes. Such measurements are necessary when comparing defects, testing for absorption and estimating the size of defects. 7.22.6 Probes Normal probes A normal probe generates longitudinal waves, which leaves the probe at a right angle to its contact surface. If the probe is in contact with a specimen, the sound wave penetrate into it. It travels in straight lines, with a certain beam spread. See Figure 7-18.
Figure 7-18:
Normal probes.
Figure 7-19:
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The crystal must be damped in order to quickly stop the oscillations after it has been excited, either by an electrical pulse or by a reflected sound wave. In this way the initial pulse and the echoes on the screen of the equipment are prevented from being too wide Dual probe (TR-probes) The near resolution can be increased considerably by using a probe with two separate crystals one for transmission and one for receiving. Figure 7-20 shows the inside of a TR- probe.
Transmitter
Perspex
Figure 7-20:
TR-probes.
The piezoelectric crystal is glued to perspex blocks, which works as a delay line for
the sound. The crystals are placed in a slight angle to the surface of the object, and turned against one another. Due to that you can detect defects right under the surface. Unfortunately this construction may give spurious echoes from surface waves. This can be avoided using a probe where the crystals are parallel to the surface of the object. In return you have a minor sensitivity for defects right under the surface. Beyond these common types you have special probes developed for specific tasks e.g. waterproof types and heat resistant types. Choice of probe Normal probes differ from one another as regards to the type of crystal, its diameter and frequency. When choosing a probe you have to evaluate the influence of these variables to the quality of the probe. Below is summarised what to consider when choosing a probe. Choosing a higher frequency gives you: 1. The possibility of detecting smaller defects. 2. Less sensitivity, shorter penetration into the material.
7-45 7-45/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
3. A longer near field. 4. Smaller angle of divergence. 5. A better resolution. Choosing a crystal with a bigger diameter gives you: 1. A better sensitivity, longer penetration into the material. 2. A longer near field. 3. A smaller angle of divergence. Choosing a crystal of another material you can obtain another combination of sensitivity and resolution.
Angle probes Angle probes are normally manufactured with frequencies between 2 and 5 MHz and the angles 35, 45, 60, 70 and 80 for testing in steel. Other frequencies and angles are available. The angles are always stated in proportion to the normal. See Figure 7-21. The probe index is marked on the side of the probe with a line.
Figure 7-21:
Angle probes.
7-46 7-46/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-22:
Angle probe.
Between the piezoelectric crystal and the exit point a wedge-shaped middle piece is placed. When the probe is brought into contact with an object, the longitudinal waves, generated from the crystal, will travel at an oblique angle to the interface between the wedge and the object. Here they will be refracted and continue into the object with a different angle. If the wedge angle is small, a part of the longitudinal waves in the wedge will continue into the object as longitudinal waves and a part as transverse waves. These two wave types will penetrate into the object in different directions, which means that it will be difficult to decide, where possible echoes comes from. Due to that the wedge is fabricated with an angle, which is larger than the 1. critical angle. The longitudinal waves are totally reflected inside the wedge. The final result is a refracted transverse wave in the specimen. The shape of the wedge results in quite a lot of spurious echoes on the screen, because a part of the sound beam being reflected at the interface and inside the wedge will be reflected back and forth and finally hit the crystal. As shown on Figure 7-22 you can provide the wedge with a crystal backing for absorption of the reflected sound beams inside the wedge. Another solution is to provide the wedge with different saw cuts in order to make the reflected sound beams not hit the crystal. If you place an angle probe on a plate the sound wave will travel between the two surfaces as shown in the Figure 7-23.
7-47 7-47/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-23:
The distance between the place, where the ultrasonic wave enters the plate and the point where it hits the top again is called the skip distance P. If you place a probe on other materials than steel with another sound velocity, the angle will change. Therefore the above mentioned factors can only be used when testing steel.
Table 7-1 shows how the angle () changes from testing steel to aluminium, copper and cast iron. Steel 35 45 60 70 80 Table 7-1: Aluminium 33 42,4 55.5 63.4 69.6 Angles in different materials. Copper 23.6 29.7 37.3 41 43.4 Cast iron 23 28 35 39 41
Field of application Angle probes are normally used for testing welds and for testing of parent material in pipes. Welds are tested for internal weld defects e.g. slag inclusions, porosity, cracks, lack of sidewall fusion and lack of penetration. Pipes are tested for material defects and for transverse and longitudinal cracks Common angle probes can be used on hot surfaces with a temperature up to 70-80 C. You should also be aware of the fact that the refracted angle may change, due to the sound velocity in the probe and in the object changes with temperature. The sensitivity of the probe will decrease with temperature, because the attenuation in the probe will increase.
7-48 7-48/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Check of probes A probe is characterised by a series of qualities, which is of importance for its function. Before you start using a complete new probe or when you have used a probe a while, it might be of interest to check that it fulfils the specifications, which are listed by the manufacturer. It can be necessary to check the following qualities: Frequency - sensitivity - resolution - the width of the initial pulse - the width of the flaw echo and the shape of the sound field. For angle probes you must check the probe index and the direction of sound. Most probes have a wear plate called the sole between the crystal and the object in order to avoid wear of the crystal. When checking the probe you should inspect and maybe have the sole replaced. Coupling media At the interface between steel and air or between air and another solid material or liquid, you will have an almost 100% reflection of the sound. In practice it is not possible to produce a direct contact without air gap between the probe and the object under test, so you will have to place a coupling media between the probe and the object. As coupling media are used oil, grease, water, glycerine or wallpaper paste. Oil or grease is normally used when testing machine elements. By continues testing of welded seams, you will of financially reasons often use water as coupling. Angle probes can be provided with special water coupling so the water comes out in the middle of the contact surface. Wallpaper paste is rather thick and therefore ideal when testing on sloping or vertical surfaces. After finishing the examination you can easily remove it with cold water. Using water as a coupling in frosty weather can be difficult because it freezes. Adding spirit is an easy and cheap solution. Better, but more expensive is the use of glycerine, which at the same time spares the hands of the user.
7-49 7-49/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.22.7
Lamination examination of rolled plates and profiles Lamination examination is carried out with a normal probe or a TR-probe. If you place a normal probe on a plate with no defects you get a row of echoes, where the mutual distance corresponds to the thickness of the plate. (Figure 7-24a). If there is lamination present (Figure 7-24b) the pulse is reflected from the lamination and the distance between the echoes is reduced corresponding to this. The size of the lamination can be estimated by moving the probe along the edge of the lamination. Normally a lamination is extensive slag inclusions in the middle of a plate, but sometimes it consists of small slag inclusions, which can be either in the same depth or distributed in the entire thickness of the plate. From this type of defect you still have echoes corresponding to the normal thickness of the plate, but between those you see small echoes from the slag inclusions (Figure 7-24c).
Figure 7-24:
7-50 7-50/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Laminated plate material is rather common. It is very dangerous to use such plates in welded constructions in places, where pull occurs at right angles to the surface of the plate. Before welding a fitting onto a plate or a profile, you should carry out a lamination examination to make sure that the material is not laminated. Figure 7-25 shows what can happen if a fitting is welded on to a laminated plate.
Figure 7-25:
The laminations arise when inclusions and hollowness from the chill mould are imperfect rolled. A plate will therefore mostly contain laminations in the areas shown on Figure 7-26.
Figure 7-26:
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Testing a plate can be done manually with a plate tester as shown on Figure 7-27 or in an automatic ultrasonic testing installation, where the plate is moved past a row of probes for example 40-80 probes according to the width of the plate. Each probe scans the plate along a line and the results are registered on a paper slip.
Figure 7-27:
The results can be stored in a computer with the specifications from the client as regards for dimensions and allowable content of laminations and slag inclusions. The computer then cuts out the plate and distributes these according to order. This system often causes that clients who do not specify permissible content of laminations and slag inclusions may have screened out plates, which may contain defects. Testing of rolled profiles for laminations in body or flanges is done in the same way as for lamination testing of plates. Another situation where it is important to check for lamination is e.g. where 2 plates are welded together. If a lamination is present in the area where you would scan with angle probes the sound beam will do as shown in Figure 7-28.
7-52 7-52/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-28:
The lamination testing of plates are normally carried out with a nominal frequency of 2 or 4 MHz. Plate thicknesses down to about 5 mm can be tested quite accurate. For testing thinner plates you must use a higher frequency or TR-probes. When using a TR-probe you normally look for defect echoes showing up before 1st bottom echo. You must also be aware that the maximum sensitivity from TR-probes can be in different depths depending on the angle between the transmitting and receiving crystal. The relationship between defect and bottom echo will vary with this angle and is different compared to a normal probe. The permissibly extent of lamination in a plate varies of course with application. If the plate is to be used to transmit great tractive forces at right angle to the surface, you must ask for complete lamination free material. For other purposes you may tolerate smaller areas with lamination. In order to carry out a lamination examination you need an agreement between the parties involved. This agreement concerns the extent of examination, the size of defects and the number of defects according to a normative reference. Examination of castings In the production of casting defects like cracks, porosity, and big gas bubbles, shrinkage and sand inclusions may occur. Ultrasonic examination of such defects can be done in most cases depending on the following conditions: The penetrating power of material structures Certain materials like steel and aluminium are easy to penetrate for ultrasonic oscillations. Others are more difficult or impossible to penetrate for example grey iron, bronze and stainless steel. The attenuation of the sound beam is due to the reflection from graphite flakes in cast iron, and from segregation's in grain boundaries in the other materials. SG-iron is easy to penetrate for ultrasonic oscillation because the graphite flakes is found in the shape of small balls. 7-53 7-53/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
For some materials the penetration power varies with the cooling rate the material has been exposed to after the casting. By quick cooling you get a fine-grained structure with small segregation in the grain boundaries and with a good penetration for ultra sound. The penetration power of a specimen depends on frequency of the sound wave. Sometimes you get sufficient penetration power to carry out the examination by choosing a low frequency e.g. 0.5 MHz. At the same time the lower limit is raised for the size of the defect, which can be detected. The reflection ability of defect surfaces In castings you can find defects with so uneven surfaces that all of the sound is reflected away, which means you do not get a flaw echo. See Figure 7-29.
Figure 7-29:
Such defects can only be found using the through transmission technique with a separated transmitter and receiver probe. As long as the receiver detects pulses from the transmitter the subject is accepted. If the receiver detects nothing, there must be a defect in the intervening material. Examination of rivet joints In rivet joints on steam boilers you can have stress corrosion or caustic breaking cracks in the plates. The cracks start from the rivet holes and propagate from one rivet to another (Figure 7-30).
7-54 7-54/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-30: Examination of rivet joints for cracks in plates and rivets. The cracks can be found with angle probes as shown in Figure 7-30. Cracks with the same character can arise in the rivet shank itself. They can be found with a normal probe from the end of the rivet. Spurious echoes
7-55 7-55/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Besides the direct and expected form echoes e.g. echoes from a back wall, an edge in a plate, a welding cap or a recess and defect echoes from internal defects in the object, you can have other echoes as well - the so called spurious echoes. These echoes arise when the sound beam can be reflected in more different ways inside the object, making the pulses return to the probe in different sound paths. These sound paths normally have different length and the pulses do not return at the same time, which means that equipment registers several echoes after one another. A delay in time of a longitudinal wave can also be due to the fact, that it has been transformed into a transverse wave on a part of the sound path, where it has travelled at the smaller velocity of the transverse wave. If the sound beam is spread out more than expected or if the subject has another form than expected, it can also give unexpected echoes on the screen. Below some common occurring spurious echoes are mentioned.
Examination of long objects with a normal probe A normal probe is placed on the top or the edge of a long object, which is narrow, compared to the sound beam. A part of the scattered sound beam will be reflected from the sides and travel in paths, which are longer than the direct sound path. This is seen as a row of successive echoes, as shown in Figure 7-31. When reflecting against the side, the longitudinal wave is split up. A part continues as a transverse wave, which again is transformed into longitudinal waves after a later reflection. Finally the sound beam returns to the probe as a longitudinal wave and you can see an echo on the screen. Figure 7-31 shows some of the possible sound paths in a long object. Sound wave 1 gives 1st bottom echo. No. 2 gives 1st spurious echo after a single reflection of the longitudinal wave on the side. No. 3 gives 2nd spurious echo after a transformation into a transverse wave on a part of the sound path. No. 4 gives the 3rd spurious echo after repeated reflections of the transverse wave. All spurious echoes shows up after the 1st bottom echo - and will not interfere with the examination, if the object has the same cross section over the whole length. If you on the other hand have recesses or shrinked on bushing, the spurious echoes will be able to cover possible echoes from defects in more distant areas of the object.
7-56 7-56/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Spurious echoes after the bottom echo when examining long ob-
7-57 7-57/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.22.8 Examination of welds Application and purpose The extensive use of ultrasound for examination of welds is due to the fact that it is one of the most reliable and cheapest ways of getting information about the quality of a weld. The method has also shown that it is suitable for automation and it is used for continuous control of welding plants, where large amounts of identical welds are produced under factory-like circumstances. For examination of manual welds the ultrasound method is suitable as well. In this case manual examination is more common and will be described in the following and which still is a model for the more simplified automatic examination. When examining a weld it is all about getting a detailed and accurate picture of the existing defects, in order to record their position and size in a report, which is independent of the used equipment and the team using the equipment. The report should give the client a satisfactory foundation in order to make his decision on the quality of the weld. The ability of a defect to reflect Because the method is build on the ability of the defects in the weld to reflect the used sound waves, we will first go through the welding defects common in practice, seen from this point. The defects can be split into planar defects, meaning defects which are very small in one direction, but has a certain size in the two other directions as e.g. cracks, lack of penetration and lack of side wall fusion see Figure 7-32, and volumetric defects such as defects, which has a certain extension on all 3 sides as e.g. slag lines, slags in lack of penetration, gas porosity and slag inclusions, see Figure 7-33.
Figure 7-32: A + B: C: D: E:
Planar defects.
Longitudinal, oblique. E.g. lack of sidewall fusion cracks Longitudinal, right angle to the surface. E.g. lack of penetration and cracks Longitudinal, parallel to surface. E.g. thin layer of slags Transverse, right angle to the surface. E.g. transverse cracks
7-58 7-58/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The planar defects will reflect the sound waves well, if the extension in both directions across the beam direction is larger than about a wavelength. The reflection will then happen according to the laws of the optics and a necessary condition for the defect to be found is the sound beam being either directly or via a close by surface reflected back to the probe. The volumetric defects will reflect sound beams more or less scattered. The best reflections you will get, is from a defect with an almost plane/level surface, which is at right angles to the wave. That is found on e.g. slag lines and lack of penetration (Figure 7-33F + G). Less good reflections you get from wormholes, even though they are lengthy in the direction of the weld (Figure 7-33 H). The worst reflection is from porosity and scattered slag inclusion (Figure 7-33 J + K).
Figure 7-33: F: G: H: J: K:
Volumetric defects.
Examination technique Because of the above mentioned possible location and orientation of welding defects the method of testing must be carried out in such a way that as many defects as possible are found. In a lot of cases where you know the welding method, you can exclude some defects and simplify the method of testing. For instance is testing for 7-59 7-59/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
transverse flaws not necessary in welding in plain carbon steel with low yield point, as transverse flaws are very rare. Butt welds in thinner plates For this you should use an angle probe placed on the surface of the plate besides the welding as shown on Figure 7-34. The sound beam travels obliquely into the plate and is reflected alternately from the lower and higher surface of the plate, so the beam describes a zigzag path in the plate. An important condition for this method to work correctly is that the plate is free of lamination, which can reflect the beam, before it reaches the opposite side.
Pos. 2
Pos. 1
Figure 7-34:
An effective test for the previous mentioned types of defects needs the following movements of the probe (Figure 7-34).
7-60 7-60/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Calibrating the range and measuring the beam angle is done in the easiest way by using an IIW-block. You find it in two types, type 1 for angle probes of almost all common sizes and type 2 for miniature angle probes. The shape of these calibration blocks are shown in Figure 7-36 + Figure 7-37.
Figure 7-36:
Calibration block 1
7-61 7-61/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-37:
Calibration block 2.
Using a normal probe on the 91 mm distance does the calibration of the range on type 1, see Figure 7-36. This distance for a longitudinal wave corresponds to a transverse wave having travelled 50 mm sound path. The successive bottom echoes from the 91 mm distance are used to calibrate the range. The normal probe is replaced with the actual angle probe. It is directed towards the 100 mm arc, see Figure 7-38 and the maximum echo is found. The probe index point will be on a level with the mark showing the centre of the arc. The probe index is marked on the side of the probe. The echo from the 100 mm arc is displaced with the delay or zero control, until it is placed on 100 mm on the actual range.
Figure 7-38:
Maximising the echo from the 25 mm or 50 mm arc does the calibration of the range on type 2. Using the 25 mm arc the first echo is placed on 25 mm with the delay or zero control and the second echo is placed on 100 mm with the material velocity control, and the probe in an unchanged position on the block. By using the 50 mm arc the first echo is placed on 50 mm with the delay control and the second echo is placed on 125 mm with the material velocity control (Figure 7-38). Measuring the beam angle is done by directing the sound beam towards the cylindrical drilled hole in the block as shown in Figure 7-39 and find the position of the probe, where maximum echo is obtained. On the grade scale, which is engraved on the side of the block, you can read the beam angle on a level with the earlier marked probe index.
7-62 7-62/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-39:
The sensitivity calibration is partly done by adjusting the pulser which - depending of the type of equipment, - can be regulated stepwise in up to 5 different values - or partly by adjusting the gain of the receiver in dB steps. Which sensitivity you choose, depend on how big defects are allowed in the welding. This information you must have and you should have a specification made by the constructor for the specific welding, which indicates the maximum allowed defects and how many minor defects is allowed pr. m weld and what types of defects you can tolerate. These tolerances can be found in national and international standards and codes or in specific procedures. In practice you manage by using a sensitivity calibration corresponding to that of a well defined, artificial defect like a cylindrical drilled hole parallel to the contact surface and in the same distance as the actual defects, gives an echo of a certain size. (Figure 7-40). It is mandatory that the test block is made of a type of steel, which has the same attenuation as the test piece and that the surface is similar to the test piece as well.
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The test is done with this calibration, so that all defect indications equal to or bigger than the indication from the drilled hole is noted in the report. It must be noticed that if a defect lays in a considerable other distance (sound path) from the probe than the drilled hole, you must adjust the sensitivity again on another drilled hole in the same distance as the defect. On basis of the above mentioned calibrations and measurements of the used equipment the location of defect and the marking can be done. Moving the probe in the previous mentioned ways tests the welding and the screen is monitored all the time. All indications, above the fixed maximum, are noted and in each case you must decide, if it is reflections from excess weld metal, undercut, backing or other outer limitations. If this is not the case the position of the defect in the horizontal direction from the probe and in depth is calculated. If a defect is big compared to the cross section of the sound beam, the limitations of the defect shall be laid down. When calculating the horizontal distance and a vertical depth you use different remedies. a) You can mathematically calculate the horizontal and the vertical projection of the read off sound path (s) see Figure 7-41, because the horizontal projection a is
a = s x sin
a s d
Figure 7-41:
cal depth "d" according to the read off sound path. and the vertical projection (d) is
d = s x cos
7-64 7-64/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.22.9 Determination of defect size Introduction Ultrasonic examination of welds and materials normally involves that the found defects are estimated partly to make sure where the place of reflection is and partly to get information about the size and type of the defect. Determination of the size of large defects In the cases, where the extension of the defect in one or in both directions, measured at right angles to the sound beam, is large compared to the sound beam, you estimate the size by an acoustic scanning, determining the boundaries of the area, from where you have the defect indicated. To get precise information from where to where the defect runs, you use the socalled half value boundaries or 6 dB-drop method as the positions of the probe, where the echo height has just reduced to half the normal height from a place on the defect. The limits are determined by displacing the probe either laterally parallel to the weld direction in order to determine the length (Figure 7-42), or forward and beckwards at right angles to the weld direction to determine the extension of the depth.
Figure 7-42:
The accuracy of the defect size based on the half values can be rather good, but sometimes you get large deviations from the actual sizes. Decisive for that is both the surface of the defect and the characteristca of the soundbeam. The method has the advantage of being quick and gives in many cases adequate accurate and unambiguous results
7-65 7-65/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Determination of the width of defects is not always necessary, but can be used to identify the places of reflection close the opposite surface. It is determined by measuring the horizontal distance to the defect from both sides of the weld. As shown in Figure 7-43 an incomplete penetration will show a relatively wide defect, a crack or a lack of penetration a quit narrow defect, and large excess penetration as a defect with a negative width.
Positive defect width
a1
a2
Defect width
a1
a2
a1
a2
Figure 7-43: A: Incomplete penetration gives great defect width B: Lack of penetration and cracks in the root give a small width C: Excess penetration gives negative defect width
7-66 7-66/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Determination of the size of small defects When a reflective surface becomes small compared to the cross section of the sound beam, you will not be able to limit it by moving the sound beam. On the other hand the amount of reflected energy and with that the echo height will give information about the size of the defect. There is however a number of factors, which plays a role, so it is not possible to give a simple connection between the size of a defect and the echo height you get on the screen. Factors having an influence on the size of the flaw echo are: 1. 2. 3. 4. 5. 6. 7. 8. Pulse energy Amplification Coupling between probe and object Type of defect, shape of defect (plane, spherical etc. ) The reflecting surface (roughness) Orientation of defects in relation to the direction of sound The position of the defect in the sound field, the distance between defect and probe The acoustic properties of the object.
One of the most important factors is the type and shape of the defect, which makes the ability of reflection vary a lot from defect to defect. Also the orientation of the defect in relation to the direction of sound is of great importance for the size of the echo. The significance of the position of the defect in the sound field. Near field and far field The size of an echo from a small reflector will generally decrease with increasing distance from the probe. It will happen in a regular way at larger distances, but if the defect gets closer to the probe, you will see large variations in echo height for small changes of distances. That is due to interference, which can occur in the socalled near field (Figure 7-44).
7-67 7-67/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-44:
Schlieren image of sound field, showing near field and far field.
Comparison with artificial defects As seen from the previous situation, it can be rather complicated to determine the size of a defect only on basis of the echo of the defect. It is therefore reasonable to try to simplify the methods to determine the size of small defects. In practice it has been proved very suitable to use a comparison of the actual indications of defects with the indications, you can get from artificial defects of known size. Such defects can be worked out with a well-defined shape e.g. as a cylindrical drilled hole, flat-bottomed cylindrical holes or right-angled grooves. They must be placed in reference blocks with such dimensions that the same distance from the probe can be reached as the actual defect distance in the test piece. An example of such a reference block with cylindrical drilled holes is shown in Figure 7-45. It is also important, that the reference block is of a material, which attenuates the sound just as much as the actual material and that the surface is of the same nature, so the same degree of coupling is reached.
Figure 7-45: Reference block with cylindrical bored holes for comparing echo sizes. 7-68 7-68/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Such artificial defects are used to calibrate the echo height to a certain height. With unchanged calibration you carry out the examination and all indications, which reaches a certain level are noted. The method of comparison is normally satisfactory for approval control, because you rarely will have difficulties concerning characteristics, sizes and surfaces of the reference blocks. One must not forget that such artificial defects normally will represent optimum conditions of reflections and you will in the principle from such a comparison only be able to find the smallest value of defects. DGS diagram If reference blocks with artificial defects cannot be produced or if it appears to be awkward to use a large number of different reference blocks, you can use the DGS diagram. (D = distance, G = gain, S = size) Such a diagram valid for 4 MHz angle probes of the type Krautkrmer MWB70 - 4 is shown in Figure 7-46. Here you can read the size of a reflecting surface on basis of the sound pressure in the reflected sound beam in relation to the distance of sound, which means the distance between the reflector and the probe index on the probe.
7-69 7-69/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-46:
7-70 7-70/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Non Destructive Testing. Applied Materials Research. The British Journal of Non-Destructive Testing. Magnafacts. Materials Evaluation. Ultrasonics. INFO, NDT. Materials Research and Standards. Das Echo. Materialprfung. Technische berwachung. Schweissen + Schneiden.
Books:
L. Filipozynski and others "Ultrasonic Methods of Testing Materials". Butterworths. England. 1966. B. Banks and others "Ultrasonic Flaw Detection in Metals. Theory and Practice". Lliffe Books Ltd. London. 1962. Benson Carlin. "Ultrasonics". McGraw-Hill Book Company, Inc. 1949. J. F. Hinsley. "Non-Destructive Testing". MacDonald & Evans Ltd. London. 1959. W. J. McGonnagle. "Non-Destructive Testing". McGraw-Hill Book Company, Inc. London. 1961. J. Blitz. "Fundamentals of Ultrasonics". Butterworths. England. 1963. Non-Destructive Testing Handbook. Vol. I-II. The Ronald Press Co. 1959. Non-Destructive Testing, Programmed Instruction Handbooks, General Dynamics. Krautkrmer. "Werkstoffprfung mit Ultraschall". 2. edition. Springer-Verlag. 1966. Vaupel. Bild-Atlas fr die zerstrungsfreie Materialprfung. I-II-III. J. Matauscheck. "Einfhrung in die Ultraschalltechnik". VEB Verlag Technik. Berlin. 1957. E. A. W. Mller. "Handbuch der zerstrungsfreien Materialprfung". R. Oldenburg. 1963. Ludwig Bergmann. Der Ultraschall. S. Hirzel Verlag, Stuttgart. 1954.
7-71 7-71/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Stahl-Eisen Lieferbedingungen 072-69. (1. edition December 1969): Ultraschallgeprftes Grobblech. Schweissen + Schneiden, Heft 6/66. Rudolf Trumpfheller, Technischer berwachungs-Verein, Essen a. V.: Abnahmeprfungen an Schweissnhten nach dem Ultraschallprfverfahren.
Deutsche Industrie Normen:
DIN 54119. Vornorm Zerststrungsfreie Prfung; Ultraschallprfung, Begriffe. DIN 54120. Vornorm Zerstrungsfreie Werkstoffprfung; Kontrol1krper 1 und seine Verwendung zur Justierung u. Kontrolle von UltraschallImpulsecho-Gerten. DIN 54122. Entwurf Zerstrungsfreie Werkstoffprfung; Kontrol1krper 2 und seine Verwendung zur Justierung u. Kontrolle von Ultraschall Impulsecho-Gerten.
ASTM Standards:
A 435-67. Standard Method and Specification for Longitudinal-Wave Ultrasonic Inspection of Steel Plates for Pressure Vessels. E 317, Part 31. Evaluating performance characteristics of pulseecho ultrasonic testing systems. Rec. practice for. E 127, Part 31. Fabricating and checking aluminium alloy ultrasonic standard reference blocks. Rec. practice for. E 214, Part 31. Immersed ultrasonic testing by the reflection method using pulsed longitudinal waves. Rec. practice for. A 578, Part 4. Longitudinal wave ultrasonic testing and inspection of plain and clad steel plates for special applications. Spec. for. A 435, Part 4. Longitudinal wave ultrasonic inspection of steel plates for pressure vessels. E 164, Part 31. Ultrasonic contact inspection of weldments. A 503, Part 4. Ultrasonic examination of large forged crankshafts. Rec. practice for. E 273, Part 31. Ultrasonic inspection of longitudinal and spiral welds of welded pipe and tubing. E 213, Part 31. Ultrasonic inspection of metal pipe and tubing for longitudinal discontinuities. A 531, Part 4. Ultrasonic inspection of turbine-generator steel retaining rings. Rec. practice for.
7-72 7-72/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
A 577, Part 4. Ultrasonic, shear wave inspection of steel plates. Spec. for. A 388, Part 4. Ultrasonic testing and inspection of heavy steel forgings. Rec. practices for. A 418, Part 4. Ultrasonic testing and inspection of turbine and generator steel rotor forgings. E 114, Part 31. Ultrasonic testing by the reflection method, using pulsed longitudinal waves induced by direct contact. Rec. practice for. E 113, Part 31. Ultrasonic testing by the resonance method. Rec. practice for.
British Standards:
BS 2704. Specification for calibration blocks and recommendations for their use in ultrasonic flaw detection. BS 3683. Glossary of terms used in non-destructive testing. BS 3923. Methods for ultrasonic examination of welds. BS 4331. Methods for assessing the performance characteristics of ultrasonic flaw detection equipment.
Ultrasonic standards: DS/EN 12062 DS/EN 1712 DS/EN 1713 DS/EN 1714 DS/EN 10160 DS/EN 10308 DS/EN 583-2 DS/ENV 583-6 DS/EN 12680-1 DS/EN 12680-2 DS/EN 12680-3 DS/EN 14127 ASME V Art. 4 Art. 5 VIII Art. 9-3 ADM HP 5/3
General rules for metallic materials UT of welded joints. Acceptance levels UT examination. Characterization of indications in welds UT of welded joints (technique) UT of steel flat product (lamination) UT of steel bars Sensitivity and range setting TOFD-technique (sizing of defects) Steel castings for general purposes Steel castings for highly stressed components Spheroidal graphite cast iron castings Ultrasonic thickness measurement UT for inservice inspection UT methods for materials and fabrication UT examination of welds Zerstrungsfreie Prfung der Schweisverbindungen
7-73 7-73/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.23
7-74 7-74/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.23.2 Thickness measurements of steel plates The thickness of a steel plate can be measured with a normal probe or a TR-probe, which is placed on the surface. See Figure 7-47.
Figure 7-47:
Before starting the measurements, a contact liquid is applied to the test locations. Further, the equipment must be calibrated. For common steel alloys, the calibration is performed by means of test blocks. For unknown alloys (or if you are not sure), the calibration is performed by adjusting the sound velocity setting of the equipment until the equipment shows the same thickness as can be measured by a slide calliper at a free edge. After calibrating the range of the ultrasonic equipment using a suitable calibration block for example a 5, 10 or 25 mm thick steel block, you can read off the thickness of the plate in different ways. Reading off the position of 1st bottom echo This method is often used, if the plate has a very uneven surface, which only gives you the 1st bottom echo. You would normally use a TR-probe. See Figure 7-48.
TR-probe
t
Figure 7-48: Thickness measurement of uneven steel plate. 7-75 7-75/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
At plate thicknesses below 20 mm, the range shall preferably be calibrated to 25 or 10 mm sound path in steel to achieve the best possible reading accuracy. Another option is to use waterproof equipment. That means that you can bring both the transducer and the apparatus with you into the water. Reading off the position of one of the successive bottom echoes If more successive bottom echoes turns up on the screen you choose one, which is placed to the right of the screen. You count the number by counting echoes from the left, and you read off its position on the screen, that is the equivalent sound path, see Figure 7-49.
The equipment time base range should be calibrated in order to give you as many bottom echoes as possible on the screen. The gain should not be greater than corresponding to 1st bottom echo from the calibration block in scale height. This will give you the most accurate reading. Digitalised equipment does not need successive echoes because the reading is done where the gate intersects with the echo you want to use.
Reading off the position of one of the successive bottom echoes, with 1st or 2nd bottom echo on 0 on the scale If the surface of the plate is covered with paint or scale or if it has just become rusty or rough, this method should be used. Doing this you avoid measuring the thickness of the layer covering the surface, no matter if it is stuck to the surface or it is coupling agent on a corroded surface. Such coatings can result in rather considerable errors. The velocity in for example water is about of the velocity in steel. A 0.5 mm thick layer of water will therefore give an error of 2 mm, if 1st bottom echo is read off. The reading of the thickness is done exactly in the same way as for the previous method. 7-76 7-76/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
You should be aware when using this method that it is necessary to place 1st or 2nd bottom echo on 0 on the scale before each measurement. See Figure 7-50.
Todays' digital equipments are using so-called "gates" which means that you do not have to move 1st or 2nd echo to zero. You place the 2 gates in order to measure between 2 successive echoes. See Figure 7-51.
Layer
Figure 7-51:
7.23.3 Special equipment Thickness measurements can also be done with special equipment, which normally only are intended for this purpose. They are called thickness gauges and they can operate with TR-probes or normal probes. The gauges normally give you the thickness in digits but some gauges have both digits and an oscilloscope. Some of these gauges are capable of compensating for layers. See Figure 7-52. 7-77 7-77/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-52:
Thickness gauge for measuring through coating (left), and (right) for use without coating.
These gauges are normally calibrated in steel. When measuring on other materials the results must be corrected as shown later. 7.23.4 Thickness measurements of hot steel plates With normal equipment you can carry out thickness measurements on objects with a temperature up to about 70 C. Using a special heat resistant probe it is possible to measure hot objects with a temperature as high as 800 C, but you must be aware of the following:
a.
The probes are normally TR-probes, where the delay line is made from heatinsulated material. The probe should be cooled in water frequently The velocity in the material in the delay line drops significant with temperature. The echoes on the screen will due to this move towards the right of the screen. Often the screen is photographed to get a fast and accurate reading or you could use the freeze function. The attenuation in the insulating material increases with temperature. Due to that the reading must be done fast and you should have enough gain reserve in the ultrasonic equipment. The coupling is often silicone, which evaporates in a few seconds at 400 C. Therefore there is no time to adjust the probe on the object to get the clearest picture on the screen. You must use a fixture, to make sure, that the contact surface is parallel with the object surface during the movement towards the object, so that you immediately have a good contact.
b.
c.
d.
7-78 7-78/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
e.
The velocity in steel decreases with increasing temperature. In the area 0-400 C it decreases with approximately 1% per 100 C. A 400 C hot object is measured approximately 4% too thick. The results must therefore be corrected.
7.23.5 Thickness measurements of other materials than steel Most metals, porcelain, glass and some plastic materials can be measured in the same way as steel. The equipment time base range should be calibrated on a calibration block of the material concerned. If this cannot be produced, but you know the velocity of the material, you can calibrate the equipment on a steel block and then read off the thickness as usual. The results have to be corrected by multiplication with a correction factor. This factor depends on the velocity in the material compared to the velocity in steel. If the sound for example only travels 0.8 times as fast in copper than in steel, the read thickness of a copper plate has to be multiplied by 0.8 to get the real thickness. Normally applies that the real thickness = the read thickness multiplied by the proportion between the velocity of the material and the velocity in steel. For velocities see Table 7-2.
t real = t read x
v material v steel
Measuring the velocity for longitudinal waves You can measure the velocity of longitudinal waves in an object with ordinary ultrasonic equipment and probe in the following way: Measure the real thickness mechanically using a gauge. Then measure the thickness in the same place with the ultrasonic equipment. The unknown velocity can be found using this formula:
v material =
7-79 7-79/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Material
Velocities in m/sec. VL longitudinalwaves 6320 2160 3300 6800-7300 4700 5240 1450 5770 4660 3830 5630 3500-5600 5900 3320 6070 2180 5460 4170 1000 2500 1480 4260 2300 2200 3980 5660 5570 1800-2200 2730 2670 5600-6200 1350 1250 1920 1740 1483
VT transverse waves 3130 700 4000-4700 2260 2640 3050 2350 2050 2960 2200-3200 3230 1670 3110 1100 2620 2410 1100 2560
Acoustic impedance Z = x VL kg 10 6 2 m sec. 17 25 24 77-102 42 46 20 10 39 31 50 25-40 45 24 27 21 104 30 39 1.4 15 2.8 1.8 3.6 14 14.5 2.0-2.7 3.2 2.8 13 3.0 1.0 2.5 1.5 1.5
M E T A L S
N O N M E T A L S
Aluminium Lead White metal Hard metal Copper Constantan Mercury Magnesium Manganin Brass Nickel Cast iron Steel Tin Titanium Bismuth Wolfram Zinc Aluminium oxide Araldit Rubber, soft Glass, flint Rubber, vulcanised Paraffin wax Ice Glass, crown Quartz glass Nylon, perlon Perspex Polystyrene Porcelain Teflon Diesel oil Glycerine Motor oil Water (20 C)
2.7 11.4 7.3 11-15 8.9 8.8 13.6 1.7 8.4 8.1 8.8 7.2 7.7 7.3 4.5 9.8 19.1 7.1 3.9 1.18 0.9 3.6 1.2 0.83 0.9 2.5 2.6 1.1-1.2 1.18 1.06 2.4 2.2 0.80 1.26 0.87 1.0
L I Q U I D S
Table 7-2:
Density, velocities and acoustic impedance for different materials. 7-80 7-80/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.24
Magnetic film gages are used to nondestructively measure the thickness of a nonmagnetic coating on ferrous substrates. Most coatings on steel and iron are measured this way. Magnetic gages use one of two principles of operation: magnetic pulloff or magnetic/electromagnetic induction. Magnetic Pull-off Magnetic pull-off gages use a permanent magnet, a calibrated spring, and a graduated scale. The attraction between the magnet and magnetic steel pulls the two together. As the coating thickness separating the two increases, it becomes easier to pull the magnet away. Coating thickness is determined by measuring this pull-off force. Thinner coatings will have stronger magnetic attraction while thicker films will have comparatively less magnetic attraction. Testing with magnetic gages is sensitive to surface roughness, curvature, substrate thickness, and the make up of the metal alloy.
Figure 7-53:
Magnetic pull-off gages are rugged, simple, inexpensive, portable, and usually do not require any calibration adjustment. They are a good, low-cost alternative in situations where quality goals require only a few readings during production. Pull-off
7-81 7-81/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
gages are typically pencil-type or rollback dial models. Pencil-type models (PosiPen shown in Fig 1) use a magnet that is mounted to a helical spring that works perpendicularly to the coated surface. Most pencil-type pull-off gages have large magnets and are designed to work in only one or two positions, which partially compensate for gravity. A more accurate version is available, which has a tiny, precise magnet to measure on small, hot, or hard-to-reach surfaces. A triple indicator ensures accurate measurements when the gage is pointed down, up, or horizontally with a tolerance of 10%. Rollback dial models (PosiTest shown in Figure 7-54) are the most common form of magnetic pull-off gage. A magnet is attached to one end of a pivoting balanced arm and connected to a calibrated hairspring. By rotating the dial with a finger, the spring increases the force on the magnet and pulls it from the surface. These gages are easy to use and have a balanced arm that allows them to work in any position, independent of gravity. They are safe in explosive environments and are commonly used by painting contractors and small powder coating operations. Typical tolerance is 5%.
Figure 7-54:
Magnetic and Electromagnetic Induction Magnetic induction instruments use a permanent magnet as the source of the magnetic field. A Hall-effect generator or magneto-resistor is used to sense the magnetic flux density at a pole of the magnet. Electromagnetic induction instruments use an alternating magnetic field. A soft, ferromagnetic rod wound with a coil of fine wire is used to produce a magnetic field. A second coil of wire is used to detect changes in magnetic flux. These electronic instruments measure the change in magnetic flux density at the surface of a magnetic probe as it nears a steel surface. The magnitude of the flux density at the probe surface is directly related to the distance from the steel substrate. By measuring flux density the coating thickness can be determined. Electronic magnetic gages (e.g. PosiTector 6000 F Series shown in Figure 7-55, PosiTest DFT Ferrous) come in many shapes and sizes. They commonly use a constant pressure probe to provide consistent readings that are not influenced by different operators. Readings are shown on a liquid crystal display (LCD). They can have options to store measurement results, perform instant analysis of readings, and out-
7-82 7-82/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
put results to a printer or computer for further examination. Typical tolerance is 1%. The manufacturers instructions should be carefully followed for most accurate results. Standard test methods are available in ASTM D 1186, D 7091-05, ISO 2178 and ISO 2808.
Eddy Current
Eddy current techniques are used to nondestructively measure the thickness of nonconductive coatings on nonferrous metal substrates. A coil of fine wire conducting a high-frequency alternating current (above 1 MHz) is used to set up an alternating magnetic field at the surface of the instrument's probe. When the probe is brought near a conductive surface, the alternating magnetic field will set up eddy currents on the surface. The substrate characteristics and the distance of the probe from the substrate (the coating thickness) affect the magnitude of the eddy currents. The eddy currents create their own opposing electromagnetic field that can be sensed by the exciting coil or by a second, adjacent coil. Eddy current coating thickness gages (e.g. PosiTector 6000 N Series) look and operate like electronic magnetic gages (see Figure 7-55). They are used to measure coating thickness over all nonferrous metals. Like magnetic electronic gages, they commonly use a constant pressure probe and display results on an LCD. They can also have options to store measurement results or perform instant analysis of readings and output to a printer or computer for further examination. The typical tolerance is 1%. Testing is sensitive to surface roughness, curvature, substrate thickness, type of metal substrate and distance from an edge. Standard methods for the application and performance of this test are available in ASTM B 244, ASTM D 1400, D 7091-05 and ISO 2360.
7-83 7-83/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
It is now common for gauges to incorporate both magnetic and eddy current principles into one unit (e.g. PosiTector 6000 FN, PosiTest DFT Combo). Some simplify the task of measuring most coatings over any metal by switching automatically from one principle of operation to the other, depending upon the substrate. These combination units are popular with painters and powder coaters. 7.24.3 Ultrasonic
The ultrasonic pulse-echo technique of ultrasonic gages (e.g. PosiTector 100 and PosiTector 200 shown in Fig 4) is used to measure the thickness of coatings on nonmetal substrates (plastic, wood, etc.) without damaging the coating. The probe of the instrument contains an ultrasonic transducer that sends a pulse through the coating. The pulse reflects back from the substrate to the transducer and is converted into a high frequency electrical signal. The echo waveform is digitized and analyzed to determine coating thickness. In some circumstances, individual layers in a multi-layer system can be measured. Typical tolerance for this device is 3%. Standard methods for the application and performance of this test are available in ASTM D 6132.
Micrometer
Micrometers are sometimes used to check coating thickness. They have the advantage of measuring any coating/substrate combination but the disadvantage of requiring access to the bare substrate. The requirement to touch both the surface of the coating and the underside of the substrate can be limiting and they are often not sensitive enough to measure thin coatings. Two measurements must be taken: one with the coating in place and the other without. The difference between the two readings, the height variation, is taken to be the coating thickness. On rough surfaces, micrometers measure coating thickness above the highest peak.
7-84 7-84/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Destructive Tests
One destructive technique is to cut the coated part in a cross section and measure the film thickness by viewing the cut microscopically. Another cross sectioning technique uses a scaled microscope to view a geometric incision through the dry-film coating. A special cutting tool is used to make a small, precise V-groove through the coating and into the substrate. Gages are available that come complete with cutting tips and illuminated scaled magnifier. While the principles of this destructive method are easy to understand, there are opportunities for measuring error. It takes skill to prepare the sample and interpret the results. Adjusting the measurement reticule to a jagged or indistinct interface may create inaccuracy, particularly between different operators. This method is used when inexpensive, nondestructive methods are not possible, or as a way of confirming nondestructive results. ASTM D 4138 outlines a standard method for this measurement system. 7.24.6 Gravimetric
By measuring the mass and area of the coating, thickness can be determined. The simplest method is to weigh the part before and after coating. Once the mass and area have been determined, the thickness is calculated using the following equation: T = m x 10 Axd where T is the thickness in micrometers, m is the mass of the coating in milligrams, A is the area tested in square centimeters, and d is the density in grams per cubic centimeter. 7-85 7-85/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
It is difficult to relate the mass of the coating to thickness when the substrate is rough or the coating uneven. Laboratories are best equipped to handle this timeconsuming and often destructive method.
As mentioned before, one of the most used methods to measure the thickness of a nonmagnetic dry coating on ferrous substrates is Magnetic Film Thickness Gages using the principle: Electromagnetic induction The measurements on greater steel-constructions can be done by the so called 80/20 rule. This measurement method specifies the number of measurements and the acceptcriterias:
Figure 7-59:
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
80/20-rule (example): 1. On the construction areas of 10 m2 is selected (minimum 5% of the surface has to be covered). Each selected area shall be connected. 2. 3. Minimum 5 fields of 50 cm2 are selected in each area. In each field measure 3 thickness measurements. Calculate the means of these 3 measurements and look at them as one single measurement. 4. Accept criteria: Only 20% of the total number of single measurements is allowed to be lower than the nominal dry coating thickness, and the lowest value from a single measurement shall be at least 80% of the nominal coating thickness.
7.24.8
Thickness Standards
Coating thickness gages are calibrated to known thickness standards. There are many sources of thickness standards but it is best to ensure they are traceable to a national measurement institute such as NIST (National Institute of Standards & Technology). It is also important to verify the standards are at least four times as accurate as the gage they will be used to calibrate. A regular check against these standards verifies the gage is operating properly. When readings do not meet the accuracy specification of the gage, the gage must be adjusted or repaired and then calibrated again.
Figure 7-60:
7-87 7-87/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Summary Film thickness in coatings can have a big impact on cost and quality. Measurement of film thickness should be a routine event for all coaters. The measurements may e.g. be performed at random locations in a homogenous area where the coating thickness may be assumed to exhibit only a random variation. These measurements may be used to determine the mean and standard deviation of the coating thickness. On the basis of these measurements it may be determined whether the coating thickness complies with the required thickness. The measurements may also be performed at a number of known problem areas such as corners, edges and inaccessible areas. Based on these measurements it may be decided to perform patch repairs of the coating. The correct gage to use depends on the thickness range of the coating, the shape and type of substrate, the cost of the gage, and how critical it is to get an accurate measurement.
7.25
Dye penetrant
This method can be used for steel structures. 7.25.1 Introduction and History of Penetrant Testing Dye penetrant inspection is a method that is used to reveal surface breaking flaws by bleed out of a colored or fluorescent dye from the flaw. The technique is based on the ability of a liquid to be drawn into a "clean" surface breaking flaw by capillary action. After a period of time called the "dwell," excess surface penetrant is removed and a developer applied. This acts as a "blotter." It draws the penetrant from the flaw to reveal its presence. Colored (contrast) penetrants require good white light while fluorescent penetrants need to be used in darkened conditions with an ultraviolet "black light". A very early surface inspection technique involved the rubbing of carbon black on glazed pottery, whereby the carbon black would settle in surface cracks rendering them visible. Later it became the practice in railway workshops to examine iron and steel components by the "oil and whiting" method. In this method, a heavy oil commonly available in railway workshops was diluted with kerosene in large tanks so that locomotive parts such as wheels could be submerged. After removal and careful cleaning, the surface was then coated with a fine suspension of chalk in alcohol so that a white surface layer was formed once the alcohol had evaporated. The object was then vibrated by being stroked with a hammer, causing the residual oil in any surface cracks to seep out and stain the white coating. This method was in use from the latter part of the 19th century through to approximately 1940, when the magnetic particle method was introduced and found to be more sensitive for the ferromagnetic iron and steels. 7-88 7-88/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-61:
A different (though related) method was introduced in the 1940's, where the surface under examination is coated with a lacquer, and after drying the surface is vibrated by hitting with a hammer, for example. This causes the brittle lacquer layer to crack generally around surface defects. The brittle lacquer (stress coat) has been used primarily to show the distribution of stresses in a part and not finding defects. Many of these early developments were carried out by Magnaflux in Chicago, IL, USA in association with the Switzer Bros., Cleveland, OH, USA. More affective penetrating oils containing highly visible (usually red) dyes were developed by Magnaflux to enhance flaw detection capability. This method, known as the visible or colour contrast dye penetrant method, is still used quite extensively today. In 1942, Magnaflux introduced the Zyglo system of penetrant inspection where fluorescent dyes were added to the liquid penetrant. These dyes would then fluoresce when exposed to ultraviolet light (sometimes referred to as "black light") rendering indications from cracks and other surface flaws more readily visible to the inspectors' eyes.
Figure 7-62:
7-89 7-89/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.25.2
Improving Detection
Figure 7-63:
Crack indication.
The advantage that a dye penetrant inspection offers over an unaided visual inspection is that it makes defects easier to see for the inspector. There are basically two ways that a penetrant inspection process makes flaws more easily seen. First, LPI produces a flaw indication that is much larger and easier for the eye to detect than the flaw itself. Many flaws are so small or narrow that they are undetectable by the unaided eye. Due to the physical features of the eye, there is a threshold below which objects cannot be resolved. This threshold of visual acuity is around 0.003 inch for a person with 20/20 vision. The second way that LPI improves the detect ability of a flaw is that it produces a flaw indication with a high level of contrast between the indication and the background which also helps to make the indication more easily seen. When a visible dye penetrant inspection is performed, the penetrant materials are formulated using a bright red dye that provides for a high level of contrast between the white developer that serves as a background as well as to pull the trapped penetrant from the flaw. When a fluorescent penetrant inspection is performed, the penetrant materials are formulated to glow brightly and to give off light at a wavelength that the eye is most sensitive to under dim lighting conditions.
Figure 7-64:
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.25.3 1.
Basic Processing of a Dye Penetrant Testing Surface Preparation: One of the most critical steps of a dye penetrant inspection is the surface preparation. The surface must be free of oil, grease, water, or other contaminants that may prevent penetrant from entering flaws. The sample may also require etching if mechanical operations such as machining, sanding, or grit blasting have been performed. These and other mechanical operations can smear the surface of the sample, thus closing the defects.
Figure 7-65: 2.
Penetrant Application: Once the surface has been thoroughly cleaned and dried, the penetrant material is applied by spraying, brushing, or immersing the parts in a penetrant bath.
Figure 7-66: 3.
Penetrant application.
Penetrant Dwell: The penetrant is left on the surface for sufficient time to allow as much penetrant as possible to be drawn from or to seep into a de-
7-91 7-91/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
fect. Penetrant dwell time is the total time that the penetrant is in contact with the part surface. Dwell times are usually recommended by the penetrant producers or required by the specification being followed. The times vary depending on the application, penetrant materials used, the material, the form of the material being inspected, and the type of defect being inspected. Minimum dwell times typically range from 5 to 60 minutes. Generally, there is no harm in using a longer penetrant dwell time as long as the penetrant is not allowed to dry. The ideal dwell time is often determined by experimentation and is often very specific to a particular application.
4.
Excess Penetrant Removal: This is a most delicate part of the inspection procedure because the excess penetrant must be removed from the surface of the sample while removing as little penetrant as possible from defects. Depending on the penetrant system used, this step may involve cleaning with a solvent, direct rinsing with water, or first treated with an emulsifier and then rinsing with water.
Figure 7-67:
5.
Developer Application: A thin layer of developer is then applied to the sample to draw penetrant trapped in flaws back to the surface where it will be visible. Developers come in a variety of forms that may be applied by dusting (dry powdered), dipping, or spraying (wet developers).
7-92 7-92/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-68:
Developer.
6.
Indication Development: The developer is allowed to stand on the part surface for a period of time sufficient to permit the extraction of the trapped penetrant out of any surface flaws. This development time is usually a minimum of 10 minutes and significantly longer times may be necessary for tight cracks. Inspection: Inspection is then performed under appropriate lighting to detect indications from any flaws which may be present. Clean Surface: The final step in the process is to thoroughly clean the part surface to remove the developer from the parts that were found to be acceptable. Common Uses of Dye Penetrant Inspection
7.
8.
7.25.4
Dye penetrant inspection is one of the most widely used nondestructive evaluation (NDE) methods. Its popularity can be attributed to two main factors, which are its relative ease of use and its flexibility. LPI can be used to inspect almost any material provided that its surface is not extremely rough or porous. Materials that are commonly inspected using LPI include the following:
Metals (aluminum, copper, steel, titanium, etc.) Glass Many ceramic materials Rubber Plastics
LPI offers flexibility in performing inspections because it can be applied in a large variety of applications ranging from automotive spark plugs to critical aircraft components. Penetrant material can be applied with a spray can or a cotton swab to inspect for flaws known to occur in a specific area or it can be applied by dipping or spraying to quickly inspect large areas. The picture to the right above shows, visible dye penetrant being locally applied to a highly loaded connecting point to check for fatigue cracking.
7-93 7-93/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Penetrant inspection systems have been developed to inspect some very large components. In this picture, DC-10 banjo fittings are being moved into a penetrant inspection system at what used to be the Douglas Aircraft Company's Long Beach, California facility. These large machined aluminum forgings are used to support the number 3 engine in the tail of a DC-10 aircraft. Dye penetrant inspection is used to inspect for flaws that breaks the surface of the sample. Some of these flaws are listed below:
Cracks Overload and impact fractures Porosity Laps Seams Pin holes in welds Lack of fusion or braising along the edge of the bond line
As mentioned above, one of the major limitations of a penetrant inspection is that flaws must be open to the surface.
7.25.5
Like all nondestructive inspection methods, dye penetrant inspection has both advantages and disadvantages. The primary advantages and disadvantages when compared to other NDE methods are summarized below.
7-94 7-94/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Primary Advantages
The method has high sensitive to small surface discontinuities. The method has few material limitations, i.e. metallic and nonmetallic, magnetic and nonmagnetic, and conductive and nonconductive materials may be inspected. Large areas and large volumes of parts/materials can be inspected rapidly and at low cost. Parts with complex geometric shapes are routinely inspected. Indications are produced directly on the surface of the part and constitute a visual representation of the flaw. Aerosol spray cans make penetrant materials very portable. Penetrant materials and associated equipment are relatively inexpensive.
Primary Disadvantages
Only surface breaking defects can be detected. Only materials with a relative nonporous surface can be inspected. Precleaning is critical as contaminants can mask defects. Metal smearing from machining, grinding, and grit or vapor blasting must be removed prior to LPI. The inspector must have direct access to the surface being inspected. Surface finish and roughness can affect inspection sensitivity. Multiple process operations must be performed and controlled. Post cleaning of acceptable parts or materials is required. Chemical handling and proper disposal is required.
7-95 7-95/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.25.6
The penetrant materials used today are much more sophisticated than the kerosene and whiting first used by railroad inspectors near the turn of the 20th century. Today's penetrants are carefully formulated to produce the level of sensitivity desired by the inspector. To perform well, a penetrant must possess a number of important characteristics. A penetrant must
spread easily over the surface of the material being inspected to provide complete and even coverage. be drawn into surface breaking defects by capillary action. remain in the defect but remove easily from the surface of the part. remain fluid so it can be drawn back to the surface of the part through the drying and developing steps. be highly visible or fluoresce brightly to produce easy to see indications. must not be harmful to the material being tested or the inspector.
All penetrant materials do not perform the same and are not designed to perform the same. Penetrant manufactures have developed different formulations to address a variety of inspection applications. Some applications call for the detection of the smallest defects possible and have smooth surface where the penetrant is easy to remove. In other applications the rejectable defect size may be larger and a penetrant formulated to find larger flaws can be used. The penetrants that are used to detect the smallest defect will also produce the largest amount of irrelevant indications.
Figure 7-69:
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Penetrant materials are classified in the various industry and government specifications by their physical characteristics and their performance. Aerospace Material Specification (AMS) 2644, Inspection Material, Penetrant, is now the primary specification used in the USA to control penetrant materials. Historically, Military Standard 25135, Inspection Materials, Penetrants, has been the primary document for specifying penetrants but this document is slowly being phased out and replaced by AMS 2644. Other specifications such as ASTM 1417, Standard Practice for Dye Penetrant Examinations, may also contain information on the classification of penetrant materials but they are generally referred back to MIL-I-25135 or AMS 2644. Penetrant materials come in two basic types. These types are listed below: Type 1 - Fluorescent Penetrants Type 2 - Visible Penetrants
Fluorescent penetrants contain a dye or several dyes that fluoresce when exposed to ultraviolet radiation. Visible penetrants contain a red dye that provides high contrast against the white developer background. Fluorescent penetrant systems are more sensitive than visible penetrant systems because the eye is drawn to the glow of the fluorescing indication. However, visible penetrants do not require a darkened area and an ultraviolet light in order to make an inspection. Visible penetrants are also less vulnerable to contamination from things such as cleaning fluid that can significantly reduce the strength of a fluorescent indication.
Figure 7-70:
Penetrants are then classified by the method used to remove the excess penetrant from the part. The four methods are listed below:
Method A - Water Washable Method B - Post Emulsifiable, Lipophilic Method C - Solvent Removable Method D - Post Emulsifiable, Hydrophilic 7-97 7-97/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Water washable (Method A) penetrants can be removed from the part by rinsing with water alone. These penetrants contain some emulsifying agent (detergent) that makes it possible to wash the penetrant from the part surface with water alone. Water washable penetrants are sometimes referred to as self-emulsifying systems. Post emulsifiable penetrants come in two varieties, lipophilic and hydrophilic. In post emulsifiers, lipophilic systems (Method B), the penetrant is oil soluble and interacts with the oil-based emulsifier to make removal possible. Post emulsifiable, hydrophilic systems (Method D), use an emulsifier that is a water soluble detergent which lifts the excess penetrant from the surface of the part with a water wash. Solvent removable penetrants (Method C) require the use of a solvent to remove the penetrant from the part. Penetrants are then classified based on the strength or detectability of the indication that is produced for a number of very small and tight fatigue cracks. The five sensitivity levels are shown below:
Level - Ultra Low Sensitivity Level 1 - Low Sensitivity Level 2 - Medium Sensitivity Level 3 - High Sensitivity Level 4 - Ultra-High Sensitivity
The major US government and industry specifications currently rely on the US Air Force Materials Laboratory at Wright-Patterson Air Force Base to classify penetrants into one of the five sensitivity levels. This procedure uses titanium and Inconel specimens with small surface cracks produced in low cycle fatigue bending to classify penetrant systems. The brightness of the indication produced is measured using a photometer. The sensitivity levels and the test procedure used can be found in Military Specification MIL-I-25135 and Aerospace Material Specification 2644, Penetrant Inspection Materials.
An interesting note about the sensitivity levels is that only four levels were originally planned but when some penetrants were judged to have sensitivities significantly less than most others in the level 1 category, the level was created.
7.25.7 Penetrants The industry and military specification that control the penetrant materials and their use all stipulate certain physical properties of the penetrant materials that must be met. Some of these requirements address the safe use of the materials, such as toxicity, flash point, and corrosiveness, and other requirements address storage and contamination issues. Still others delineate properties that are thought to be primarily responsible for the performance or sensitivity of the penetrants. The properties of
7-98 7-98/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
penetrant materials that are controlled by AMS 2644 and MIL-I-25135E include flash point, surface wetting capability, viscosity, contact angle, color, brightness, ultraviolet stability, thermal stability, water tolerance, and removability. How some of these properties can affect penetrant testing are described next.
Some properties of a penetrant Capillary Action Capillary action is the tendency of certain liquids to travel or climb when exposed to small openings. In nature there are many examples of capillary action. Plants and trees have a network similar to capillary tubes that draw water upward supplying nourishment. The earth brings water to the surface through the capillary action of the earth's exterior. Dye Penetrant Inspection Capillary action is the phenomena that makes dye penetrant inspection possible. All of the steps that are taken in the process of conducting a penetrant test, from precleaning through the actual evaluation of the results, is done to promote capillary action. Precleaning When a part is precleaned, everything is removed that will prevent the penetrant from entering discontinuities and therefore, interfere with capillary action. Once the surface is clean and dry, the penetrant is applied. The penetrant is then drawn into the discontinuities through capillary action, see figure below.
Figure 7-71:
If the part is not clean and dry, less penetrant and possibly none, will be drawn into the discontinuities. Discontinuities that would have been revealed may be overlooked.
7-99 7-99/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Applying Developer Following the removal of the excess penetrant, a developer is applied. The developer induces reverse capillary action to take place. Penetrant is drawn from the discontinuities into the developer in the same way that the fibers of a paper towel absorb or blot a liquid, see figure below.
Figure 7-72:
Surface Tension There are many factors in capillary action; among these are surface tension, cohesion, wetting ability, adhesion and contact angle. Each of these factors has a strong influence in the performance of capillary action. Of these, surface tension is one of the two most important factors. Water in a pond exhibits surface tension when it supports the weight of an insect - a spider or mosquito for example. The insect is supported by a molecular membrane created by the attraction (cohesiveness) of one water molecule to another. Each water molecule is attracted laterally and vertically (above and below) to adjacent molecules. The molecules on the surface are attracted only laterally and below because of the absence of molecules above them. This change in attraction between surface molecules creates the effect of a stretched membrane on the surface of the water strong enough to support small objects. Water has high surface tension because of the strong cohesive attraction between the molecules of water. The amount of surface tension will vary between different liquids depending upon how cohesive the molecules are. Wetting Ability and Contact Angle The second most important factor in capillary action is wetting ability. How well a liquid wets the surface of a specimen is referred to as its wetting ability. The wetting ability of a liquid is determined by the contact angle produced when a liquid meets a surface. The cohesive force that determines surface tension competes with the adhesive properties of a liquid producing a specific degree of contact angle. Adhesion
7-100 7-100/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Adhesion is how strongly the molecules of a liquid are attracted to a particular surface. If a capillary tube is placed in a beaker of water, the water will rise in the tube to a level higher than the water surrounding the tube. The water climbs in the tube because the molecules of water are more strongly attracted to the inside surface of the tube than they are to each other. The stronger the attraction between the molecules of a liquid and a surface, the smaller will be the contact angle and the higher a liquid will rise in a capillary tube. Adhesive Properties of A Liquid When the adhesive properties of a liquid are stronger than the cohesive properties the contact angle will be less than 90 and the liquid will rise in a capillary tube. With a contact angle of less than 90, the liquid is said to have good wetting ability (Figure 7-73 A). The smaller the contact angle, the more the liquid will wet the surface and exhibit greater capillary rise. If the contact angle of a liquid is 90 , it will have poor wetting ability and will not rise in a capillary tube (Figure 7-73 B).
7-101 7-101/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Capillary tube
Droplet Liquid
A. Good wetting ability is obtained when the contact angle is less than 90 A. A contact angle less than 90 will cause capillary rise
Capillary tube
Droplet Liquid
B. Poor wetting ability is obtained when the contact angle is 90 B. A contact angle of 90 provides no capillary rise or depression
Cohesive Properties of A Liquid When the cohesive properties of a liquid are stronger than the adhesive properties the contact angle will be greater than 90 (Figure 7-73 C) and the liquid will descend in a capillary tube. Mercury, as an example, has a large contact angle and will descend in a capillary tube to a level below that of the surrounding liquid. The mercury will not wet the surface because of the large contact angle. 7-102 7-102/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Penetrant Contact Angle Penetrant materials must have a very small contact angle in order to make a good penetrant. Some have contact angles close to 0. Water by itself does not make a good penetrant material, because in addition to having high surface tension, it also has a large contact angle when compared to the oil base that is used in most penetrants. If a wetting agent is added to water, reducing surface tension and the contact angle, it becomes a good penetrant base material that is used in some test applications. Surface Condition Surface condition of the material - roughness, cleanliness, etc., will have an effect on the size of the contact angle. It will change the adhesive properties of the liquid and as a result wetting ability. A surface layer of film will lower adhesion, increase the contact angle, and reduce wetting ability. Size of Opening Another consideration in capillary action is the size of the opening. The narrower the opening, the stronger the capillary action. A liquid with good wetting ability will rise further in a capillary tube that has a small diameter than it will in a tube that has a large diameter (Figure 7-74 A). If a liquid exhibits capillary depression (lack of wetting ability) it will be depressed further in a small tube than a large tube (Figure 7-74, below).
Color and Fluorescent Brightness The color of the penetrant material is of obvious importance in visible dye penetrant inspection, as the dye must provide good contrast against the developer or part being inspected. Remember from our earlier discussion of contrast sensitivity that generally the higher the contrast, the easier objects are to see. The dye used in visible dye penetrant is usually vibrant red but other colors can be purchased for special applications. 7-103 7-103/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-75:
When fluorescent materials are involved, the effect of color and fluorescence is not so straightforward. LPI materials fluoresce because they contain one or more dyes that absorb electromagnetic radiation over a particular wavelength and the absorption of photons leads to changes in the electronic configuration of the molecules. Since the molecules are not stable at this higher energy state, they almost immediately re-emit the energy. There is some energy loss in the process causing the photons to be re-emitted at a slightly longer wavelength, which is in the visible range. The radiation absorption and emission could take place a number of times until the desired color and brightness is achieved. Two different fluorescent colors can be mixed to interact by a mechanism called cascading. The emission of visible light by this process involves one dye absorbing ultraviolet radiation to emit a band of radiation that makes a second dye glow. Since the human eye is the most commonly used sensing device, most penetrants are designed to fluoresce as close as possible to the eyes' peak response. Penetrant Brightness Fluorescent brightness was erroneously once thought to be the controlling factor with respect to flaw detection sensitivity. Measurements have been made to evaluate the intrinsic brightness of virtually all commercially available penetrants and they all have about the same brightness. Intrinsic brightness values are determined for thick liquid films and the dimensional threshold of fluorescence is a more important property. The measurement of fluorescent brightness is detailed in ASTM E-1135, "Standard Test Method for Comparing the Brightness of Fluorescent Penetrants."
7-104 7-104/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Fluorescent brightness.
Emulsifiers
When removal of the penetrant from the defect due to over-washing of the part is a concern, a post emulsifiable penetrant system can be used. Post emulsifiable penetrants require a separate emulsifier to break the penetrant down and make it water washable. Most penetrant inspection specifications classify penetrant systems into four methods of excess penetrant removal. These are listed below: 1. 2. 3. 4. Method A: Water-Washable Method B: Post Emulsifiable, Lipophilic Method C: Solvent Removable Method D: Post Emulsifiable, Hydrophilic
Method C relies on a solvent cleaner to remove the penetrant from the part being inspected. Method A has emulsifiers built into the penetrant liquid that makes it possible to remove the excess penetrant with a simple water wash. Method B and D penetrants require an additional processing step where a separate emulsification agent is applied to make the excess penetrant more removable with a water wash. Lipophilic emulsification systems are oil-based materials that are supplied in readyto-use form. Hydrophilic systems are water-based and supplied as a concentrate that must be diluted with water prior to use. Lipophilic emulsifiers (Method B) were introduced in the late 1950's and work with both a chemical and mechanical action. After the emulsifier has coated the surface of the object, mechanical action starts to remove some of the excess penetrant as the mixture drains from the part. During the emulsification time, the emulsifier diffuses into the remaining penetrant and the resulting mixture is easily removed with a water spray. Hydrophilic emulsifiers (Method D) also remove the excess penetrant with mechanical and chemical action but the action is different because no diffusion takes place. Hydrophilic emulsifiers are basically detergents that contain solvents and surfactants. The hydrophilic emulsifier breaks up the penetrant into small quantities and prevents
7-105 7-105/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
these pieces from recombining or reattaching to the surface of the part. The mechanical action of the rinse water removes the displaced penetrant from the part and causes fresh remover to contact and lift newly exposed penetrant from the surface.
The hydrophilic post emulsifiable method (Method D) was introduced in the mid 1970's and since it is more sensitive than the lipophilic post emulsifiable method it has made the later method virtually obsolete. The major advantage of hydrophilic emulsifiers is that they are less sensitive to variation in the contact and removal time. While emulsification time should be controlled as closely as possible, a variation of one minute or more in the contact time will have little effect on flaw detectability when a hydrophilic emulsifier is used. However, a variation of as little as 15 to 30 seconds can have a significant effect when a lipophilic system is used.
Emulsifier.
Developers
The role of the developer is to pull the trapped penetrant material out of defects and to spread the developer out on the surface of the part so it can be seen by an inspector. The fine developer particles both reflect and refract the incident ultraviolet light, allowing more of it to interact with the penetrant, causing more efficient fluorescence. The developer also allows more light to be emitted through the same mechanism. This is why indications are brighter than the penetrant itself under UV light. Another function that some developers performs is to create a white background so there is a greater degree of contrast between the indication and the surrounding background.
7-106 7-106/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Developer Forms The AMS 2644 and Mil-I-25135 classify developers into six standard forms. These forms are listed below: 1. 2. 3. 4. 5. 6. Form a - Dry Powder Form b - Water Soluble Form c - Water Suspendible Form d - Nonaqueous Type 1 Fluorescent (Solvent Based) Form e - Nonaqueous Type 2 Visible Dye (Solvent Based) Form f - Special Applications
The developer classifications are based on the method that the developer is applied. The developer can be applied as a dry powder, or dissolved or suspended in a liquid carrier. Each of the developer forms has advantages and disadvantages.
Dry Powder Dry powder developer is generally considered to be the least sensitive but it is inexpensive to use and easy to apply. Dry developers are white, fluffy powders that can be applied to a thoroughly dry surface in a number of ways. The developer can be applied by dipping parts in a container of developer, or by using a puffer to dust parts with the developer. Parts can also be placed in a dust cabinet where the developer is blown around and allowed to settle on the part. Electrostatic powder spray guns are also available to apply the developer. The goal is to allow the developer to come in contact with the whole inspection area. Unless the part is electrostatically charged, the powder will only adhere to areas where trapped penetrant has wet the surface of the part. The penetrant will try to wet the surface of the penetrant particle and fill the voids between the particles, which brings more penetrant to the surface of the part where it can be seen. Since dry powder developers only stick to the part where penetrant is present, the dry developer does not provide a uniform white background as the other forms of developers do. Having a uniform light background is very important for a visible inspection to be effective and since dry developers do not provide one, they are seldom used for visible inspections. When a dry developer is used, 7-107 7-107/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
indications tend to stay bright and sharp since the penetrant has a limited amount of room to spread. Water Soluble As the name implies, water soluble developers consist of a group of chemicals that are dissolved in water and form a developer layer when the water is evaporated away. The best method for applying water soluble developers is by spraying it on the part. The part can be wet or dry. Dipping, pouring, or brushing the solution on to the surface is sometimes used but these methods are less desirable. Aqueous developers contain wetting agents that cause the solution to function much like dilute hydrophilic emulsifier and can lead to additional removal of entrapped penetrant. Drying is achieved by placing the wet but well drained part in a recalculating warm air dryer with the temperature held between 70 and 75F. If the parts are not dried quickly, the indications will will be blurred and indistinct. Properly developed parts will have an even, pale white coating over the entire surface. Water Suspendible Water suspendible developers consist of insoluble developer particles suspended in water. Water suspendible developers require frequent stirring or agitation to keep the particles from settling out of suspension. Water suspendible developers are applied to parts in the same manner as water soluble developers. Parts coated with a water suspendible developer must be forced dried just as parts coated with a water soluble developer are forced dried. The surface of a part coated with a water suspendible developer will have a slightly translucent white coating.
Nonaqueous Nonaqueous developers suspend the developer in a volatile solvent and are typically applied with a spray gun. Nonaqueous developers are commonly distributed in aerosol spray cans for portability. The solvent tends to pull penetrant from the indications by solvent action. Since the solvent is highly volatile, forced drying is not required. A nonaqueous developer should be applied to a thoroughly dried part to form a slightly translucent white coating.
7-108 7-108/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Special Applications Plastic or lacquer developers are a special developers that are primarily used when a permanent record of the inspection is required.
7.25.10 Preparation of Part One of the most critical steps in the penetrant inspection process is preparing the part for inspection. All coatings, such as paints, varnishes, plating, and heavy oxides must be removed to ensure that defects are open the surface of the part. If the parts have been machined, sanded, or blasted prior to the penetrant inspection, it is possible that a thin layer of metal may have smeared across the surface and closed off defects. It is even possible for metal smearing to occur as a result of cleaning operations such as grit or vapor blasting. This layer of metal smearing must be removed before inspection. Contaminants Coatings, such as paint, are much more elastic than metal and will not fracture even though a large defect may be present just below the coating. The part must be thoroughly cleaned as surface contaminates can prevent the penetrant from entering a defect. Surface contaminants can also lead to a higher level of background noise since the excess penetrant may be more difficult to remove.
Common coatings and contaminates that must be removed include: paint, dirt, flux, scale, varnish, oil, etchant, smut, plating, grease, oxide, wax, decals, machining fluid, rust, and residue from previous penetrant inspections. Some of these contaminants would obviously prevent penetrant from entering defects and it is, therefore, clear that they must be removed. However, the impact of other contaminants such as the residue from previous penetrant inspections is less clear, but they can have a disastrous affect on the inspection. Take the link below to review some of the research that has been done to evaluate the effects of contaminants on LPI sensitivity. A good cleaning procedure will remove all contamination from the part and not leave any residue that may interfere with the inspection process. It has been found that some alkaline cleaners can be detrimental to the penetrant inspection process if they have silicates in concentrations above 0.5 percent. Sodium metasilicate, sodium silicate, and related compounds can adhere to the surface of parts and form a coating that prevents penetrant entry into cracks. Researchers in Russia have also found that some domestic soaps and commercial detergents can clog flaw cavities and reduce the wettability of the metal surface, thus, reducing the sensitivity of the pene-
7-109 7-109/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
trant. Conrad and Caudill found that media from plastic media blasting was partially responsible for loss of LPI indication strength. Microphotographs of cracks after plastic media blasting showed media entrapment in addition to metal smearing.
Figure 7-78:
It is very important that the material being inspected has not been smeared across its own surface during machining or cleaning operations. It is well recognized that machining, honing, lapping, hand sanding, hand scraping, shot peening, grit blasting, tumble deburring, and peening operations can cause a small amount of the material to smear on the surface of some materials. It is perhaps less recognized that some cleaning operations, such as steam cleaning, can also cause metal smearing in the softer materials. Take the link below to learn more about metal smearing and its affects on LPI.
7.25.11 Selection of a Penetrant Technique The selection of a dye penetrant system is not a straightforward task. There are a variety of penetrant systems and developer types that are available for use, and one set of penetrant materials will not work for all applications. Many factors must be considered when selecting the penetrant materials for a particular application. These factors include the sensitivity required, materials cost, number of parts and size of area requiring inspection, and portability.
7-110 7-110/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-79:
Penetrant System.
When sensitivity is the primary consideration for choosing a penetrant system, the first decision that must be made is whether to use fluorescent dye penetrant, or visible dye penetrant. Fluorescent penetrants are generally more capable of producing a detectable indication from a small defect because the human eye is more sensitive to a light indication on a dark background and the eye is naturally drawn to a fluorescent indication. The graph below presents a series of curves that show the contrast ratio required for a spot of a certain diameter to be seen. The curves in Figure 7-80 show that for indications spots larger than 0.076 mm (0.003 inch) in diameter, it does not really matter if it is a dark spot on a light background or a light spot on a dark background. However, when a dark indication on a light background is further reduced in size, it is no longer detectable even though contrast is increased. Furthermore, with a light indication on a dark background, indications down to 0.003 mm (0.0001 inch) were detectable when the contrast between the flaw and the background was high enough. From this data, it can be seen why a fluorescent penetrant offers an advantage over visible penetrant for finding very small defects. Data presented by De Graaf and De Rijk supports this statement. They inspected "Identical" fatigue cracked specimens using a red dye penetrant and a fluorescent dye penetrant. The fluorescent penetrant found 60 defects while the visible dye was only able to find 39 of the defects.
7-111 7-111/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-80:
Under certain conditions, the visible penetrant may be a better choice. When fairly large defects are the subject of the inspection, a high sensitivity system may not be warranted and may result in a large number of irrelevant indications. Visible dye penetrants have also been found to give better results when surface roughness is high or when flaws are located in areas such as weldments. Since visible dye penetrants do not require a darkened area for the use of an ultraviolet light, visible systems are more easy to use in the field. Solvent removable penetrants, when properly applied can have the highest sensitivity and are very convenient to use but are usually not practical for large area inspection or in highvolume production settings. Another consideration in the selection of a penetrant system is whether water washable, post-emulsifiable or solvent removable penetrants will be used. Post-emulsifiable systems are designed to reduce the possibility of over-washing, which is one of the factors known to reduce sensitivity. However, these systems add another step, and thus cost, to the inspection process. Penetrants are evaluated by the US Air Force according to the requirements in MIL-I25135 and each penetrant system is classified into one of five sensitivity levels. This procedure uses titanium and Inconel specimens with small surface cracks produced
7-112 7-112/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
in low cycle fatigue bending to classify penetrant systems. The brightness of the indications produced after processing a set of specimens with a particular penetrant system is measured using a photometer. A procedure for producing and evaluating the penetrant qualification specimens was reported on by Moore and Larson at the 1997 ASNT Fall Conference. Most commercially available penetrant materials are listed in the Qualified Products List of MIL-I-25135 according to their type, method and sensitivity level. Visible dye and dual-purpose penetrants are not classified into sensitivity levels as fluorescent penetrants are. The sensitivity of a visible dye penetrant is regarded as level 1 and largely dependent on obtaining good contrast between the indication and the background. 7.25.12 Penetrant Application and Dwell Time The penetrant material can be applied in a number of different ways which include spraying, brushing, or immersing the parts in a penetrant bath. The method of penetrant application has little effect on the inspection sensitivity but an electrostatic spraying method is reported to produce slightly better results than other methods. Once the part is covered in penetrant it must be allowed to dwell so the penetrant has time to enter any defect present.
Figure 7-81:
Drain-dwell.
There are basically two dwell mode options, immersion-dwell (keeping the part immersed in the penetrant during the dwell period) or drain-dwell (letting the part drain during the dwell period). Prior to a study by Sherwin, the immersion-dwell mode was generally considered to be more sensitive but recognized to be less economical because more penetrant was washed away and emulsifiers were contaminated more rapidly. The reasoning for thinking this method was more sensitive was that the penetrant was more migratory and more likely to fill flaws when kept com-
7-113 7-113/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
pletely fluid and not allowed to loose volatile constituents by evaporation. However, Sherwin showed that if the specimens are allowed to drain-dwell, the sensitivity is higher because the evaporation increases the dyestuff concentration of the penetrant on the specimen. As pointed-out earlier, sensitivity increases as the dyestuff concentration increases. Sherwin also cautions that the samples being inspected should be placed outside the penetrant tank wall so that vapors from the tank do not accumulate and dilute the dyestuff concentration of the penetrant on the specimen.
Figure 7-82:
Immersion-dwell.
Penetrant Dwell Time Penetrant dwell time is the total time that the penetrant is in contact with the part surface. The dwell time is important because it allows the penetrant the time necessary to be drawn or to seep into a defect. Dwell times are usually recommended by the penetrant producers or required by the specification being followed. The time required to fill a flaw depends on a number of variables which include the following:
The surface tension of the penetrant. The contact angle of the penetrant. The dynamic shear viscosity of the penetrant, which can vary with the diameter of the capillary. The viscosity of a penetrant in microcapillary flaws is higher than its viscosity in bulk, which slows the infiltration of the tight flaws. The atmospheric pressure at the flaw opening. The capillary pressure at the flaw opening. The pressure of the gas trapped in the flaw by the penetrant. The radius of the flaw or the distance between the flaw walls.
7-114 7-114/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The density or specific gravity of the penetrant. Microstructural properties of the penetrant.
The ideal dwell time is often determined by experimentation and is often very specific to a particular application. For example, AMS 2647A requires that the dwell time for all aircraft and engine be at least 20 minutes while the ASTM E1209 only requires a 5 minute dwell time for parts made of titanium and other heat resistant alloys. Generally, there is no harm in using a longer penetrant dwell time as long as the penetrant is not allowed to dry.
7.25.13 Penetrant Removal Process The penetrant removal procedure must effectively remove the penetrant from the surface of the part without removing an appreciable amount of entrapped penetrant from the defect. If the removal process extracts penetrant from the flaw, the flaw indication will be reduced by a proportional amount. If the penetrant is not effectively removed from the part surface, the contrast between the indication and the background will be reduced. As discussed in Contrast Sensitivity Section, as the contrast increases so does visibility of the indication.
Figure 7-83:
7-115 7-115/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Removal Method Penetrant systems are classified into four methods of excess penetrant removal. These include the following: 1. 2. 3. 4. Method A: Water-Washable Method B: Post Emulsifiable, Lipophilic Method C: Solvent Removable Method D: Post Emulsifiable, Hydrophilic
Method C, Solvent Removable, is used primarily for inspecting small localized areas as this method requires hand wiping the surface with a cloth moistened with the solvent remover, and this process is too labor intensive for most production situations. Of the three production penetrant inspection methods, Method A, Water-Washable, is the most economical to apply. Water-washable or self-emulsifiable penetrants contain an emulsifier as an integral part of the formulation. The excess penetrant may be removed from the object surface, with a simple water rinse. These materials have the property of forming relatively viscous gels upon contact with water, which results in the formation of gel-like plugs in surface openings. While they are completely soluble in water, given enough contact time, the plugs offer a brief period of protection against rapid wash removal. Thus, water-washable penetrant systems provide ease of use and a high level of sensitivity. When removal of the penetrant from the defect due to over-washing of the part is a concern, a post-emulsifiable penetrant system can be used. Post-emulsifiable penetrants require a separate emulsifier to break the penetrant down and make it water washable. The emulsifier is usually applied by dipping the object. Hydrophilic emulsifiers may also be sprayed on the object but spraying is not recommended for lipophilic emulsifiers because it can result in non-uniform emulsification if not properly applied. Brushing the emulsifier on to the part is not recommended because the bristles of the brush may force emulsifier into discontinuities causing the entrapped penetrant to be removed. The emulsifier is allowed sufficient time to react with the penetrant on the surface of the part but not given time to make its way into defects to react with the trapped penetrant. The penetrant that has reacted with the emulsifier is easily cleaned away. Controlling the reaction time is of essential importance when using a post-emulsifiable system. If the emulsification time is too short, an excessive amount of penetrant will be left on the surface leading to high background levels. If the emulsification time is too long, the emulsifier will react with the penetrant entrapped in discontinuities making it possible to deplete the amount needed to form an indication.
7-116 7-116/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-84:
The hydrophilic post emulsifiable method (Method D) is more sensitive than the lipophilic post emulsifiable method (Method B). Since these methods are generally only used when very high sensitivity is needed, Method D is almost always used making Method B virtually obsolete. The major advantage of hydrophilic emulsifiers is that they are less sensitive to variation in the contact and removal time. While emulsification time should be controlled as closely as possible, a variation of one minute or more in the contact time will have little effect on flaw detectability when a hydrophilic emulsifier is used, but a variation of as little as 15 to 30 seconds can have a significant effect when a lipophilic system is used. Using an emulsifier involves adding a couple of steps to the penetrant process and ,therefore, slightly increases the cost of an inspection. When using an emulsifier, the penetrant process includes the following steps (extra steps in bold): 1. pre-clean part, 2. apply penetrant and allow to dwell, 3. pre-rinse to remove first layer of penetrant, 4. apply hydrophilic emulsifier,and allow contact for specified time, 5. rinse to remove excess penetrant, 6. dry part, 7. apply developer and allow part to develop, and 8. inspect. (see some of the steps below).
Penetrant
To the left: Part with penetrant. To the right: Part with penetrant
7-117 7-117/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Vandafskylning Water-Washing
Figure 7-86:
To the left: The emulsifier has emulsified the pene-trant on the parts surface, but not the penetrant in the discontinuitie. To the right: Excess penetrant and emulsifier on the parts surface are being removed with water.
Rinse Method and Time for Water-Washable Penetrants The method used to rinsing the excess penetrant from the object surface and the time of the rinse should be controlled so as to prevent overwashing. It is generally recommended that a coarse spray rinse or an air agitated, immersion wash tank be used. When a spray is being used, it should be directed at a 45 angle to the part surface so as to not force water directly into any discontinuities that may be present. The spray or immersion time should be kept to a minimum through frequent inspections of the remaining background level.
Hand Wiping of Solvent Removable Penetrants When a solvent removable penetrant is used, care must also be taken to carefully remove the penetrant from the part surface while removing as little as possible from the flaw. The first step in this cleaning procedure is to dry wipe the surface of the part in one direction using a white lint free cotton rag. One dry pass in one direction is all that should be used to remove as much penetrant as possible. Next, the surface should be wiped with one pass in one direction with a cleaner moistened rag. One dry pass followed by one damp pass is all that is recommended. Additional wiping may sometimes be necessary but keep in mind that with every additional wipe, some of the entrapped
7-118 7-118/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
penetrant will be removed and inspection sensitivity will be reduced. To study the effects of the wiping process, Japanese researchers manufactured a test specimen out of acrylic plates that allowed them to view the movement of the penetrant in a narrow cavity. The sample consisted of two pieces of acrylic with two thin sheets of vinyl clamped between as spaces. The plates were clamped in the corners and all but one of the edges sealed. The unsealed edge acted as the flaw. The clearance between the plates varied from 15 microns (0.059 inch) at the clamping points to 30 microns (0.118 inch) at the midpoint between the clamps. The distance between the clamping points is believed to be 30 mm (1.18 inch).
Although the size of the flaw represented by this specimen is large, an interesting observation was made. They found that when the surface of the specimen was wiped with a dry cloth, penetrant was blotted and removed from the flaw at the corner areas were the clearance between the plate was least. When the penetrant at the side areas was removed, penetrant moved horizontally from the center area to the ends of the simulated crack where capillary forces are stronger. Therefore, across the crack length, the penetrant surface has a parabola-like shape where the liquid is at the surface in the corners but depressed in the center. This shows that each time the cleaning cloth touches the edge of a crack, penetrant is lost from the defect. This also explains why the bleedout of an indication is often largest at the corners of cracks. 7.25.14 Use and Selection of a Developer The use of developer is almost always recommended. One study reported that the output from a fluorescent penetrant could be multiplied by up to seven times when a suitable powder developer was used. Another study showed that the use of developer can have a dramatic effect on the probability of detection (POD) of an inspection. When a Haynes Alloy 188, flat panel specimen with a low-cycle fatigue crack was inspected without a developer, a 90 % POD was never reached with crack lengths as long as 19 mm (0.75 inch). The operator detected only 86 of 284 cracks and had 70 false-calls. When a developer was used, a 90 % POD was reached at 2 mm (0.077 inch), with the inspector identifying 277 of 311 cracks with no false-calls. However some authors have reported that in special situations the use of a developer may actually reduce sensitivity. These situations primarily occur when large, well defined defects are being inspected on a surface that contains many nonrelevant indications that cause excessive bleedout.
7-119 7-119/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-87:
Type of Developer Used and Method of Application Nonaqueous developers are generally recognized as the most sensitive when properly applied. There is less agreement on the performance of dry and aqueous wet developers but the aqueous developers are usually considered more sensitive. Aqueous wet developers form a finer matrix of particles that is more in contact with the part surface. However, if the thickness of the coating becomes too great, defects can be masked. Also aqueous wet developers can cause leaching and blurring of indications when used with water washable penetrants. The relative sensitivities of developers and application techniques as ranked in Volume II of the Nondestructive Testing Handbook are shown in the table below. There is general industry agreement with this table, but some industry experts feel that water suspendible developers are more sensitive than water-soluble developers. Sensitivity ranking of developers per the Nondestructive Testing Handbook.
7-120 7-120/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Developer Form Nonaqueous Wet Solvent Plastic Film water-soluble Water Suspendible water-soluble Water Suspendible Dry Dry Dry Dry
Method of Application Spray Spray Spray Spray Immersion Immersion Dust Cloud (Electrostatic) Fluidized Bed Dust Cloud (Air Agitation) Immersion (Dip)
Table 7-4 lists the main advantages and disadvantages of the various developer types.
7-121 7-121/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Developer
Advantages
Disadvantages
Does not form contrast background so cannot be used with visible systems Difficult to assure entire part surface has been coated Coating is translucent and provides poor contrast (not recommended for visual systems) Indications for water washable systems are dim and blurred
Dry
Indications tend to remain brighter and more distinct over time Easily to apply
Ease of coating entire part Soluble White coating for good contrast can be produced which work well for both visible and fluorescent systems
Ease of coating entire part Indications are bright and sharp Suspendible White coating for good contrast can be produced which work well for both visible and fluorescent systems Very portable Easy to apply to readily accessible surfaces White coating for good contrast can be produced which work well for both visible and fluorescent systems Indications show-up rapidly and are well defined Provides highest sensitivity Difficult to apply evenly to all surfaces More difficult to clean part after inspection
Nonaqueous
7-122 7-122/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.25.15 Quality Control Temperature The temperature of the penetrant materials and the part being inspected can have an effect on the results. Temperatures from 27 to 49oC (80 to 120oF) are reported in the literature to produce optimal results. Many specifications allow testing in the range of 4 to 52oC (40 to 125oF). A tip to remember is that surfaces that can be touched for an extended period of time without burning the skin are generally below 52oC (125oF). Since the surface tension of most materials decrease as the temperature increases, raising the temperature of the penetrant will increase the wetting of the surface and the capillary forces. Of course, the converse is also true and lowering the temperature will have a negative effect on the flow characteristics. Raising the temperature will also raise the speed of evaporation of penetrants, which can have a positive or negative effect on sensitivity. The impact will be positive if the evaporation serves to increase the dye concentration of the penetrant trapped in a flaw up to the concentration quenching point and not beyond. Higher temperatures and more rapid evaporation will have a negative effect if the dye concentration is caused to exceed the concentration quenching point or the flow characteristics are changed to the point where the penetrant does not readily flow. The method of processing a hot part was once commonly employed. Parts were either heated or processed hot off the production line. In its days, this served to increase inspection sensitivity by increasing the viscosity of the penetrant. However, the penetrant materials used today have 1/2 to 1/3 the viscosity of the penetrants on the market in the 1960's and 1970's. Heating the part prior to inspection is no longer necessary and no longer recommended. Penetrant The quality of a penetrant inspection is highly dependent on the quality of the penetrant materials used. Only products meeting the requirements of an industry specification, such as AMS 2644, should be used. The performance of a penetrant can be affected by contamination and aging. Contamination by another liquid will change the surface tension and contact angle of the solution, and virtually all organic dyes deteriorate over time resulting in a loss of color or fluorescent response. Therefore, regular checks must be performed to insure that the material performance has not degraded. When the penetrant is first received from the manufacturer, a sample of the fresh solution should be collected and stored as a standard for future comparison. The standard specimen should be stored in an opaque glass or metal, sealed container. Penetrants that are in-use should be compared regularly against the standard specimen to detect changes in color, odor and consistency. When using fluorescent penetrants, a brightness comparison per the requirements of ASTM E 1417 is also often required. This check involves placing a drop of the standard and the in-use pene-
7-123 7-123/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
trants on a piece of Whatman #4 filter paper and making a side by side comparison of the brightness of the two spots under UV light. Additionally, the water content of water washable penetrants must be checked regularly. When water contaminates oil-based penetrants, the surface tension and contact angle of the mixture will increase since water has a higher surface tension than most oil-based penetrants In self-emulsifiable penetrants, water contamination can produce a gel break or emulsion inversion when the water concentration becomes high enough. The formation of the gel is an important feature during the washing processes but must be avoided until the stage in the process. Data indicates that the water contamination must be significant (greater than 10%) for gel formation to occur. Most specification limit water contamination to around 5% to be conservative. Non-water-based, water washable penetrants are checked using the procedure specified in ASTM D95 or ASTM E 1417. Water-based, water washable penetrants are checked with a refractometer. The rejection criteria is different for different penetrants so the requirements of the qualifying specification or the manufacturer's instructions must be consulted. Application of the Penetrant The application of the penetrant is the step of the process that requires the least amount of control. As long as the surface being inspected receives a generous coating of penetrant, it really doesn't matter how the penetrant is applied. Generally, the application method is an economic or convenience decision. It is important that the part be thoroughly cleaned and dried. Any contaminates or moisture on the surface of the part or within a flaw can prevent the penetrant material from entering the defect. The part should also be cool to the touch. The recommended range of temperature is 4 to 52oC (39 to 125 F). Wash Temperature and Pressure The wash temperature and pressure and time are three parameters that are typically controlled in penetrant inspection process specification. A coarse spray or an immersion wash tank with air agitation is often used. When the spray method is used, the water pressure is usually limited to 276 kN/m2 (40 psi). The temperature range of the water is usually specified as a wide range (ex. 10 to 38C (50 to 100 F) in AMS 2647A.) A low-pressure, coarse water spray will force less water into flaws to dilute and/or remove trapped penetrant and weaken the indication. The temperature will have an effect on the surface tension of the water and warmer water will have more wetting action than cold water. Warmer water temperatures may also make emulsifiers and detergent more effective. The wash time should only be as long as necessary to decrease the background to an acceptable level. Frequent visual checks of the part should be made to determine when the part has be adequately rinsed.
7-124 7-124/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Drying Process The temperature used to dry parts after the application of aqueous wet developer or prior to the application of a dry powder or a nonaqueous wet developer, must be controlled to prevent "cooking" of the penetrant in the defect. High drying temperature can affect penetrants in a couple of ways. First, some penetrants can fade at high temperatures due to dye vaporization or sublimation. Second, high temperatures can cause the penetrant to dry in the the flaw preventing it from migrating to the surface to produce an indication. To prevent harming the penetrant material, drying temperature should be kept to under 71 degree centigrade. The drying should be limited to the minimum length of time necessary to thoroughly dry the component being inspected.
Developer The function of the developer is very important in a penetrant inspection. It must draw out of the discontinuity a sufficient amount of penetrant to form an indication, and it must spread the penetrant out on the surface to produce a visible indication. In a fluorescent penetrant inspection, the amount of penetrant brought to the surface must exceed the dye's thin film threshold of fluorescence of the indication will not fluoresce. Additionally, the developer makes fluorescent indications appear brighter than indications produced with the same amount of dye but without the developer. In order to accomplish these functions, a developer must adhere to the part surface and result in a uniform, highly porous layer with many paths for the penetrant to be moved due to capillary action. Some developers are applied wet and other dry, but the desired end result is always a uniform, highly porous, surface layer. Since the quality control requirements for each of the developer types is slightly different, they will be covered individually. Dry Powder Developer A dry powder developer should be checked daily to ensure that it is fluffy and not caked. It should be similar to fresh powdered sugar and not granulated like powered soup. It should also be relatively free from specks of fluorescent penetrant material from previous inspection. This check is performed by spreading a sample of the developer out and examining it under UV light. If there are ten or more fluorescent specks in an 10 cm diameter area, the batch should be discarded. Apply a light coat of the developer by immersing the test component or dusting the surface. After the development time, excessive powder can be removed by gently blowing on the surface with air not exceeding 35 kPa or 5 psi.
Wet Soluble/Suspendible Developer Wet soluble developer must be completely dissolved in the water and wet suspendible developer must be thoroughly mixed prior to application. The concentration of powder in the carrier solution must be controlled in these developers. The concentra-
7-125 7-125/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
tion should be checked at least weekly using a hydrometer to make sure it meets the manufacturer's specification. To check for contamination, the solution should be examined weekly using both white light and UV light. If a scum is present or the solution fluoresces, it should be replaced. Some specification require that a clean aluminum panel be dipped in the developer, dried, and examined for indications of contamination by fluorescent penetrant materials. These developers are applied immediately after the final wash. A uniform coating should be applied by spraying, flowing or immersion of the component. They should never be applied with a brush. Care should be taken to avoid a heavy accumulation of the developer solution in crevices and recesses. Prolonged contact of the component with the developer solution should be avoided in order to minimize dilution or removal of the penetrant from discontinuities. Solvent Suspendible Solvent suspendible developers are typically supplied in an sealed aerosol spray can. Since the developer solution is in a sealed vessel, direct check of the solution are not possible. However, the way that the developer is dispensed must be monitored. The spray developer should produce a fine, even coating on the surface of the part. Make sure the can is well shaken and apply a thin coating to a test article. If the spray produces spatters or other an uneven coating the can should be discarded. When applying a solvent suspendible developer, it is up to the inspector to control the thickness of the coating. When using a visible penetrant system, the developer coating must be thick enough to provide a white contrasting background but not heavy enough to mask indications. When using a fluorescent penetrant system, a very light coating should be used. The developer should be applied under white light condition and should appear evenly transparent. Development Time Part should be allowed to develop for a minimum of 10 minutes and no more than 2 hours before inspecting. Lighting After a component has been properly processed, it is ready for inspection. While automated vision inspection systems are sometimes used, the focus here will be on inspection performed visually by a human inspector as this is the dominate method. Proper lighting is of great importance when visually inspecting a surface for a penetrant indication. Obviously, the lighting requirements are different for an inspection conducted using a visible dye penetrant than they are for an inspection conducted using a fluorescent dye penetrant. The lighting requirements for each of these techniques, as well as how light measurements are made, is discussed below.
7-126 7-126/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-88:
UV-lamp.
Lighting for Visible Dye Penetrant Inspections When using a visible penetrant, the intensity of the white light is of principal importance. Inspections can be conducted using natural lighting or artificial lighting. When using natural lighting, it is important to keep in mind that daylight varies from hour to hour so inspector must stay constantly aware on the lighting conditions and make adjustment when needed. To improve uniformity in lighting from one inspection to the next, the use of artificial lighting is recommended. Artificial lighting should be white whenever possible and white flood or halogen lamps are most commonly used. The light intensity is required to be 100 foot-candles at the surface being inspected. It is advisable to choose a white light wattage that will provide sufficient light, but avoid excessive reflected light that could distract from the inspection.
Lighting for Fluorescent Penetrant Inspections Then a fluorescent penetrant is being employed, the ultraviolet illumination and the visible light inside the inspection booth is important. Penetrant dyes are excited by the UV of 365-nm wavelength and emit visible light somewhere in the green-yellow range between 520 and 580 nm. The source of ultraviolet light (UV) is often a mercury arc lamp with a filter. The lamps emit many wavelengths and a filter is used to remove all but the UV and a small amount of visible light between 320 and 410 nm. Visible light of wavelengths above 410 nm interferes with contrast, and UV emissions below 320 nm include some hazardous wavelengths. Standards and procedures require verification of lens condition and light intensity. Black lights should never be used with a cracked filter as output of white light and harmful black light will be increased. The cleanliness of the filter should also be checked as a coating of solvent carrier, oils, or other foreign materials can reduce the intensity by up to as much as 50%. The filter should be checked visually and cleaned as necessary before warm-up of the light. 7-127 7-127/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Since fluorescent brightness is linear with respect to ultraviolet excitation, a change in the intensity of the light (from age or damage) and a change in the distance of the light source from the surface being inspected will have a direct impact on the inspection. For UV lights used in component evaluations, the normally accepted intensity is 1000 microwatts per square centimeter when measured at 15 inches from the filter face (requirements can vary from 800 to 1200 W/cm2). The required check should be performed when a new bulb is installed, at startup of the inspection cycle, if a change in intensity is noticed, or every eight hours of continuous use. Regularly checking the intensity of UV lights is very important because bulbs loose intensity over time. In fact, a bulb that is near the end of its operating life will often have an intensity of only 25 percent of its original output. A maximum intensity of 5000 w/cm2 is recommended at the inspection area. The reason is that the fluorescent part of the penetrant is sensitive to UV-light in fact, the UV-light can make the flourescens fade away. Consequencely, a UV-lamp should never be placed directly above the tank containing fluorescent penetrant liquid. Black light intensity will also be affected by voltage variations. A bulb that produces acceptable intensity at 120 volts will produce significantly less at 110 volts. For this reason it is important to provide constant voltage to the light. Also, most UV light must be warmed up prior to use and should be on for at least 15 minutes before beginning an inspection. When performing a fluorescent penetrant inspection, it is important to keep white light to a minimum as it will significantly reduce the inspectors ability to detect fluorescent indications. Light levels of less than 2 fc are required by most procedures with some procedures requiring less than 0.5 fc at the inspection surface. Procedures require a check and documentation of ambient white light in the inspection area. When checking black light intensity at 15 inches a reading of the white light produced by the black light may be required to verify white light is being removed by the filter. Light Measurement Light intensity measurements are made using a radiometer. A radiometer is an instrument that translate light energy into an electrical current. Light striking a silicon photodiode detector causes a charge to build up between internal layers. When an external circuit is connected to the cell, an electrical current is produced. This current is linear with respect to incident light. Some radiometers have the ability to measure both black and white light, while others require a separate sensor for each measurement. Whichever type is used, the sensing area should be clean and free of any materials that could reduce or obstruct light reaching the sensor. Radiometers are relatively unstable instruments and readings often change considerable over time. Therefore, they should be calibrated at least every six months. Ultraviolet light measurements should be taken using a fixture to maintain a minimum distance of 15 inches from the filter face to the sensor. The sensor should be centered in the light field to obtain and record the highest reading. UV spot lights are often focused, so intensity readings will vary considerable over a small area. White lights are seldom
7-128 7-128/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
focused and depending on the wattage, will often produce in excess of the 100 fc at 15 inches. Many specifications do not require the white light intensity check to be conducted at a specific distance. 7.25.16 System Performance Check System performance checks involve processing a test specimen with known defects to determine if the process will reveal discontinuities of the size required. The specimen must be processed following the same procedure used to process production parts. A system performance check is typically required daily, at the reactivation of a system after maintenance or repairs, or any time the system is suspected of being out of control. As with penetrant inspections in general, results are directly dependent on the skill of the operator and, therefore, each operator should process a panel. The ideal specimen is a production item that has natural defects of the minimum acceptable size. Some specification delineate the type and size of the defects that must be present in the specimen and detected. Surface finish is will affect washability so the check specimen should have the same surface finish as the production parts being processed. If penetrant systems with different sensitivity levels are being used, there should be a separate specimen for each system. There are some universal test specimens that can be used if a standard part is not available. The most commonly used test specimen is the TAM or PSM panel. These panel are usually made of stainless steel that has been chrome plated on one half and surfaced finished on the other half to produced the desired roughness. The chrome plated section is impacted from the back side to produce a starburst set of cracks in the chrome. There are five impacted areas to produce range of crack sizes. Each panel has a characteristic signature and variances in that signature are indications of process variance. Panel patterns as well as brightness are indicators of process consistency or variance, see picture above. Care of system performance check specimens is critical. Specimens should be handled carefully to avoid damage. They should be cleaned thoroughly between uses and storage in a solvent is generally recommended. Before processing a specimen, it should be inspected under UV light to make sure that it is clean and not already producing an indication.
7-129 7-129/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.25.17 Nature of the Defect The nature of the defect can have a large affect on sensitivity of a dye penetrant inspection. Sensitivity is defined as the smallest defect that can be detected with a high degree of reliability. Typically, the crack length at the sample surface is used to define size of the defect. A survey of any probability-of-detection curve for penetrant inspection will quickly lead one to the conclusion that crack length has a definite affect on sensitivity. However, the crack length alone does not determine whether a flaw will be seen or go undetected. The volume of the defect is likely to be the more important feature. The flaw must be of sufficient volume so that enough penetrant will bleed back out to a size that is detectable by the eye or that will satisfy the dimensional thresholds of fluorescence.
Figure 7-89:
tion (POD) curve from the Nondestructive Evaluation (NDE) Capabilities Data Book. Above is an example of fluorescent penetrant inspection probability of detection (POD) curve from the Nondestructive Evaluation (NDE) Capabilities Data Book. Please note that this curve is specific to one set of inspection conditions and should not be interpreted to apply to other inspection situations. 7-130 7-130/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
small round defects than small linear defects. Small round defects are generally easier to detect for several reasons. First, they are typically volumetric defects that can trap significant amounts of penetrant. Second, round defects fill with penetrant faster than linear defects. One research effort found that elliptical flaw with length to width ratio of 100, will take the penetrant nearly 10 times longer to fill than a cylindrical flaw with the same volume. deeper flaws than shallow flaws. Deeper flaws will trap more penetrant than shallow flaws, and they are less prone to over washing. flaws with a narrow opening at the surface than wide open flaws. Flaws with narrow surface openings are less prone to over washing. flaws on smooth surfaces than on rough surfaces. The surface roughness of the part primarily affects the removability of a penetrant. Rough surfaces tend to trap more penetrant in the various tool marks, scratches, and
Removing the penetrant from the surface of the part is more difficult and a higher level of background fluorescence or over washing may occur.
pits that make up the surface.
flaws with rough fracture surfaces than smooth fracture surfaces. The surface roughness that the fracture faces is a factor in the speed at which a penetrant enters a defect. In general, the penetrant spreads faster over a surface as the surface roughness increases. It should be noted that a particular penetrant may spread slower than others on a smooth surface but faster than the rest on a rougher surface. flaws under tensile or no loading than flaws under compression loading. In a 1987 study at the University College London, the effect of crack closure on detectability was evaluated. Researchers used a four-point bend fixture to place tension and compression loads on specimens that were fabricated to contain fatigue cracks. All cracks were detected with no load and with tensile loads placed on the parts. However, as compressive loads were placed on the parts, the crack length steadily decreased as load increased until a load was reached when the crack was no longer detectable.
7.25.18 Health & Safety Precautions in Dye Penetrant Inspection When proper health and safety precautions are followed, dye penetrant inspection operations can be completed without harm to inspection personnel. However, there are a number of health and safety related issues that must be addressed. Since each inspection operation will have its own unique set of health and safety concerns that must be addressed, only a few of the most common concerns will be discussed here.
7-131 7-131/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Chemical Safety Whenever chemicals must be handled, certain precautions must be taken as directed by the material safety data sheets (MSDS) for the chemicals. Before working with a chemical of any kind, it is highly recommended that the MSDS be reviewed so that proper chemical safety and hygiene practices can be followed. Some of the penetrant materials are flammable and, therefore, should be used and stored in small quantities. They should only be used in a well ventilated area and ignition sources avoided. Eye protection should always be worn to prevent contact of the chemicals with the eyes. Many of the chemicals used contain detergents and solvents that can dermatitis. Gloves and other protective clothing should be worn to limit contact with the chemicals. Ultraviolet Light Safety Ultraviolet (UV) light or "black light" as it is sometimes called, has wavelengths ranging from 180 to 400 nanometers. These wavelengths place UV light in the invisible part of the electromagnetic spectrum between visible light and X-rays. The most familiar source of UV radiation is the the sun and is necessary in small doses for certain chemical processes to occur in the body. However, too much exposure can be harmful to the skin and eyes. Excessive UV light exposure can cause painful sunburn, accelerate wrinkling and increase the risk of skin cancer. UV light can cause eye inflammation, cataracts, and retinal damage. Because of their close proximity, laboratory devices, like UV lamps, deliver UV light at a much higher intensity than the sun and, therefore, can cause injury much more quickly. The greatest threat with UV light exposure is that the individual is generally unaware that the damage is occurring. There is usually no pain associated with the injury until several hours after the exposure. Skin and eye damage occurs at wavelengths around 320 nm and shorter which is well below the 365 nm wavelength, where penetrants are designed to fluoresce. Therefore, UV lamps sold for use in LPI application are almost always filtered to remove the harmful UV wavelengths. The lamps produce radiation at the harmful wavelengths so it is essential that they be used with the proper filter in place and in good condition.
7-132 7-132/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.25.19 References and Resources Cartz, Louis, Nondestructive Testing, ASM Intl, Metals Park, OH, 1995, ISBN: 0871705176 Introduction to Capillary Testing Theory, Borovikov, A.S. (Ed.), Minsk, Nauka i Tekhnika Publishing, 1988 Liquid Penetrant Testing, Nondestructive Testing Handbook, Volume 2, Tracy, Noel (Tech. Ed.), Moore, Patrick (Ed.) American Society for Nondestructive Testing, Columbus, OH, 1999, ISBN 1-57117-028-6 Larson, B.F., Study of the Factors Affecting the Sensitivity of Liquid Penetrant Inspections: Review of Literature Published from 1970 to 1998, FAA Technical Report Number DOT/FAA/AR-01/95, Office of Aviation Research, Washington, DC, Jan 2002 (pdf 1.0 meg) Flaherty, J. J., History of Penetrants: The First 20 Years, 1941-61, Materials Evaluation, Vol. 44, No. 12, November 1986, pp. 1371-1374, 1376, 1378, 1380, 1382 Boisvert, B.W., Hardy, G., Dorgan, J.F., and Selner, R.H., The Fluorescent Penetrant Hydrophilic Remover Process, Materials Evaluation, February 1983, pp. 134-137. Sherwin, A. G., Overremoval Propensities of the Prewash Hydrophilic Emulsifier Fluorescent Penetrant Process, Materials Evaluation, March 1993, pp. 294-299. Robinson, Sam J., Here Today, Gone Tomorrow! Replacing Methyl Chloroform in the Penetrant Process, Materials Evaluation, Vol. 50, No. 8, August 1992, pp. 936-946. Rummel, W., Cautions on the Use of Commercial Aqueous Precleaners for Penetrant Inspection, Materials Evaluation, Vol. 16, No. 5, August 1998, pp. 950-952. Glazkov, Y.A., Some Technological Mistakes in the Application of Capillary Inspection to Repairs of Gas Turbin Engines, translation from Defektoskopiya - The Soviet Journal of Nondestructive Testing, Vol. 26, No. 3, New York, NY Plenum/Consultants Bureau, January 1990, pp. 361-367. Glazkov, Yu . A., Bruevich, E.P., and Samokhin, N.L, Special Features of Application of Aqueous Solutions of Commercial Detergents in Capillary Flaw Inspection, Defektoskopiya - The Soviet Journal of Nondestructive Testing, Vol. 19, No. 8, August 1982, pp. 83-87. De Graaf, E. and De Rijk, P., Comparison Between Reliability, Sensitivity, and Accuracy of Nondestructive Inspection Methods, 13th Symposium on Nondestructive Evaluation Proceedings, San Antonio, TX, published by NTIAC, Southwest Research Institute, San Antonio, TX, April 1981, pp. 311-322. 7-133 7-133/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Thomas, W.E., An Analytic Approach to Penetrant Performance, 1963 Lester Honor Lecture, Nondestructive Testing, Vol. 21, No. 6, Nov.-Dec. 1963, pp. 354-368. Senda, T., Maeda, N., Kato, M., Ebata, M., Ooka, K., and Miyoshi, S., Factors Involved in Formation of Penetrant Testing Indications, NDE in the Nuclear Industry: Proceedings of the 6th International Conference, Zurich, Switzerland, November December 1984, pp. 807-810. Brittain, P. I., The Amplifying Action of Developer Powders, QUALTEST 3 Conference, Cincinnati OH, Oct 1984. Rummel, W. D., Probability of Detection as a Quantitative Measure of Nondestructive Testing End-To-End Process Capabilities, Materials Evaluation, January 1998, pp. 35. Nondestructive Testing Handbook, Vol. 2, Liquid Penetrant Tests, Robert McMaster, et al., American Society for Nondestructive Testing, 1982, pp. 283-319. Rummel, W.D. and Matzkanin, G. A., Nondestructive Evaluation (NDE) Capabilities Data Book, Published by the Nondestructive Testing Information Analysis Center (NTIAC), NTIAC #DB-95-02, May 1996. Alburger, J.R., Dimensional Transition Effects in Visible Color and Fluorescent Dye Liquids, Proceedings, 23rd Annual Conference, Instrument Society of America, Vol. 23, Part I, Paper No. 564. Deutsch, S. A, Preliminary Study of the Fluid Mechanics of Liquid Penetrant Testing, Journal of Research of the National Bureau of Standards, Vol. 84, No. 4, July-August 1979, pp. 287-291. Kauppinen, P. and Sillanpaa, J., Reliability of Surface Inspection Methods, Proceedings of the 12th World Conference on Nondestructive Testing, Amsterdam, Netherlands, Vol. 2, Elsevier Science Publishing, Amsterdam, 1989, pp. 1723-1728. Vaerman, J. F., Fluorescent Penetrant Inspection Process, Automatic Method for Sensitivity Quantification, Proceedings of 11th World Conference on Nondestructive Testing, Volume III, Las Vegas, NV, November 1985, pp. 1920-1927. Thomas, W.E., An Analytic Approach to Penetrant Performance, 1963 Lester Honor Lecture, Nondestructive Testing, Vol. 21, No. 6, Nov.-Dec. 1963, pp. 354-368. Clark, R., Dover, W.D., and Bond, L.J., The Effect of Crack Closure on the Reliability of NDT Predictions of Crack Size, NDT International, Vol. 20, No. 5, Guildford, United Kingdom, Butterworth Scientific Limited, October 1987, pp. 269-275.
7-134 7-134/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.26
7-135 7-135/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.26.2 Basic Principles In theory, magnetic particle inspection (MPI) is a relatively simple concept. It can be considered as a combination of two non-destructive testing methods: magnetic flux leakage testing and visual testing. Consider a bar magnet. It has a magnetic field in and around the magnet. Any place that a magnetic line of force exits or enters the magnet is called a pole. A pole where a magnetic line of force exits the magnet is called a north pole and a pole where a line of force enters the magnet is called a south pole. When a bar magnet is broken in the centre of its length, two complete bar magnets with magnetic poles on each end of each piece will result. If the magnet is just cracked but not broken completely in two, a north and south pole will form at each edge of the crack. The magnetic field exits the North Pole and renters at the South Pole. The magnetic field spreads out when it encounter the small air gap created by the crack because the air cannot support as much magnetic field per unit volume as the magnet can. When the field spreads out, it appears to leak out of the material and, thus, it is called a flux leakage field. If iron particles are sprinkled on a cracked magnet, the particles will be attracted to and cluster not only at the poles at the ends of the magnet but also at the poles at the edges of the crack. This cluster of particles is much easier to see than the actual crack and this is the basis for magnetic particle inspection.
Figure 7-91:
7-136 7-136/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The first step in a magnetic particle inspection is to magnetize the component that is to be inspected. If any defects on or near the surface are present, the defects will create a leakage field. After the component has been magnetized, iron particles, either in a dry or wet suspended form, are applied to the surface of the magnetized part. The particles will be attracted and cluster at the flux leakage fields, thus forming a visible indication that the inspector can detect. 7.26.3 History of Magnetic Particle Inspection Magnetism is the ability of matter to attract other matter to it. The ancient Greeks were the first to discover this phenomenon in a mineral they named magnetite. Later on Bergmann, Becquerel, and Faraday discovered that all matter including liquids and gasses were affected by magnetism, but only a few responded to a noticeable extent. The earliest known use of magnetism to inspect an object took place as early as 1868. Cannon barrels were checked for defects by magnetizing the barrel then sliding a magnetic compass along the barrel's length. These early inspectors were able to locate flaws in the barrels by monitoring the needle of the compass. This was a form of non-destructive testing but the term was not commonly used until some time after World War I. In the early 1920s, William Hoke realized that magnetic particles (colour metal shavings) could be used with magnetism as a means of locating defects. Hoke discovered that a surface or subsurface flaw in a magnetized material caused the magnetic field to distort and extend beyond the part. This discovery was brought to his attention in the machine shop. He noticed that the metallic grindings from hard steel parts, which were being held by a magnetic chuck while being ground, formed patterns on the face of the parts which corresponded to the cracks in the surface. Applying a fine ferromagnetic powder to the parts caused a build up of powder over flaws and formed a visible indication. The image shows a 1928 Electro-Magnetic Steel Testing Device (MPI) made by the Equipment and Engineering Company Ltd. (ECO) of Strand, England. In the early 1930s, magnetic particle inspection (MPI) was quickly replacing the oiland-whiting method (an early form of the liquid penetrant inspection) as the method of choice by the railroad to inspect steam engine boilers, wheels, axles, and the tracks. Today, the MPI inspection method is used extensively to check for flaws in a large variety of manufactured materials and components. MPI is used to check materials such as steel bar stock for seams and other flaws prior to investing machining
7-137 7-137/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
time during the manufacturing of a component. Critical automotive components are inspected for flaws after fabrication to ensure that defective parts are not placed into service. MPI is used to inspect some highly loaded components that have been inservice for a period of time. For example, many components of high performance race cars are inspected whenever the engine, drive train and other systems are overhauled. MPI is also used to evaluate the integrity of structural welds on bridges, storage tanks, and other safety critical structures. 7.26.4 Magnetism Magnets are very common items in the workplace and household. Uses of magnets range from holding pictures on the refrigerator to causing torque in electric motors. Most people are familiar with the general properties of magnets but are less familiar with the source of magnetism. The traditional concept of magnetism centres around the magnetic field and what is know as a dipole. The term "magnetic field" simply describes a volume of space where there is a change in energy within that volume. This change in energy can be detected and measured. The location where a magnetic field can be detected exiting or entering a material is called a magnetic pole. Magnetic poles have never been detected in isolation but always occur in pairs and, thus, the name dipole. Therefore, a dipole is an object that has a magnetic pole on one end and a second equal but opposite magnetic pole on the other. A bar magnet can be considered a dipole with a north pole at one end and South Pole at the other. A magnetic field can be measured leaving the dipole at the North Pole and returning the magnet at the South Pole. If a magnet is cut in two, two magnets or dipoles are created out of one. This sectioning and creation of dipoles can continue to the atomic level. Therefore, the source of magnetism lays in the basic building block of all matter...the atom.
The Source of Magnetism All matter is composed of atoms, and atoms are composed of protons, neutrons and electrons. The protons and neutrons are located in the atom's nucleus and the electrons are in constant motion around the nucleus. Electrons carry a negative electrical charge and produce a magnetic field as they move through space. A magnetic field is produced whenever an electrical charge is in motion. The strength of this field is called the magnetic moment. This may be hard to visualize on a subatomic scale but consider electric current flowing through a conductor. When the electrons (electric current) are flowing through the conductor, a magnetic field forms around the conductor. The magnetic field can be detected using a compass. The magnetic field will place a force on the compass needle, which is another example of a dipole. 7-138 7-138/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Since all matter is comprised of atoms, all materials are affected in some way by a magnetic field. However, not all materials react the same way. This will be explored more in the next section. 7.26.5 Magnetic Materials When a material is placed within a magnetic field, the magnetic forces of the material's electrons will be affected. This effect is known as Faraday's Law of Magnetic Induction. However, materials can react quite differently to the presence of an external magnetic field. This reaction is dependent on a number of factors such as the atomic and molecular structure of the material, and the net magnetic field associated with the atoms. The magnetic moments associated with atoms have three origins. These are the electron orbital motion, the change in orbital motion caused by an external magnetic field, and the spin of the electrons. In most atoms, electrons occur in pairs. Each electron in a pair spins in the opposite direction. So when electrons are paired together, their opposite spins cause there magnetic fields to cancel each other. Therefore, no net magnetic field exists. Alternately, materials with some unpaired electrons will have a net magnetic field and will react more to an external field. Most materials can be classified as ferromagnetic, diamagnetic or paramagnetic. Diamagnetic metals have a very weak and negative susceptibility to magnetic fields. Diamagnetic materials are slightly repelled by a magnetic field and the material does not retain the magnetic properties when the external field is removed. Diamagnetic materials are solids with all paired electron and, therefore, no permanent net magnetic moment per atom. Diamagnetic properties arise from the realignment of the electron orbits under the influence of an external magnetic field. Most elements in the periodic table, including copper, silver, and gold, are diamagnetic. Paramagnetic metals have a small and positive susceptibility to magnetic fields. These materials are slightly attracted by a magnetic field and the material does not retain the magnetic properties when the external field is removed. Paramagnetic properties are due to the presence of some unpaired electrons and from the realignment of the electron orbits caused by the external magnetic field. Paramagnetic materials include magnesium, molybdenum, lithium, and tantalum. Ferromagnetic materials have a large and positive susceptibility to an external magnetic field. They exhibit a strong attraction to magnetic fields and are able to retain their magnetic properties after the external field has been removed. Ferromagnetic materials have some unpaired electrons so their atoms have a net magnetic moment. They get their strong magnetic properties due to the presence of magnetic domains. In these domains, large numbers of atoms moments (1012 to 1015) are aligned parallel so that the magnetic force within the domain is strong. When a ferromagnetic material is in the unmagnified state, the domains are nearly randomly organized and the net magnetic field for the part as a whole is zero. When a magnetizing force is applied, the domains become aligned to produce a strong
7-139 7-139/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
magnetic field within the part. Iron, nickel, and cobalt are examples of ferromagnetic materials. Components with these materials are commonly inspected using the magnetic particle method. 7.26.6 Magnetic Domains Ferromagnetic materials get their magnetic properties not only because their atoms carry a magnetic moment but also because the material is made up of small regions known as magnetic domains. In each domain, all of the atomic dipoles are coupled together in a preferential direction. This alignment develops as the material develops its crystalline structure during solidification from the molten state. Magnetic domains can be detected using Magnetic Force Microscopy (MFM) and images of the domains like the one shown below can be constructed.
Magnetic Force Microscopy (MFM) image showing the magnetic domains in a piece of heat treated carbon steel.
Figure 7-92:
domains in a piece of heat treated carbon steel. During solidification a trillion or more atom moments are aligned parallel so that the magnetic force within the domain is strong in one direction. Ferromagnetic materials are said to be characterized by "spontaneous magnetization" since they obtain saturation magnetization in each of the domains without an external magnetic field being applied. Even though the domains are magnetically saturated, the bulk material may not show any signs of magnetism because the domains develop themselves are randomly oriented relative to each other. Ferromagnetic materials become magnetized when the magnetic domains within the material are aligned. This can be done by placing the material in a strong external magnetic field or by passing electrical current through the material. Some or all of the domains can become aligned. The more domains that are aligned, the stronger the magnetic field in the material. When all of the domains are aligned, the material is said to be magnetically
saturated. When a
material is magnetically saturated, no additional amount of external magnetization force will cause an increase in its internal level of magnetization. An example of these domains is shown in Figure 7-93. 7-140 7-140/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Magnetized Material
7.26.7
Magnetic Field In and Around a Bar Magnet As discussed previously a magnetic field is a change in energy within a volume of space. The magnetic field surrounding a bar magnet can be seen in the magnetograph below. A magnetograph can be created by placing a piece of paper over a magnet and sprinkling the paper with iron filings. The particles align themselves with the lines of magnetic force produced by the magnet. The magnetic lines of force show where the magnetic field exits the material at one pole and reenters the material at another pole along the length of the magnet. It should be noted that the magnetic lines of force exist in three-dimensions but are only seen in two dimensions in the image.
It can be seen in the magnetograph that there are poles all along the length of the magnet but that the poles are concentrated at the ends of the magnet. The area where the exit poles are concentrated is called the magnet's North Pole and the area where the entrance poles are concentrated is called the magnet's South Pole.
7-141 7-141/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Magnetic Fields in and around Horseshoe and Ring Magnets Magnets come in a variety of shapes and one of the more common is the horseshoe (U) magnet. The horseshoe magnet has north and south poles just like a bar magnet but the magnet is curved so the poles lie in the same plane. The magnetic lines of force flow from pole to pole just like in the bar magnet. However, since the poles are located closer together and a more direct path exists for the lines of flux to travel between the poles, the magnetic field is concentrated between the poles. If a bar magnet was placed across the end of a horseshoe magnet or if a magnet was formed in the shape of a ring, the lines of magnetic force would not even need to enter the air. The value of such a magnet where the magnetic field is completely contained with the material probably has limited use. However, it is important to understand that the magnetic field can flow in loop within a material when the concept of circular magnetism is later covered.
General Properties of Magnetic Lines of Force Magnetic lines of force have a number of important properties, which include:
They seek the path of least resistance between opposite magnetic poles. In a single bar magnet as shown to the right, they attempt to form closed loop from pole to pole. They never cross one another. They all have the same strength. Their density decreases (they spread out) when they move from an area of higher permeability to an area of lower permeability. Their density decreases with increasing distance from the poles. They are considered to have direction as if flowing, though no actual movement occurs. They flow from the South Pole to the North Pole within the material and North Pole to South Pole in air.
7-142 7-142/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.26.8 Electromagnetic Fields Magnets are not the only source of magnetic fields. In 1820, Hans Christian Oersted discovered that an electric current flowing through a wire caused a nearby compass to deflect. This indicated that the current in the wire was generating a magnetic field. Oersted studied the nature of the magnetic field around the long straight wire. He found that the magnetic field existed in circular form around the wire and that the intensity of the field was directly proportional to the amount of current carried by the wire. He also found that the strength of the field was strongest close to the wire and diminished with distance from the conductor until it could no longer be detected. In most conductors, the magnetic field exists only as long as the current is flowing (i.e. an electrical charge is in motion). However, in ferromagnetic materials the electric current will cause some or all of the magnetic domains to align and a residual magnetic field will remain. Oersted also noticed that the direction of the magnetic field was dependent on the direction of the electrical current in the wire. A three-dimensional representation of the magnetic field is shown below. There is a simple rule for remembering the direction of the magnetic field around a conductor. It is called the right-hand rule. If a person grasps a conductor in ones right hand with the thumb pointing in the direction of the current, the fingers will circle the conductor in the direction of the magnetic field.
7-143 7-143/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
A word of caution about the right-hand rule For the right-hand rule to work, one important thing that must be remembered about the direction of current flow. Standard convention has current flowing from the positive terminal to the negative terminal. This convention is credited to Benjamin Franklin who theorized that electric current was due to a positive charge moving from the positive terminal to the negative terminal. However, it was later discovered that it is the movement of the negatively charged electron that is responsible for electrical current. Rather than changing several centuries of theory and equations, Franklin's convention is still used today.
7.26.9 Magnetic Field Produced by a Coil When a current carrying conductor is formed into a loop or several loops to form a coil, a magnetic field develops that flows through the centre of the loop or coil along the longitudinal axis and circles back around the outside of the loop or coil. The magnetic field circling each loop of wire combines with the fields from the other loops to produce a concentrated field down the centre of the coil. A loosely wound coil is illustrated below to show the interaction of the magnetic field. The magnetic field is essentially uniform down the length of the coil when it is wound tighter.
7-144 7-144/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The strength of a coil's magnetic field increases not only with increasing current but also with each loop that is added to the coil. A long straight coil of wire is called a solenoid and can be used to generate a nearly uniform magnetic field similar to that of a bar magnet. The concentrated magnetic field inside a coil is very useful in magnetizing ferromagnetic materials for inspection using the magnetic particle testing method. Please be aware that the field outside the coil is weak and is not suitable for magnetize ferromagnetic materials.
7.26.10 Quantifying Magnetic Properties (Magnetic Field Strength, Flux Density, Total Flux and Magnetization) Until now, only the qualitative features of the magnetic field have been discussed. However, it is necessary to be able to measure and express quantitatively the various characteristics of magnetism. Unfortunately, a number of unit conventions are in use as shown below. SI units will be used in this material. The advantage of using SI units is that they are traceable back to an agreed set of four base units - meter, kilogram, second, and Ampere.
The units for magnetic field strength H are ampere/meter. A magnetic field strength of 1 ampere/meter is produced at the center of a single circular conductor of diameter 1 meter carrying a steady current of 1 ampere. The number of magnetic lines of force cutting through a plane of a given area at a right angle is known as the magnetic flux density B. The flux density or magnetic induction has the tesla as its unit. One tesla is equal to 1 Newton/ (A/m). From these units it can be seen that the flux density is a measure of the force applied to a particle by the magnetic field. The Gauss is CGS unit for flux density and is commonly used by US industry. One gauss represents one line of flux passing through one square centimeter of air oriented 90 degrees to flux flow. The total number of lines of magnetic force in a material is called magnetic flux. The strength of the flux is determined by the number of magnetic domains that are aligned within a material. The total flux is simply the flux density applied over an area. Flux carries the unit of a weber, which is simply a tesla-square meter.
7-145 7-145/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The magnetization is a measure of the extent to which an object is magnetized. It is a measure of the magnetic dipole moment per unit volume of the object. Magnetization carries the same units as a magnetic field; amperes/meter.
7.26.11 The Hysteresis Loop and Magnetic Properties A great deal of information can be learned about the magnetic properties of a material by studying its hysteresis loop. A hysteresis loop shows the relationship between the induced magnetic flux density B and the magnetizing force H. It is often referred to as the B-H loop. An example hysteresis loop is shown below.
7-146 7-146/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The loop is generated by measuring the magnetic flux B of a ferromagnetic material while the magnetizing force H is changed. A ferromagnetic material that has never been previously magnetized or has been thoroughly demagnetized will follow the dashed line as H is increased. As the line demonstrates, the greater the amount of current applied (H+), the stronger the magnetic field in the component (B+). At point "a" almost all of the magnetic domains are aligned and an additional increase in the magnetizing force will produce very little increase in magnetic flux. The material has reached the point of magnetic saturation. When H is reduced back down to zero, the curve will move from point "a" to point "b." At this point, it can be seen that some magnetic flux remains in the material even though the magnetizing force is zero. This is referred to as the point of retentivity on the graph and indicates the remanence or level of residual magnetism in the material. (Some of the magnetic domains remain aligned but some have lost there alignment.) As the magnetizing force is reversed, the curve moves to point "c", where the flux has been reduced to zero. This is called the point of coercivity on the curve. (The reversed magnetizing force has flipped enough of the domains so that the net flux within the material is zero.) The force required to remove the residual magnetism from the material, is called the coercive force or coercivity of the material. As the magnetizing force is increased in the negative direction, the material will again become magnetically saturated but in the opposite direction (point "d"). Reducing H to zero brings the curve to point "e." It will have a level of residual magnetism equal to that achieved in the other direction. Increasing H back in the positive direction will return B to zero. Notice that the curve did not return to the origin of the graph because some force is required to remove the residual magnetism. The curve will take a different path from point "f" back the saturation point where it with complete the loop. From the hysteresis loop, a number of primary magnetic properties of a material can be determined. 1. Retentivity - A measure of the residual flux density corresponding to the saturation induction of a magnetic material. In other words, it is a material's ability to retain a certain amount of residual magnetic field when the magnetizing force is removed after achieving saturation. (The value of B at point B on the hysteresis curve.) Residual Magnetism or Residual Flux - the magnetic flux density that remains in a material when the magnetizing force is zero. Note that residual magnetism and retentivity are the same when the material has been magnetized to the saturation point. However, the level of residual magnetism may be lower than the retentivity value when the magnetizing force did not reach the saturation level. Coercive Force - The amount of reverse magnetic field which must be applied to a magnetic material to make the magnetic flux return to zero. (The value of H at point C on the hysteresis curve.)
2.
3.
7-147 7-147/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
4.
Permeability - A property of a material that describes the ease with which a magnetic flux is established in the component. Reluctance - Is the opposition that a ferromagnetic material shows to the establishment of a magnetic field. Reluctance is analogous to the resistance in an electrical circuit.
5.
7.26.12 Permeability As previously mentioned, permeability is a material property that describes the ease with which a magnetic flux is established in the component. It is the ratio of the flux density to the magnetizing force and, therefore, represented by the following equation:
= /
It is clear that this equation describes the slope of the curve at any point on the hysteresis loop. The permeability value given in papers and reference materials is usually the maximum permeability or the maximum relative permeability. The maximum permeability is the point where the slope of the B/H curve for unmagnetized material is the greatest. This point is often taken as the point where a straight line from the origin is tangent to the B/H curve. The relative permeability is arrived at by taking the ratio of the material's permeability to the permeability in free space (air).
where:
7-148 7-148/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The shape of the hysteresis loop tells a great deal about the material being magnetized. The hysteresis curves of two different materials are shown in the graph. Relative to the other material, the materials with the wide hysteresis loop has:
Lower Permeability Higher Retentivity Higher Coercivity Higher Reluctance Higher Residual Magnetism
Higher Permeability Lower Retentivity Lower Coercivity Lower Reluctance Lower Residual Magnetism.
In magnetic particle testing the level of residual magnetism is important. Residual magnetic fields are affected by the permeability, which can be related to the carbon content and alloying of the material. A component with high carbon content will have low permeability and will retain more magnetic flux than a material with low carbon content.
7.26.13 Magnetic Field Orientation and Flaw Detectability To properly inspect a component for cracks or other defects, it is important to understand that orientation between the magnetic lines of force and the flaw is very important. There are two general types of magnetic fields that can be established within a component.
7-149 7-149/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
A longitudinal magnetic field has magnetic lines of force that run parallel to the long axis of the part. Longitudinal magnetization of a component can be accomplished using the longitudinal field set up by a coil or solenoid. It can also be accomplished using permanent or electromagnets. A circular magnetic field has magnetic lines of force that run circumferentially around the perimeter of a part. A circular magnetic field is induced in an article by either passing current through the component or by passing current through a conductor surrounded by the component.
The type of magnetic field established is determined by the method used to magnetize the specimen. Being able to magnetize the part in two directions is important because the best detection of defects occurs when the lines of magnetic force are established at right angles to the longest dimension of the defect. This orientation creates the largest disruption of the magnetic field within the part and the greatest flux leakage at the surface of the part. As can be seen in the image below, if the magnetic field is parallel to the defect, the field will see little disruption and no flux leakage field will be produced.
An orientation of 45 to 90 degrees between the magnetic field and the defect is necessary to form an indication. Since defects may occur in various and unknown direc-
7-150 7-150/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
tions, each part is normally magnetized in two directions at right angles to each other. If the component below is considered, it is known that passing current through the part from end to end will establish a circular magnetic field that will be 90 degrees to the direction of the current. Therefore, defects that have a significant dimension in the direction of the current (longitudinal defects) should be detectable. Alternately, transverse-type defects will not be detectable with circular magnetization.
7.26.14 Magnetization of Ferromagnetic Materials There are a variety of methods that can be used to establish a magnetic field in a component for evaluation using magnetic particle inspection. It is common to classify the magnetizing methods as either direct or indirect.
Magnetization Using Direct Induction (Direct Magnetization) With direct magnetization, current is passed directly through the component. Recall that whenever current flows a magnetic field is produced. Using the right-hand rule, which was introduced earlier, it is known that the magnetic lines of flux form normal to the direction of the current and form a circular field in and around the conductor. When using the direct magnetization method, care must be taken to ensure that good electrical contact is established and maintained between the test equipment and the test component. Improper contact can result in arcing that may damage the component. It is also possible to overheat components in areas of high resistance such as the contact points and in areas of small cross-sectional area. There are several ways that direct magnetization is commonly accomplished. One way involves clamping the component between two electrical contacts in a special 7-151 7-151/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
piece of equipment. Current is passed through the component and a circular magnetic field is established in and around the component. When the magnetizing current is stopped, a residual magnetic field will remain within the component. The strength of the induced magnetic field is proportional to the amount of current passed through the component. A second technique involves using clamps or prods, which are attached or placed in contact with the component. Electrical current flows through the component from contact to contact. The current sets up a circular magnetic field around the path of the current.
Magnetization Using Indirect Induction (Indirect Magnetization) Indirect magnetization is accomplished by using a strong external magnetic field to establish a magnetic field within the component. As with direct magnetization, there are several ways that indirect magnetization can be accomplished. The use of permanent magnets is a low cost method of establishing a magnetic field. However, their use is limited due to lack of control of the field strength and the difficulty of placing and removing strong permanent magnets from the component.
Electromagnets in the form of an adjustable horseshoe magnet (called a yoke) eliminate the problems associated with permanent magnets and are used extensively in industry. Electromagnets only exhibit a magnetic flux when electric current is flowing around the soft iron core. When the magnet is placed on the component, a magnetic field is established between the north and south poles of the magnet. Another way of indirectly inducting a magnetic field in a material is by using the magnetic field of a current carrying conductor. A circular magnetic field can be established in cylindrical components by using a central conductor. Typically, one or more cylindrical components are hung from a solid copper bar running through the inside diameter. Current is passed through the copper bar and the resulting circular magnetic field establishes a magnetic field within the test components.
7-152 7-152/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The use of coils and solenoids is a third method of indirect magnetization. When the length of a component is several times larger than its diameter, a longitudinal magnetic field can be established in the component. The component is placed longitudinally in the concentrated magnetic field that fills the centre of a coil or solenoid. This magnetization technique is often referred to as a "coil shot."
7.26.15 Magnetizing Current As seen in the previous pages, electric current is often used to establish the magnetic field in components during magnetic particle inspection. Alternating current and direct current are the two basic types of current commonly used. Current from single phase 110 volts, to three phase 440 volts are used when generating an electric field in a component. Current flow is often modified to provide the appropriate field within the part. The type of current used can have an effect on the inspection results so the types of currents commonly used will be briefly reviewed. Direct Current Direct current (DC) flows continuously in one direction at a constant voltage. A battery is the most common source of direct current. As previously mentioned, current is said to flow from the positive to the negative terminal when in actuality the electrons flow in the opposite direction. DC is very desirable when performing magnetic particle inspection in search of subsurface defects because DC generates a magnetic field that penetrates deeper into the material. In ferromagnetic materials, the magnetic field produced by DC generally penetrates the entire cross-section of the component; whereas, the field produced using alternating current is concentrated in a thin layer at the surface of the component. Alternating Current Alternating current (AC) reverses in direction at a rate of 50 or 60 cycles per second. In the United States, 60 cycles current are the commercial norm but 50 cycles current are common in many countries. Since AC is readily available in most facilities, it is convenient to make use of it for magnetic particle inspection. However, when AC is used to induce a magnetic field in ferromagnetic materials the magnetic field will be limited to narrow region at the surface of the component. This phenomenon is known as "skin effect" and it occurs because induction is not a spontaneous reaction and the rapidly reversing current does not allow the domains down in the material time to align. Therefore, it is recommended that AC be used only when the inspection is limited to surface defects.
7-153 7-153/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Rectified Alternating Current Clearly, the skin effect limits the use of AC since many inspection applications call for the detection of subsurface defects. However, the convenient access to AC, drive its use beyond surface flaw inspections. Luckily, AC can be converted to current that is very much like DC through the process of rectification. With the use of rectifiers, the reversing AC can be converted to a one-directional current. The three commonly used types of rectified current are described below.
Half Wave Rectified Alternating Current (HWAC) When single phase alternating current is passed through a rectifier, current is allowed to flow in only one direction. The reverse half of each cycle is blocked out so that a one directional, pulsating current is produced. The current rises from zero to a maximum and then returns to zero. No current flows during the time when the reverse cycle is blocked out. The HWAC repeats at same rate as the unrectified current (50 or 60 hertz typical). Since half of the current is blocked out, the amperage is half of the unaltered AC. This type of current is often referred to as half wave DC or pulsating DC. The pulsation of the HWAC helps magnetic particle indications form by vibrating the particles and giving them added mobility. This added mobility is especially important when 7-154 7-154/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
using dry particles. The pulsation is reported to significantly improve inspection sensitivity. HWAC is most often used to power electromagnetic yokes. Full Wave Rectified Alternating Current (FWAC) (Single Phase) Full wave rectification inverts the negative current to positive current rather than blocking it out. This produces a pulsating DC with no interval between the pulses. Filtering is usually performed to soften the sharp polarity switching in the rectified current. While particle mobility is not as good as half-wave AC due to the reduction in pulsation, the depth of the subsurface magnetic field is improved. Three Phase Full Wave Rectified Alternating Current Three phase current is often used to power industrial equipment because it has more favourable power transmission and line loading characteristics. This type of electrical current is also highly desirable for magnetic particle testing because when it is rectified and filtered, the resulting current very closely resembles direct current. Stationary magnetic particle equipment wired with three phase AC will usually have the ability to magnetize with AC or DC (three phase full wave rectified), providing the inspector with the advantages of each current form. 7.26.16 Longitudinal Magnetic Fields, Distribution and Intensity When the length of a component is several times larger than its diameter, a longitudinal magnetic field can be established in the component. The component is often placed longitudinally in the concentrated magnetic field that fills the centre of a coil or solenoid. This magnetization technique is often referred to as a "coil shot." The magnetic field travels through the component from end to end with some flux loss along its length as shown in the image to the right. Keep in mind that the magnetic lines of flux occur in three dimensions and are only shown in 2D in the image. The magnetic lines of flux are much denser inside the ferromagnetic material than in air because ferromagnetic materials have much higher permeability than air. When the concentrated flux within the material comes to the air at the end of the component, it must spread out since the air can not support as many lines of flux per unit volume. To keep from crossing as they spread out, some of the magnetic lines of flux are forced out the side of the component. When a component is magnetized along its complete length, the flux loss is small along its length. Therefore, when a component is uniform in cross section and magnetic permeability, the flux density will be relatively uniform throughout the component. Flaws that run normal to the magnetic lines of flux will disturb the flux lines and often cause a leakage field at the surface of the component.
7-155 7-155/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
When a component with considerable length is magnetized using a solenoid, it is possible to magnetize only a portion of the component. Only the material within the solenoid and about the same width on each side of the solenoid will be strongly magnetized. At some distance from the solenoid, the magnetic lines of force will abandon their longitudinal direction, leave the part at a pole on one side of the solenoid and return to the part at an opposite pole on the other side of the solenoid. This occurs because the magnetizing force diminishes with increasing distance from the solenoid, and, therefore, the magnetizing force may only be strong enough to align the magnetic domains within and very near the solenoid. The unmagnetized portion of the component will not support as much magnetic flux as the magnetized portion and some of the flux will be forced out of the part as illustrated in the image below. Therefore, a long component must be magnetized and inspected at several locations along its length for complete inspection coverage.
Solenoid - An electrically energized coil of insulated wire, which produces a magnetic field within the coil.
7-156 7-156/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.26.17 Circular Magnetic Fields, Distribution and Intensity As discussed previously, when current is passed through a solid conductor, a magnetic field forms in and around the conductor. The following statements can be made about the distribution and intensity of the magnetic field.
The field strength varies from zero at the center of the component to a maximum at the surface. The field strength at the surface of the conductor decreases as the radius of the conductor increases when the current strength is held constant. (However, a larger conductor is capable of carrying more current.) The field strength outside the conductor is directly proportional to the current strength. Inside the conductor the field strength is dependent on the current strength, magnetic permeability of the material, and if magnetic, the locations on the B-H curve. The field strength outside the conductor decreases with distance from the conductor.
In the images below, the magnetic field strength is graphed versus distance from the center of the conductor. It can be seen that in a nonmagnetic conductor carrying DC, the internal field strength rises from zero at the center to a maximum value at the surface of the conductor. The external field strength decrease with distance from the surface of the conductor. When the conductor is a magnetic material, the field strength within the conductor is much greater than it was in the nonmagnetic conductor. This is due to the permeability of the magnetic material. The external field is exactly the same for the two materials provided the current level and conductor radius are the same.
7-157 7-157/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The magnetic field distribution in and around a solid conductor of a nonmagnetic material carrying direct current.
The magnetic field distribution in and around a solid conductor of a magnetic material carrying direct current.
When the conductor is carrying alternating current, the internal magnetic field strength rises from zero at the center to a maximum at the surface. However, the field is concentrated in a thin layer near the surface of the conductor. This is known as the "skin effect." The skin effect is evident in the field strength versus distance graph for a magnet conductor shown to the right. The external field decreases with increasing distance from the surface as it does with DC. It should be remembered that with AC the field is constantly varying in strength and direction.
The magnetic field distribution in and around a solid conductor of a magnetic material carrying alternating current.
7-158 7-158/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
In a hollow circular conductor there is no magnetic field in the void area. The magnetic field is zero at the inside wall surface and rises until it reaches a maximum at the outside wall surface. As with a solid conductor, when the conductor is a magnetic material, the field strength within the conductor is much greater than it was in the nonmagnetic conductor due to the permeability of the magnetic material. The external field strength decrease with distance from the surface of the conductor. The external field is exactly the same for the two materials provided the current level and conductor radius are the same.
The magnetic field distribution in and The magnetic field distribution in and around a hollow conductor of a nonmag- around a hollow conductor of a magnetic netic material carrying direct current. material carrying direct current.
7-159 7-159/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
When AC is passed through a hollow circular conductor the skin effect concentrates the magnetic field at the outside diameter of the component. As can be seen in the field distribution images, the field strength at the inside surface of hollow conductor carrying a circular magnetic field produced by direct magnetization is very low. Therefore, the direct method of magnetization is not recommended when inspecting the inside diameter wall of a hollow component for shallow defects. The field strength increases rather rapidly as one moves in from the ID so if the defect has significant depth, it may be detectable.
The magnetic field distribution in and around a hollow conductor of a magnetic material carrying alternating current.
However, a much better method of magnetizing hollow components for inspection of the ID and OD surfaces is with the use of a central conductor. As can be seen in the field distribution image to the right, when current is passed through a nonmagnetic central conductor (copper bar) the magnetic field produced on the inside diameter surface of a magnetic tube is much greater and the field is still strong enough for defect detection on the OD surface.
The magnetic field distribution in and around a nonmagnetic central conductor carrying DC inside a hollow conductor of a magnetic material.
7-160 7-160/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.26.18 Demagnetization After conducting a magnetic particle inspection, it is usually necessary to demagnetize the component. Remanent magnetic fields can:
affect machining by causing cuttings to cling to a component. interfere with electronic equipment such as a compass. create a condition known as "arc blow" in the welding process. Arc blow may cause the weld arc to wonder or filler metal to be repelled from the weld. cause abrasive particle to cling to bearing or faying surfaces and increase wear.
Removal of a field may be accomplished in several ways. This random orientation of the magnetic domains can be achieved most effectively by heating the material above its curie temperature. The Curie temperature for low carbon steel is 770 degrees C or 1390 degrees F. When steel is heated above its curie temperature, it will become austenitic and loose its magnetic properties. When it is cooled back down it will go through a reverse transformation and will contain no residual magnetic field. The material should also be placed with it long axis in an east-west orientation to avoid any influence of the Earth's magnetic field. It is often inconvenient to heat a material above it Curie temperature to demagnetize it so another method that returns the material to a nearly unmagnetized state is commonly used. Subjecting the component to a reversing and decreasing magnetic field will return the dipoles to a nearly randomly oriented throughout the material. This can be accomplished by pulling a component out and away from a coil with AC passing through it. The same can also be accomplished using an electromagnetic yoke with AC selected. Also, many stationary magnetic particle inspection units come with a demagnetization feature that slowly reduces the AC in a coil in which the component is placed. A field meter is often used to verify that the residual flux has been removed from a component. Industry standards usually require that the magnetic flux be reduced to less than 3 gauss after completing a magnetic particle inspection.
7-161 7-161/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.26.19 Measuring Magnetic Fields When performing a magnetic particle inspection, it is very important to be able to determine the direction and intensity of the magnetic field. As discussed previously, the direction of the magnetic field should be between 45 and 90 degrees to the longest dimension of the flaw for best detectability. The field intensity must be high enough to cause an indication to form, but not too high or nonrelevant indications may form that could mask relevant indications. To cause an indication to form, the field strength in the object must produce a flux leakage field that is strong enough to hold the magnetic particles in place over a discontinuity. Flux measurement devices can provide important information about the field strength. Since it is impractical to measure the actual field strength within the material, all the devices measure the magnetic field that is outside of the material. There are a number of different devices that can be used to detect and measure an external magnetic field. The two devices commonly used in magnetic particle inspection are the field indicator and the Hall Effect meter, which is also often called a gauss meter. Pie gages and shims are devices that are often used to provide an indication of the field direction and strength but do not actually yield a quantitative measure. They will be discussed in a later section.
Field Indicators Field indicators are small mechanical devices that utilize a soft iron vane that will be deflected by a magnetic field. The X-ray image below shows the inside working of a field meter looking in from the side. The vane is attached to a needle that rotates and moves the pointer for the scale. Field indicators can be adjusted and calibrated so that quantitative information can be obtained. However, the measurement range of field indicators is usually small due to the mechanics of the device. The one shown to the right has a range from plus twenty gauss to minus twenty gauss. This limited range makes them best suited for measuring the residual magnetic field after demagnetization.
7-162 7-162/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Hall-Effect (Gauss/Tesla) Meter A Hall-effect meter is an electronic device that provides a digital readout of the magnetic field strength in gauss or tesla units. The meters use a very small conductive or semiconductor element at the tip of the probe. Electric current is passed through the conductor. In a magnetic field, the magnetic field exerts a force on the moving electrons which tends to push them to one side of the conductor. A build-up of charge at the sides of the conductors will balance this magnetic influence, producing a measurable voltage between the two sides of the conductor. The presence of this measurable transverse voltage is called the Hall-effect after Edwin H. Hall who discovered it in 1879.
The voltage generated Vh can be related to the external magnetic field by the following equation. Vh = I B Rh / b Where: Vh is the voltage generated. I is the applied direct current. B is the component of the magnetic field that is at a right angle to the direct current in the Hall element. Rh is the Hall Coefficient of the Hall element. b is the thickness of the Hall element. Probes are available with either tangential (transverse) or axial sensing elements. Probes can be purchased in a wide variety of sizes and configurations and with different measurement ranges. The probe is placed in the magnetic field such that the magnetic lines of force intersect the major dimensions of the sensing element at a right angle. Placement and orientation of the probe is very important and will be discussed in a later section.
7-163 7-163/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.26.20 Portable Magnetizing Equipment To properly inspect a part for cracks or other defects, it is important to become familiar with the different types of magnetic fields and the equipment used to generate them. As discussed previously, one of the primary requirements for detection of a defect in a ferromagnetic material is that the magnetic field induced in the part must intercept the defect at a 45 to 90 degrees angle. Flaws that are normal (90 degrees) to the magnetic field will produce the strongest indications because they disrupt more of the magnet flux.
Therefore, for proper inspection of a component, it is important to be able to establish a magnetic field in at least two directions. A variety of equipment exists to establish the magnetic field for MPI. One way to classify equipment is based on its portability. Some equipment is designed to be portable so that inspections can be made in the field and some is designed to be stationary for ease of inspection in the laboratory or manufacturing facility. Portable equipment will be discussed first. Permanent magnets Permanent magnets are sometimes used for magnetic particle inspection as the source of magnetism. The two primary types of permanent magnets are bar magnets and horseshoe (yoke) magnets. These industrial magnets are usually very strong and may require significant strength to remove them from a piece of metal. Some permanent magnets require over 50 pounds of force to remove them from the surface. Because it is difficult to remove the magnets from the component being inspected, and sometimes difficult and dangerous to place the magnets, their use is not particularly popular. However, permanent magnets are sometimes used by a diver for inspection in an underwater environment or other areas, such as in an explosive environment, where electromagnets cannot be used. Permanent magnets can also be made small enough to fit into tight areas where electromagnets might not fit.
7-164 7-164/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Electromagnets Today, most of the equipment used to create the magnetic field used in MPI are based on electromagnetism. It uses an electrical current to produce the magnetic field. An electromagnetic yoke is a very common piece of equipment that is used to establish a magnetic field. It is basically made by wrapping an electrical coil around a piece of soft ferromagnetic steel. A switch is included in the electrical circuit so that the current and, therefore, also the magnetic field can be turn on and off. They can be powered with alternating current from a wall socket or by direct current from a battery pack. This type of magnet generates a very strong magnetic field in a local area where the poles of magnet touch the part to be inspected. Some yokes can lift weights in excess of 40 pounds.
Figure 7-94:
To the left: Portable yoke with battery pack. To the right: Portable
7-165 7-165/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Prods Prods are handheld electrodes that are pressed against the surface of the component being inspected to make contact for passing electrical current through the metal. The current passing between the prods creates a circular magnetic field around the prods that is can be used in magnetic particle inspection. Prods are typically made from copper and have an insulated handle to help protect the operator. One of the prods has a trigger switch so that the current can be quickly and easily turned on and off. Sometimes the two prods are connected by an insulator as shown in the image to facilitate one hand operation. This is referred to as a dual prod and is commonly used for weld inspections. If proper contact is not maintained between the prods and the component surface, electrical arcing can occur and cause damage to the component. For this reason, the uses of prods are not allowed when inspecting aerospace and other critical components. To help to prevent arcing, the prod tips should be inspected frequently to ensure that they are not oxidized, covered with scale or other contaminant, or damaged. The figure below shows two prods used to create a current through a conducting part. The resultant magnetic field roughly depicted gives an estimation of the patterns expected with magnetic particle on an unflawed surface.
Portable Coils and Conductive Cables Coils and conductive cables are used to establish a longitudinal magnetic field within a component. When a preformed coil is used, the component is placed against the inside surface on the coil. Coils typically have three or five turns of a copper cable within the molded frame. A foot switch is often used to energize the coil. Conductive cables are wrapped around the component. The cable used is typically 00 extra flexi-
7-166 7-166/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
ble or 0000 extra flexible. The number of wraps is determined by the magnetizing force needed and, of course, the length of the cable. Normally the wraps are kept as close together as possible. When using a coil or cable wrapped into a coil, amperage is usually expressed in ampere-turns. Ampere-turns are the amperage shown on the amp meter times the number of turns in the coil.
Figure 7-95:
Portable Power Supplies Portable power supplies are used to provide the necessary electricity to the prods, coils or cables. Power supplies are commercially available in a variety of sizes. Small power supplies generally provide up to 1,500 A of half wave direct current or alternating current when used with a 4.5 meter 0000 cable. They are small and light enough to be carried and operate on either 120 V or 240 V electrical services. When more power is necessary, mobile power supplies can be used. These units come with wheel so that they can be rolled where needed. These units also operate on 120 V or 240 V electrical services and can provide up to 6,000 A of AC or half-wave DC when 9 meters or less of 0000 cable is used.
7-167 7-167/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.26.21 Stationary Magnetizing Equipment Stationary magnetic particle inspection equipment is designed for use in laboratory or production environment. The most common stationary system is the wet horizontal (bench) unit. Wet horizontal units are designed to allow for batch inspections of a variety of components. The units have head and tail stocks, similar to a lathe but with electrical contact that the part can be clamped between for the production of a circular magnetic field using direct magnetization. The tail stock can be moved and locked into place to accommodate parts of various lengths. To assist the operator in clamping the parts, the contact on the headstock can be moved pneumatically via a foot switch. Most units also have a movable coil that can be moved into place so the indirect magnetization can be used to produce a longitudinal magnetic field. Most coils have five turns and can be obtained in a variety of sizes. The wet magnetic particle solution is collected and held in a tank. A pump and hose system is used to apply the particle solution to the components being inspected. Either the visible or fluorescent particles can be used. Some of the systems offers a variety of options in electrical current used for magnetizing the component. The operator has the option to use AC, half wave DC, or full wave DC. In some units, a demagnetization feature is built in, which uses the coil and decaying AC. To inspect a part using a head-shot, the part is clamped between two electrical contact pads. The magnetic solution, called a bath, is then flowed over the surface of the part. The bath is then interrupted and a magnetizing current is applied to the part for a short duration, typically 0.5 to 1.5 seconds. (Precautions should be taken to prevent burning or overheating of the part.) A circular field flowing around the circumference of the part is created. Leakage fields from defects then attract the particles forming indications.
7-168 7-168/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
When the coil is used to establish a longitudinal magnetic field within the part, the part is placed on the inside surface of the coil. Just as done with a head shot, the bath is then flowed over the surface of the part. A magnetizing current is applied to the part for a short duration, typically 0.5 to 1.5 seconds, just after coverage with the bath is interrupted. (Precautions should be taken to prevent burning or overheating of the part.) Leakage fields from defects attract the particles forming visible indications.
The wet horizontal unit can also be used to establish a circular magnetic field using a central conductor. This type of a setup is used to inspect parts that are hollow such as gears, tubes, and other ring-shaped objects. A central conductor is an electrically conductive bar that is usually made of copper or aluminium. The bar is inserted through the centre of the hollow part and the bar is then clamped between the contact pads. When current is passed through the central conductor, a circular magnetic field flows around the bar and enters into the part or parts being inspected.
7.26.22 Multidirectional Magnetizing Equipment Multidirectional units allow the component to be magnetized in two directions, longitudinally and circumferentially, in rapid succession. Therefore, inspections are conducted without the need for a second shot. In multidirectional units, the two fields are balanced so that the field strengths are equal in both directions. These quickly changing balanced fields produce a multidirectional field in the component providing detection of defects lying in more than one direction. Just as in conventional wet-horizontal systems, the electrical current used in multidirectional magnetization may be alternating, half-wave direct, or full-wave. It is also possible to use a combination of currents depending on the test applications. Multidirectional mag7-169 7-169/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
netization can be used for a large number of production applications, and high volume inspections. To determine adequate field strength and balance of the rapidly changing fields, it requires a little more effort when multidirectional equipment is used. It is desirable to develop the technique using a component with known defects oriented in at least two directions, or a manufactured defect standard. Quantitative Quality Indicators (QQI) are also often used to verify the strength and direction of magnetic fields. 7.26.23 Lights Magnetic particle inspection can be performed using particles that are highly visible under white lighting conditions or particles that are highly visible under ultraviolet lighting conditions. When an inspection is being performed using the visible color contrast particles, no special lighting is required as long as the area of inspection is well lit. A light intensity of at least 1000 lux (100 fc) is recommended when visible particles are used, but a variety of light sources can be used. When fluorescent particles are used, special ultraviolet light must be used. Fluorescence is defined as the property of emitting radiation as a result of and during exposure to radiation. Particles used in fluorescent magnetic particle inspections are coated with a material that produces light in the visible spectrum when exposed to the near-ultraviolet light. This "particle glow" provides high contrast indications on the component anywhere particles collect. Particles that fluoresce yellowgreen are most common because this color matches the peak sensitivity of the human eye under dark conditions. However, particles that fluoresce red, blue, yellow, and green colors are available. Ultraviolet Light Ultraviolet light or "black light" is light in the 1,000 to 4,000 Angstroms (100 to 400 nm) wavelength range in the electromagnetic spectrum. It is a very energetic form of light that is invisible to the human eye. Wavelengths above 4,000 Angstroms fall into the visible light spectrum and are seen as the color violet. UV is separated according to wavelength into three classes: A, B, and C. The shorter the wavelength, the more energy that is carried in the light and the more dangerous it is to the human cells.
7-170 7-170/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The desired wavelength range for use in nondestructive testing is between 3,500 and 3,800 Angstroms with a peak wavelength at about 3,650 A. This wavelength range is used because it is in the UV-A range, which is the safest to work with. UV-B will do an effective job of causing substances to fluoresce, however, it should not be used because harmful effects such as skin burns, and eye damage can occur. This wavelength of radiation is found in the arc created during the welding process. UV-C (1,000 to 2,800) is even more dangerous to living cells and is used to kill bacteria in industrial and medical settings. The desired wavelength range for use in NDT is obtained by filtering the ultraviolet light generated by the light bulb. The output of a UV bulb spans a wide range of wavelengths. The short wave lengths of 3,120 A to 3,340 A are produced in low levels. A peak wavelength of 3650 A is produced at a very high intensity. Wavelengths in the visible violet range (4050 A to 4350 A), green-yellow (5460 A), yellow (6220 A) and orange (6770 A) are also usually produced. The filter allows only radiation in the range of 3200 to 4000 angstroms and a little visible dark purple to pass.
Basic Ultraviolet Lights UV bulbs come in a variety on shapes and sizes. The more common types are the low pressure tube, high pressure spot, and the high pressure flood types. The tubular black light is similar in construction to the tubular fluorescent lights used for office or home illumination. These lights use a low pressure mercury vapor arc. Tube lengths of 6 to 48 inches are common. The low pressure bulbs are most often used to provide general illumination to large areas rather than for illumination of components to be inspected. These bulbs generate a relatively large amount of white light that is a concern as inspection specifications require less than two foot candles of white light at the inspection surface. Flood lights are also used to illuminate the inspection area as they provide even illumination over a large area. Intensity levels for flood lamps are relatively low because the energy is spread over a large area. They generally do not generate the required UV light intensity at the given distance that specifications require.
7-171 7-171/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Spot lights on the other hand provide concentrated energy that can be directed to the area of inspection. A spot light will generate a six inch diameter circle of high intensity light when held fifteen inches from the inspection surface. 100 watt mercury vapor lights are most commonly used, but higher wattage's are available. In the high pressure mercury vapor spot or flood lamps, UV light is generated by a quartz tube inside the bulb. This tube contains two electrodes that establish an arc. The distance between electrodes is such that a starting electrode must be used. A resister limits the current to the starting electrode that establishes the initial arc that vaporizes the mercury in the tube. Once this low level arc is established and the mercury is vaporized the arc between the main electrodes is established. It takes approximately five minutes to "warm up" and establish the arc between the main electrodes. This is why specifications require a "warm up time" before using the high pressure mercury vapor lights. Flood and spot black lights produce large amounts of heat and should be handled with caution to prevent burns. This condition has been eliminated by newer designs that include cooling fans. The arc in the bulb can be upset when exposed to an external magnetic field, such as that generated by a coil. Care should be taken not to bring the lamp close to strong magnetic fields, but if the arc is upset and extinguished, it must be allowed to cool before it can be safely restarted.
High Intensity Ultraviolet Lights The 400 watt metal halide bulbs or "super lights" can be found in some facilities. This super bright light will provide adequate lighting over an area of up to ten times of that covered by the 100 watt bulb. Due to their high intensity, excessive light reflecting from the surface of a component is a concern. Moving the light a greater distance from the inspection area will generally reduce this glare. Another type of high intensity light available is the micro discharge light. This particular light produces up to ten times the amount of UV light conventional lights produce. Readings of up to 60,000 uW/cm2 at 15 inches can be achieved.
7-172 7-172/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.26.24 Magnetic Field Indicators Determining whether a magnetic field is of adequate strength and in the proper direction is critical when performing magnetic particle testing. As discussed previously, knowing the direction of the field is important because the field should be as close to perpendicular to the defect as possible and no more than 45 degrees from normal. Being able to evaluate the field direction and strength is especially important when inspecting with a multidirectional machine, because when the fields are not balanced properly a vector field will be produced that may not detect some defects. There is actually no easy to apply method that permits an exact measurement of field intensity at a given point within a material. In order to measure the field strength it is necessary to intercept the flux lines. This is impossible without cutting into the material and cutting the material would immediately change the field within the part. However, cutting a small slot or hole into the material and measuring the leakage field that crosses the air gap with a Gauss meter is probably the best way to get an estimate of the actual field strength within a part. Nevertheless, there are a number of tools and methods available that are used to determine the presence and direction of a field surrounding the component. Gauss Meter or Hall Effect Gage A Gauss meter with a Hall Effect probe is commonly used to measure the tangential field strength on the surface of the part. As discussed in some detail on the "Measuring Magnetic Fields" page, the Hall Effect is the transverse electric field created in a conductor when placed in a magnetic field. Gauss meters, also called Tesla meters, are used to measure the strength of a field tangential to the surface of the magnetized test object. The meters measure the intensity of the field in the air adjacent to the component when a magnetic field is applied. The advantages of Hall Effect devices are; they provide a quantitative measure of the strength of magnetizing force tangential to the surface of a test piece, they can be used for measurement of residual magnetic fields, and they can be used repetitively. Their main disadvantages are that they must be periodically calibrated, and they cannot be used to establish the balance of fields in multidirectional applications. Quantitative Quality Indicator (QQI) The Quantitative Quality Indicators (QQI) or Artificial Flaw Standard are often the preferred method of assuring proper field direction and adequate field strength. The use of QQI is also the only practical way of ensuring balanced field intensity and direction in multiple-direction magnetization equipment. QQI are often used in conjunction with a Gauss meter to establish the inspection procedure for a particular component. They are used with the wet method only and, as other flux sharing devices; they can only be used when continuous magnetization is used. The QQI is a thin strip of either 0.002 or 0.004 inch thick AISI 1005 steel. A photo etch process is used to inscribe a specific pattern, such as concentric circles or a plus sign. QQI are nominally 3/4 inch square, but miniature shims are also available. QQI
7-173 7-173/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
must be in intimate contact with the part being evaluated. This is accomplished by placing the shim on a part etched side down, and taping or gluing it to the surface. The component is then magnetized and particles applied. When the field strength is adequate, the particles will adhere over the engraved pattern and provide information about the field direction. When a multidirectional technique is used, a balance of the fields is noted when all areas of the QQI produce indications. Some of the advantages of QQI are: they can be quantified and related to other parameters; they can accommodate virtually any configuration with suitable selection; and they can be reused with careful application and removal practices. Some of the disadvantages are: the application process is somewhat slow, the parts must be clean and dry; shims cannot be used as a residual magnetism indicator as they are a flux sharing device; they can be easily damaged with improper handling and will corrode if not cleaned and properly stored.
Figure 7-96:
Left is a photo of a typical QQI shim. The photo on the right shows
the indication produced by the QQI when it is applied to the surface a part and a magnetic field is established that runs across the shim from right to left.
Pie Gage The pie gage is a disk of highly permeable material divided into four, six, or eight sections by nonferromagnetic material. The division serve as artificial defects that radiate out in different directions from the centre. Diameter of the gage is to 1 inch. The divisions between the low carbon steel pie sections are to be no greater than 1/32 inch. The sections are furnace brazed and copper plated. The gage is placed on the test piece copper side up, and the test piece is magnetized. After particles are applied, and excess removed, the indications provide the inspector the orientation of the magnetic field. 7-174 7-174/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The principal application is on flat surfaces such as weldments or steel castings where dry powder is used with a yoke or prods. The pie gage is not recommended for precision parts with complex shapes, for wet-method applications, or for proving field magnitude. The gage should be demagnetized between readings. Several of the main advantages of the pie gage are: it is easy to use and it can be used indefinitely without deterioration. The pie gage has several disadvantages, which include: it retains some residual magnetism so indications will prevail after removal of the source of magnetization, it can only be used in relatively flat areas, and it cannot be reliably used for determination of balanced fields in multidirectional magnetization.
Slotted Strips Slotted strips, also known as Burmah-Castrol Strips, are pieces of highly permeable ferromagnetic material with slots of different widths. They are placed on the test object as it is inspected. The indications produced on the strips give the inspector a general idea of the field strength in a particular area. Advantages of these strips are: they are relatively easily applied to the component; they can be used successfully with either the wet or dry method when using the continuous magnetization; they are repeatable as long as orientation to the magnetic field is maintained and they can be used repetitively. Disadvantages include: they cannot be bent to complex configuration; and they are not suitable for multidirectional field applications since they indicate defects in only one direction.
7-175 7-175/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.26.25 Magnetic Particles As mentioned previously, the particles that are used for magnetic particle inspection are a key ingredient as they form the indications that alert the inspector to defects. Particles start out as tiny milled (a machining process) pieces of iron or iron oxide. A pigment (somewhat like paint) is bonded to their surfaces to give the particles colour. The metal used for the particles has high magnetic permeability and low retentivity. High magnetic permeability is important because it makes the particles attract easily to small magnetic leakage fields from discontinuities, such as flaws. Low retentivity is important because the particles themselves never become strongly magnetized so they do not stick to each other or the surface of the part. Particles are available in a dry mix or a wet solution.
Dry Magnetic Particles Dry magnetic particles can typically be purchased in red, black, grey, yellow and several other colours so that a high level of contrast between the particles and the part being inspected can be achieved. The size of the magnetic particles is also very important. Dry magnetic particle products are produced to include a range of particle sizes. The fine particles are around 50 m (0.002 inch) in size are about three times smaller in diameter and more than 20 times lighter than the coarse particles (150 m or 0.006 inch), which make them more sensitive to the leakage fields from very small discontinuities. However, dry testing particles cannot be made exclusively of the fine particles. Coarser particles are needed to bridge large discontinuities and to reduce the powder's dusty nature. Additionally, small particles easily adhere to surface contamination, such as remanent dirt or moisture, and get trapped in surface roughness features producing a high level of background. It should also be recognized that finer particles will be more easily blown away by the wind and, therefore, windy conditions can reduce the sensitivity of an inspection. Also, reclaiming the dry particles is not recommended because the small particles are less likely to be recaptured and the "once used" mix will result in less sensitive inspections.
7-176 7-176/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The particle shape is also important. Long, slender particles tend to align themselves along the lines of magnetic force. However, research has shown that if dry powder consists only of long, slender particles, the application process would be less than desirable. Elongated particles come from the dispenser in clumps and lack the ability to flow freely and form the desired "cloud" of particles floating on the component. Therefore, globular particles are added that are shorter. The mix of globular and elongated particles results in a dry powder that flows well and maintains good sensitivity. Most dry particle mixes have particle with L/D ratios between one and two. Wet Magnetic Particles Magnetic particles are also supplied in a wet suspension such as water or oil. The wet magnetic particle testing method is generally more sensitive than the dry because the suspension provides the particles with more mobility and makes it possible for smaller particles to be used since dust and adherence to surface contamination is reduced or eliminated. The wet method also makes it easy to apply the particles uniformly to a relatively large area. Wet method magnetic particles products differ from dry powder products in a number of ways. One way is that both visible and fluorescent particles are available. Most nonfluorescent particles are ferromagnetic iron oxides, which are either black or brown in colour. Fluorescent particles are coated with pigments that fluoresce when exposed to ultraviolet light. Particles that fluoresce green-yellow are most common to take advantage of the peak colour sensitivity of the eye but other fluorescent colours are also available. (For more information on the colour sensitivity of the eye, see the penetrant inspection material.)
The particles used with the wet method are smaller in size than those used in the dry method for the reasons mentioned above. The particles are typically 10 m (0.0004 inch) and smaller and the synthetic iron oxides have particle diameters around 0.1 m (0.000004 inch). This very small size is a result of the process used to form the particles and is not particularly desirable, as the particles are almost too fine to set-
7-177 7-177/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
tle out of suspension. However, due to their slight residual magnetism, the oxide particles are present mostly in clusters that settle out of suspension much faster than the individual particles. This makes it possible to see and measure the concentration of the particles for process control purposes. Wet particles are also a mix of long slender and globular particles. The carrier solutions can be water- or oil-based. Water-based carriers form quicker indications, are generally less expensive, present little or no fire hazard, give off no petrochemical fumes, and are easier to clean from the part. Water-based solutions are usually formulated with a corrosion inhibitor to offer some corrosion protection. However, oil-based carrier solutions offer superior corrosion and hydrogen embrittlement protection to those materials that are prone to attack by these mechanisms. 7.26.26 Suspension Liquids Suspension liquids used in the wet magnetic particle inspection method can be either a well refined light petroleum distillate or water containing additives. Petroleumbased liquids are the most desirable carriers because they provided good wetting of the surface of metallic parts. However, water-based carriers are used more because of low cost, low fire hazard, and the ability to form indications quicker than solvent-based carriers. Water-based carriers must contain wetting agents to disrupt surface films of oil that may exist on the part and to aid in the dispersion of magnetic particles in the carrier. The wetting agents create foaming as the solution is moved about, so anti-foaming agents must be added. Also, since water promotes corrosion in ferrous materials, corrosion inhibitors are usually added as well. Petroleum based carriers are primarily used in systems where maintaining the proper particle concentration is a concern. The petroleum based carriers require less maintenance because they evaporate at a slower rate than the water-based carriers. Therefore, petroleum based carriers might be a better choice for a system that only gets occasional use and adjusting the carrier volume with each use is undesirable. Modern solvent carriers are specifically designed with properties that have flash points above 200 degrees F and keep nocuous vapours low. Petroleum carriers are required to meet certain specifications such as AMS 2641.
7-178 7-178/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.26.27 Testing Practices Dry Particle Inspection In this magnetic particle testing technique, dry particles are dusted onto the surface of the test object as the item is magnetized. Dry particle inspection is well suited for the inspections conducted on rough surfaces. When an electromagnetic yoke is used, the AC or half wave DC current creates a pulsating magnetic field that provides mobility to the powder. The primary applications for dry powders are ungrounded welds and rough as-cast surfaces. Dry particle inspection is also used to detect shallow subsurface cracks. Dry particles with half wave DC is the best approach when inspecting for lack-of-root penetration in welds of thin materials. Half wave DC with prods and dry particles is commonly used when inspecting large castings for hot tears and cracks.
Steps in performing an inspection using dry particles Prepare the part surface - the surface should be relatively clean but this is not as critical as it is with liquid penetrant inspection. The surface must be free of grease, oil or other moisture that could keep particles from moving freely. A thin layer of paint, rust or scale will reduce test sensitivity but can sometimes be left in place with adequate results. Specifications often allow up to 0.003 inch (0.076 mm) of a nonconductive coating (such as paint) and 0.001 inch max (0.025 mm) of a ferromagnetic coating (such as nickel) to be left on the surface. Any loose dirt, paint, rust or scale must be removed. Apply the magnetizing force - Use permanent magnets, an electromagnetic yoke, prods, a coil or other means to establish the necessary magnetic flux. Dust on the dry magnetic particles - Dust on a light layer of magnetic particles. Gently blow off the excess powder - With the magnetizing force still applied, remove the excess powder from the surface with a few gently puffs of dry air. The force of the air needs to be strong enough to remove the excess particle but not strong enough to dislodge particle held by a magnetic flux leakage field. Terminate the magnetizing force - If the magnetic flux is being generated with an electromagnet or an electromagnetic field, the magnetizing force should be terminated. If permanent magnets are being used, they can be left in place. Inspect for indications - Look for areas where the magnetic particles are clustered. 7-179 7-179/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Wet Suspension Inspection Wet suspension magnetic particle inspection, or more commonly wet magnetic particle inspection, involves applying the particles while they are suspended in a liquid carrier. Wet magnetic particle inspection is most commonly performed using a stationary, wet, horizontal inspection unit but suspensions are also available in spray cans for use with an electromagnetic yoke. A wet inspection has several advantages over a dry inspection. First, all the surfaces of the component can be quickly and easily covered with a relatively uniform layer of particles. Second, the liquid carrier provides mobility to the particles for an extended period of time, which allows enough particles to float to small leakage fields to form a visible indication. Therefore, wet inspection is considered best for detecting very small discontinuities on smooth surfaces. On rough surfaces, however, the particles (which are much smaller in wet suspensions) can settle in the surface valleys and loose mobility rendering them less effective than dry powders under these conditions. Steps in performing an inspection using wet suspensions Prepare the part surface - Just as is required with dry particle inspections, the surface should be relatively clean. The surface must be free of grease, oil and other moisture that could prevent the suspension from wetting the surface and preventing the particles from moving freely. A thin layer of paint, rust or scale will reduce test sensitivity, but can sometimes be left in place with adequate results. Specifications often allow up to 0.003 inch (0.076 mm) of a nonconductive coating (such as paint) and 0.001 inch max (0.025 mm) of a ferromagnetic coating (such as nickel) to be left on the surface. Any loose dirt, paint, rust or scale must be removed. Apply the suspension - The suspension is gently sprayed or flowed over the surface of the part. Usually, the stream of suspension is diverted from the part just before the magnetizing field is applied. Apply the magnetizing force - The magnetizing force should be applied immediately after applying the suspension of magnetic particles. When using a wet horizontal inspection unit, the current is applied in two or three short busts (1/2 second) which helps to improve particle mobility. Inspect for indications - Look for areas where the magnetic particles are clustered. Surface discontinuities will produce a sharp indication. The indications from subsurface flaws will be less defined and loose definition as depth increases. 7-180 7-180/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.26.28 Inspection using Magnetic Rubber The magnetic rubber technique was developed for detecting very fine cracks and is capable of revealing finer cracks than other magnetic techniques. Additionally, the technique can be use to examine difficult to reach areas, such as the threads on the inside diameter of holes, where the moulded plugs can be removed and examined under ideal conditions and magnification if desired. The trade-off, of course, is that inspection times are much longer. The technique uses a liquid (uncured) rubber containing suspended magnetic particles. The rubber compound is applied to the area to be inspected on a magnetized component. Inspections can be performed using either an applied magnetic field, which is maintained while the rubber sets (active field), or the residual field from magnetization of the component prior to pouring the compound. A dam of modelling clay is often used to contain the compound in the region of interest. The magnetic particles migrate to the leakage field caused by a discontinuity. As the rubber cures, discontinuity indications remain in place on the rubber. The rubber is allowed to completely set, which takes from 10 to 30 minutes. The rubber cast is removed from the part. The rubber conforms to the surface contours and provides a reverse replica of the surface. The rubber cast is examined for evidence of discontinuities, which appear as dark lines on the surface of the moulding. The moulding can be retained as a permanent record of the inspection. Magnetic rubber methods require similar magnetizing systems used for dry method magnetic particle tests. The system may include yokes, prods, clamps, coils or central conductors. Alternating, direct current, or permanent magnets may be used to draw the particles to the leakage fields. The direct current yoke is the most common magnetization source for magnetic rubber inspection. 7.26.29 Continuous and Residual Magnetization Techniques In magnetic particle inspection, the magnetic particles can either be applied to the component while the magnetizing force is applied, or after it has been stopped. Continuous magnetization describes the technique where the magnetizing force is applied and maintained while the magnetic particles are dusted or flowed onto the surface of the component. In a wet horizontal testing unit the application of the particles is stopped just before the magnetizing force is applied but since particles are still flowing over and covering the surface, this is considered continuous magnetization. Residual magnetization, on the other hand, describes the technique where the magnetizing force is applied to magnetize the component and then stopped before applying the magnetic particles. Only the residual field of the magnetized component is used to attract magnetic particles and produce an indication.
7-181 7-181/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The continuous technique is generally chosen when maximum sensitivity is required because it has two distinct advantages over the residual technique. First, the magnetic flux will be highest when current is flowing and, therefore, leakage fields will also be strongest. Field strength in a component depends primarily on two variables...the applied magnetic field strength and the permeability of the test object. Viewing the upper right portion of the hysteresis loop below, it is evident that the magnetic flux will be strongest when the magnetizing force is applied. If the magnetizing force is strong enough, the flux density will reach the point of saturation. When the magnetizing force is removed, the flux density will drop to the retentivity point. The two grey traces show the path the flux density would follow if the magnetizing force was applied and removed at levels below that required to reach saturation. It can be seen that the flux density is always highest while the magnetizing current is applied. This is independent of the permeability of a material.
However, the permeability of the material is very important. High permeability materials do not retain a strong magnetic field so flux leakage fields will be extremely weak or nonexistent when the magnetizing force is removed. Therefore, materials with high magnetic permeability are not suited for inspection using the residual technique. When the residual technique is used to inspect materials with low permeability, care should be taken to ensure that the residual field is of the necessary strength to produce an indication. Defects should be relatively large and surface breaking to have a high probability of detection using the residual method. 7-182 7-182/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The second advantage of the continuous technique is that when current is used to generate the magnetizing force, it can provide added particle mobility. Alternating or pulsed direct current will cause the particles to vibrate and move slightly on the surface of the part. This movement allows the particles to travel to leakage sites. More particles mean brighter indications compared to those formed using the residual technique. One disadvantage of the continuous method is that heating of the component occurs when using direct magnetization. For example, when prods are used, they may create areas of localized heating when the continuous technique is used. This may be acceptable on components that will be further processed removing this condition but machined or in-service components may be adversely affected by this condition. While generally not recommended, the residual technique does have its uses. It is commonly used in automated inspection systems to inspect materials with high retentivity. To speed throughput, automated systems often magnetize the parts and then submerge them in an agitated magnetic particle bath or pass them through a spray station. Closely controlled automated systems provided good results using the residual magnetism technique. 7.26.30 Field Direction and Intensity Field Direction As discussed previously, determining the direction of the field is important when conducting a magnetic particle inspection because the defect must produce a significant disturbance in the magnetic field to produce an indication. It is difficult to detect discontinuities that intersect the magnetic field at an angle less than 45 degrees. When the orientation of a defect is not well established, components should be magnetized in a minimum of two directions at approximately right angles to each other. Depending on the geometry of the component, this may require longitudinal magnetization in two or more directions, multiple longitudinal and circular magnetization or circular magnetization in multiple directions. Determining strength and direction of the fields is especially critical when inspecting with a multidirectional machine. If the fields are not balanced a vector field will be produced that may not detect some defects. Depending on the application, pie gages, QQI, and a gauss meter can be used to check the field direction. The PIE gage is generally only used with dry powder inspections. QQI shims can be used in a variety of applications but are the only method recommended for use in establishing balanced fields when using multidirectional equipment. Field Strength The applied magnetic field must have sufficient strength to produce a satisfactory indication, but not so strong that it produces nonrelevant indications or limits particle mobility. If the magnetizing current is excessively high when performing a wet fluorescent particle inspection, particles can be attracted to the surface of the part and not allowed to migrate to the flux leakage fields of defects. When performing a dry
7-183 7-183/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
particle inspection, an excessive longitudinal magnetic field will cause furring. Furring is when magnetic particles build up at the magnetic poles of a part. When the field strength is excessive, the magnetic field is forced out of the part before reaching the end of the component and the poles along its length attract particle and cause high background levels. Adequate field strength may be determined by:
performing an inspection on a standard specimen that is similar to the test component and has known or artificial defects of the same type, size, and location as those expected in the test component. QQI shims can sometimes be used as the artificial defects. using a gauss meter with a Hall Effect probe to measure to the peak values of the tangent field at the surface of the part in the region of interest. Most specifications call for a field strength of 30 to 60 gauss at the surface when the magnetizing force is applied. looking for light furring at the ends pipe and bar when performing dry particle inspections of pipe, bar and other uncomplicated shapes.
Formula for calculating current levels should only be used to estimate current requirements. The magnetic field strength resulting from calculations should be assessed for adequacy using one of the two method discussed above. Likewise, published current level information should also be used only as a guide unless the values have been established for the specific component and target defects of the inspection at hand. Using a PIE Gage A PIE gage is placed copper side up and held in contact with the component as the magnetic field and particles are applied. Indications of the leakage fields provide a visual representation of defect direction within the component. PIE gages work well on flat surfaces, but if the surface is concave or convex inaccurate readings may occur. The PIE gage is a flux sharing device and requires good contact to provide accurate readings.
Using Quantitative Quality Indicator (QQI) Shims Quantitative Quality Indicator (QQI) flaw shims are used to establish proper field direction and to ensure adequate field strength during technique development. The QQI flaw shim is the most efficient means of determining balance and effectiveness of fields. The QQI are also flux sharing devices and must be properly attach so as not allow particle to be trapped under the artificial flaw. Application using super glue is the preferred way of attaching the artificial flaw, but does not allow for reuse of the shims. Shims can also be attached with tape applied to just the edge of the shim. It is recommended that the tape be impervious to oil, not be fluorescent, and be 1/4 to 1/2 inch in width. 7-184 7-184/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
The QQI must be applied to locations on the component where the flux density may very. One example would be the center area of a yoke or Y shaped component. Often the flux density will be near 0 in this area. If two legs of a Y are in contact with the pad in circular magnetization it must be determined if current is flowing evenly through each leg. A QQI on each leg would be appropriate under such conditions. QQI's can be used to establish system threshold values for a defect of a given size. By attaching a QQI shim with three circles (40%, 30% and 20% of shim thickness) to the threshold values for a specific area of the component can be established. Begin by applying current at low amperage and slowly increasing it until the largest flaw is obtained. The flux density sound is verified and recorded using a Hall effects probe. The current is then increased until the second circle is identified and the flux density is again recorded. As the current is raised more, the third ring is identified and the current values are recorded.
Hall Effects Gauss Meter There is several types of Hall effects probes that can be used to measure the magnetic field strength. Transverse probes are the type most commonly used to evaluate the field strength in magnetic particle testing. Transverse probes have the Hall Effect element mounted in a thin, flat stem and they are used to make measurements between two magnetic poles. Axial probes have the sensing element mounted such that the magnetic flux in the direction of the long axis of the probe is measured. To make a measurement with a transverse probe, the probe is positioned such that the flat surface of the Hall Effect element is transverse to the magnetic lines of flux. The Hall Effect voltage is a function of the angle at which the magnetic lines of flux pass through the sensing element. The greatest Hall Effect voltage occurs when the lines of flux pass perpendicularly through the sensing element. If not perpendicular, the output voltage is related to the cosine of the difference between 90 degrees and the actual angle. The peak field strength should be measured when the magnetizing force is applied. The field strength should be measured in all areas of the component to be inspected.
7-185 7-185/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.26.31 Particle Concentration and Condition Particle Concentration The concentration of particles in the suspension is a very important parameter in the inspection process and must be closely controlled. The particle concentration is checked after the suspension is prepared and continued regularly as part of the quality system checks. ASTM E-1444-01 requires concentration checks to be performed every eight hours or every shift change. The standard process used to perform the check requires agitating the carrier for a minimum of thirty minutes to ensure even particle distribution. A sample is then taken in a pear-shaped 100 ml centrifuge tube having a stem graduated to 1.0 ml in 0.05 ml increments for fluorescent particles, and graduated to 1.5 ml. in 0.1 ml increments for visible particles. The sample is then demagnetized so that the particles do not clump together while settling. The sample must then remain undisturbed for a minimum of 60 minutes for a petroleum-based carrier or 30 minutes for a water-based carrier, unless shorter times have been documented to produce results similar to the longer settling times. The volume of settled particles is then read. Acceptable ranges are 0.1 to 0.4 ml for fluorescent particles and 1.2 to 2.4 ml for visible particles. If the particle concentration is out of the acceptable range, particles or the carrier must be added to bring the solution back in compliance with the requirement. Particle loss is often attributed to "drag out". Drag out occurs because the solvent easily runs off components and is recaptured in the holding tank. Particles, on the other hand, tend to adhere to components, or be trapped in geometric features of the component. These particles will be "drug out" or lost to the system, and will eventually need to be replaced.
Particle Condition After the particles has settled, they should be examined for brightness and agglomeration. Fluorescent particles should be evaluated under ultraviolet light and visible particles under white light. The brightness of the particles should be evaluated weekly by comparing the particles in the test solution to those in an unused reference solution that was saved when the solution was first prepared. The brightness of the two solutions should be relatively the same. Additionally, the particles should appear loose and not lumped together. If the brightness or the agglomeration of the particles is noticeably different from the reference solution, the bath should be replaced. 7-186 7-186/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.26.32 Lighting Magnetic particle inspection predominately relies on visual inspection to detect any indications that are formed. Therefore, lighting is a very important element of the inspection process. Obviously, the lighting requirements are different for an inspection conducted using visible particles than they are for an inspection conducted using fluorescent particles. The lighting requirements for each of these techniques, as well as how light measurements are made, are discussed below. Light Requirements When Using Visible Particles Magnetic particle inspections conducted using visible particles can be conducted using natural lighting or artificial lighting. When using natural lighting, it is important to keep in mind that daylight varies from hour to hour. Inspector must stay constantly aware of the lighting conditions and make adjustments when needed. To improve uniformity in lighting from one inspection to the next, the use of artificial lighting is recommended. Artificial lighting should be white whenever possible and white flood or halogen lamps are most commonly used. The light intensity is required to be 100 foot-candles at the surface being inspected. It is advisable to choose a white light wattage that will provide sufficient light, but avoid excessive reflected light that could distract from the inspection. Light Requirements When Using Fluorescent Particles Ultraviolet Lighting When performing a magnetic particle inspection using fluorescent particles, the condition of the ultraviolet light and the ambient white light must be monitored. Standards and procedures require verification of lens condition and light intensity. Black lights should never be used with a cracked filter as output of white light and harmful black light will be increased. The cleanliness of the filter should also be checked as a coating of solvent carrier, oils, or other foreign materials can reduce the intensity by up to as much as 50%. The filter should be checked visually and cleaned as necessary before warm-up of the light. For UV lights used in component evaluations, the normally accepted intensity is 1000 microwatts per square centimetre when measured at 15 inches from the filter face (requirements can vary from 800 to 1200). The required check should be performed when a new bulb is installed, at start-up of the inspection cycle, if a change in intensity is noticed, or every eight hours if in continuous use. Regularly checking the intensity of UV lights is very important because bulbs loose intensity over time. In fact, a bulb that is near the end of its operating life will often have an intensity of only 25 percent of its original output. Black light intensity will also be affected by voltage variations, so it is important to provide constant voltage to the light. A bulb that produces acceptable intensity at 120 volts will produce significantly less at 110 volts.
7-187 7-187/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Ambient White Lighting When performing a fluorescent magnetic particle inspection, it is important to keep white light to a minimum as it will significantly reduce the inspectors ability to detect fluorescent indications. Light levels of less than 2 fc are required by most procedures with some procedures requiring less than 0.5 fc at the inspection surface. When checking black light intensity at 15 inches a reading of the white light produced by the black light may be required to verify white light is being removed by the filter. White Light for Indication Confirmation While white light is held to a minimum in fluorescent inspections, procedures may require that indications be evaluated under white light. The white light requirements here are the same as when performing an inspection with visible particles. The minimum light intensity at the surface being inspected must be 100 foot-candles. Light Measurement Light intensity measurements are made using a radiometer. A radiometer is an instrument that translates light energy into an electrical current. Light striking a silicon photodiode detector causes a charge to build up between internal layers. When an external circuit is connected to the cell, an electrical current is produced. This current is linear with respect to incident light. Some radiometers have the ability to measure both black and white light, while others require a separate sensor for each measurement. Whichever type used, the sensing area should be clean and free of any materials that could reduce or obstruct light reaching the sensor. Radiometers are relatively unstable instruments and readings often change considerably over time. Therefore, they must be calibrated regularly. They should be calibrated at least every six months. A unit should be checked to make sure its calibration is current before taking any light readings. Ultraviolet light measurements should be taken using a fixture to maintain a minimum distance of 15 inches from the filter face to the sensor. The sensor should be centred in the light field to obtain and record the highest reading. UV spot lights are often focused so intensity readings will vary considerable over a small area. White lights are seldom focused and depending on the wattage, will often produce in excess of the 100 fc at 15 inches. Many specifications do not require the white light intensity check to be conducted at a specific distance. 7-188 7-188/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.26.33 Eye Consideration Eye Adaptation Just as lighting is an important consideration in the inspection process, so is the eyes response to light. Scientists have recently discovered that a special, tiny group of cells at the back of the eye help tell the brain how much light there is, causing the pupil to get bigger or smaller. The change in pupil diameter is not instantaneous and, therefore, eyes must be given time to adapt to changing lighting conditions. When performing a fluorescent magnetic particle inspection, the eye must be given time to adapt to the darkness of the inspection booth before beginning to look for indications. Dark adaptation time of at least one minute is required by most procedures. Some studies recommend adaptation time of five minutes if entering an inspection area from direct sunlight. Technicians should be aware of, and adhere to, the adaptation time procedures requirements as it is quite easy to overlook and begin inspection before the eyes have adjusted to the darkened conditions. Eyeball Fluorescence When ultraviolet light enters the human eye, the fluid that fills the eye fluoresces. This condition is called eyeball fluorescence, and while it is considered harmless, it is annoying and interferes with vision while it exists. When working around ultraviolet lights, one should be careful not to look directly into lights and to hold spot lights to avoid reflection. UV light will be reflected from surfaces just as white light will and, therefore it is advisable to consider placement of lights to avoid this condition. Special filtered glasses may be worn by the inspector to remove all UV light from reaching the eyes but allowing yellow-green light from fluorescent indications to pass. Technicians should never wear darkened or photo chromatic glasses as these glasses also filter or block light from fluorescent indications. 7.26.34 Examples of Visible Dry Indications One of the advantages that a magnetic particle inspection has over some of the other nondestructive evaluation methods is that flaw indications generally resemble the actual flaw. This is not the case with NDT methods such as ultrasonic and eddy current inspection, where an electronic signal must be interpreted. When magnetic particle inspection is used, cracks on the surface of the part appear as sharp lines that follow the path of the crack. Flaws that exist below the surface of the part are less defined and more difficult to detect. Below are some examples of magnetic particle indications produced using dry particles.
7-189 7-189/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-97:
Figure 7-98:
Figure 7-99:
7-190 7-190/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-101:
7-191 7-191/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.26.35 Examples of Fluorescent Wet Indications The indications produced using the wet magnetic particles are sharper than dry particle indications formed on similar defects. When fluorescent particles are used, the visibility of the indications is greatly improved because the eye is drawn to the "glowing" regions in the dark setting. Below are a few examples of fluorescent wet magnetic particle indications.
Figure 7-103:
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7-193 7-193/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-106:
7-194 7-194/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.27
Strain gauging
In principle this method can be used on steel, concrete and masonry structures. It is most commonly used on steel structures. The strain gauge is the fundamental sensing element in many types of sensors such as pressure sensors, load cells, torque sensors etc. The strain gauge may also be used to monitor the strain at a given location on a structure. Most strain gauges are foil types. When the foil is subjected to strain the resistance of the foil changes in a defined way. When the strain gauge is connected into a Wheatstone bridge the change of the resistance and thereby the change of strain in the strain gauge may be registered. 7.27.1 Measurement Principle In principle, a strain gauge is an electrical-resistance wire bonded to a backing sheet of foil. It is usually cemented onto the object to be measured. When the object is loaded, the gauge follows the deformation of the material, and its resistance changes in proportion to the deformation. The extremely small changes in the resistance in the gauge are recorded, and then used to determine the strain in the material.
Figure 7-108:
a c
a) b) c) d) e)
7-195 7-195/311
d
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
e
FORCE Technology 2006
7.27.3 Applications Strain gauge measurements shed light on deformation, stresses and loads in just about any structure. Strain gauges are often used in conjunction with live load testing of bridges. The strain gauges are placed at locations where the live load causes large strains and at hot-spots where the live-load may lead to fatigue fracture. These locations are determined on the basis of a Finite Element analysis of the considered structure. A number of different types of sensors are available. A sensor may consist of a single strain gauge measuring the strain in only one direction, two strain gauges measuring the strain in two directions (usually at an angle of 90o) or three strain gauges. Using a sensor consisting of three strain gauges allow the principal strains to be determined (in the case of plane strain). Properly mounting a strain gauge is critical to its performance in ensuring that the applied strain of a material is accurately transferred through the adhesive and backing material to the foil itself. The mounting of strain gauges should therefore always be performed by trained professionals. Strain gauge measurements are usually only used for structures with a complicated geometry and/or structures where some parameters influencing the stress distribution are unknown or subject to considerable uncertainty. The method is not typical for bridge inspection and is primarily used to monitor bridges having high loads or high deterioration levels. The gauge length required depends on the material of the structure being tested and the strain gradient induced. Strain gauges are readily available with gauge lengths ranging from less than 1 mm, for high strain gradients in homogeneous materials, to 100 mm for small strain gradients in concrete. For direct attachments to masonry structures the latter gauges are appropriate. The results of the measurements may be used to calibrate the Finite Element model used to evaluate the stress distribution in the structure. This assures a more reliable evaluation of the load-carrying capacity and/or fatigue life of the structure. Examples:
Machines - also rotating parts Bridges Offshore structures Cranes Pressure vessels Concrete structures Prototypes of all kinds. 7-196 7-196/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Figure 7-109:
Squeezing Machine.
Figure 7-110:
MR-Train shaft.
7-197 7-197/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7-198 7-198/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Measurement results from the bridge. The above curves show how the wind influences on the rail. The straight downward going line between the peaks shows how a moderate wind (as it turns) influences on the rail. The peaks in the red circle show a blast of wind. The measurements are done in very short intervals, about 200 readings per/sec.
Figure 7-112:
Telescope-crane.
Figure 7-113:
Pressure Vessel.
7.27.4 Structural design Characteristic of the development of structural designs is that, at some point in the design phase, computations have to be made to determine the strength of the vari-
7-199 7-199/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
ous sections of the structure. Often, advanced software will be used in this connection employing e.g. the Finite Element Method (FEM). Only if all the assumptions - such as size of load, load frequency, and support conditions - are completely correct will such computations give a true picture of structural deformation and stress. Once the structure has been erected, strain gauges placed at critical points in the structure can be used to monitor very closely whether the assumptions made were satisfactory or whether the computational model needs to be adjusted. Strain gauge measurements provide a solid basis for making changes to give the structure the desired lifetime, and so that future structures can be built as efficiently and economically as possible. Below an example: Measurements are done on a shaft of a ferry. The curves shows vibration and torsion as it turns.
7.27.5 Fitness for purpose Computational methods in modem fracture mechanics can be used to evaluate the fitness for purpose of defective structures, e.g. structures with welding defects. These methods require a precise determination of the type, size and location of such
7-200 7-200/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
defects. This can be done using non-destructive test methods: ultrasound, eddy current, and others. It is also necessary to know precisely the stresses in the defective areas. Strain gauge measurements are the most reliable non-destructive test method in this situation, and it should be used where there is uncertainty about the stresses. Strain gauging results can be used to assess: Safety Remaining lifetime Mode of vibration Repair and reinforcement options.
Figure 7-114:
7.27.6 Testing and documentation Strain Gauge measurements can be performed as: Static testing, such as is used in the certification of pressure vessels Dynamic testing - recording variations in load over time Wireless signal transfer - measuring rotary machine parts using telemetric equipment Testing with a friction gauge - a quick and economical test of stress variations Residual stress testing, e.g. checking results of stress-relief annealing Bolt tension testing - measurement of bolt preload. 7-201 7-201/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7-202 7-202/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.27.7 Mechanical Strain Gauge There are many types of mechanical strain gauges, but the most common is the Demec gauge. The Demec gauge is essentially a bar with a small conical point at one end and a spring-loaded pivoted lever at the other. On one side of the pivot is another small conical point and a dial gauge records the levers position on the other side of the pivot. Small targets with central conical recesses are stuck to the structure, with their distance apart being equal to the Demec gauge length in the middle of its measuring range.
The relationship between the dial gauge reading and strain depends on the gauge length and lever arm ratio and is provided with the Demec gauge. Demec gauges are available with gauge lengths from 50 to 2000 mm, but 200 mm is the most common and well suited for masonry structures. The accuracy of the Demec gauge of 200 mm is 6 x 10-6. This is, however only theoretical as the accuracy also depends on the skills of the operator. To compensate for measuring mistakes made by the operator it is advantageous to use the same operator.
7-203 7-203/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.28
Principle
7-204 7-204/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.29
Radiography (RT)
Radiography involves the use of penetrating gamma or X-radiation to examine parts and products for imperfections. An X-ray generator or radioactive isotope is used as a source of radiation. Radiation is directed through a part and onto film or other imaging media. The resulting shadowgraph shows the dimensional features of the part. Possible imperfections are indicated as density changes on the film in the same manner as medical X-ray shows broken bones.
The principle
X-ray generator
Isotope
X-ray on bridgedeck
7-205 7-205/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.30
Sonic Methods
For relatively homogeneous materials ultrasonic methods are well-established procedures but the high resolution at shallow depths of penetration of high frequency causes the ultrasonic methods to be inappropriate for masonry. Instead sonic methods have been developed. The principle of operation of the sonic method is that one face of the structure is hit with a hammer and the impact is recorded by an adjacent accelerometer. Another accelerometer on the opposite face of the structure records the arrival of the transmitted compression wave, and the time between transmission and reception is calculated (time domain).
If this procedure is repeated over a regular grid, variations in the transmission time of the compression wave over the structure can be graphed. The transmission time of the compression wave depends on the density of material and the presence of voids through which the wave will not travel. If a void is present, the wave will travel around it and thereby lengthening the transmission time. If the void is large or near either the transmit-side or the receive-side accelerometer, a signal may not be received at all. A study of the variation in transmission times over a structure will indicate changes in density or the presence and extent of voids. The sonic methods have been used with success on structures where both sides of the structures are accessible. However internal construction features such as changes in wall thickness, internal arches and changes in fill material from concrete to rubble or earth are hard to interpret. 7-206 7-206/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
7.31
Accelerometers
An accelerometer produces a continuous electrical record of acceleration in a given direction. It is a small device which is easy to use and calibrate and therefore it is often used for investigating masonry structures. Accelerometers have been used to make comparative measurements of effects of different loading events or the effect of a loading event on different structures. This usage may be able to identify that one loading event causes twice the acceleration of another, but as the acceleration cannot be related to the strength of the structure, not much will have been learned. Accelerometers can be used to obtain displacement by double integration of the acceleration signal. Thereby displacements due to transient loading can be measured in locations which are otherwise not easily accessible. The displacements are not very accurate at low frequencies depending on the characteristics of the particular accelerometer. Furthermore the structures fail because of excessive stress or strain, not excessive acceleration, which is very much a secondorder effect.
7-207 7-207/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
8.
8.1
Economic analysis
General
When a bridge deteriorates very often rehabilitation work is necessary. It may not obvious which repair methods to use, and at what time to carry out the works. In other words, the optimum repair strategy is not obvious. Many factors may influence the choice of repair strategy, such as:
The urgency of repair. The repair cost estimate. The available funds. The traffic hazards caused by damage to the bridges. The inconvenience to the public in case of closure of the bridge (no trains passing the line, ). The inconvenience during the repair works. Other repair works on the same railway line.
An economic analysis is carried out as part of the special inspection in order to select the repair strategy which is economically optimum (most profitable) for the bridge owner or for the society as a whole. It has to be noted, that this analysis is not included in the extended principal inspection. The economic analysis takes into account only those factors, which can be measured in the 'unit': money. The analysis is carried out in order to determine which strategy is the optimum for the bridge, given the premises at the time of decision. When a strategy is selected for the bridge (which includes activities over a period of 20 to 30 years), it does not mean that the decision-makers are stuck with this strategy for the next 20 or 30 years. Circumstances may change, and another strategy may become more profitable. The development of traffic volume, the interest rate, the inflation may change, and the development of damage to the structure may not be as expected. If the decision makers suspect that a chosen strategy is no longer the optimum, a new special inspection has to be carried out, including an economic analysis, in order to determine which strategy is the optimum given the new premises. Obviously, if the chosen strategy implies a replacement or an exhaustive repair of a bridge, there is not much room for a later change of strategy.
8-1 8-1/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
8.2
Identify the relevant repair strategies. Determine the size and distribution of costs (repair, maintenance) connected to each strategy. Calculate the present value of each strategy. Choose the strategy with the lowest present value as the economical optimum strategy.
In connection with a repair work it is often necessary to choose between various strategies. Shall one choose an expensive repair with a long service life or a cheaper repair with a short service life ? Another problem is the time at which the repair should be carried out. Should it be done as soon as possible, can it be deferred, or can it wait until the structure is replaced ? An economic calculation method that can help in such decisions - the present value method - will be described in the following. In the present value calculations, the costs for repairs, operation and maintenance may be calculated year for year within a chosen time-horizon; the timing of each cost is based on the service life of each repair. The annual amounts are then discounted back to the initial year using a given discount rate. In this way the present value of each years expenditure is obtained. By summing the present values, a value for the strategy in question is obtained that can be compared with the corresponding value for other strategies. The strategy for which the cumulative present value is lowest is the economic optimum for the structure considered in isolation. The cumulative present value makes it possible to compare strategies in which the costs are spread over varying periods, as all costs are converted to the initial year. The further in the future a cost falls due, the lower is the present value of the cost. This effect is proportional to the discount rate adopted. To put it simply, the present value is the amount that must be deposited in the bank today to cover a cost that will fall due at the time the repair is carried out. The present value is calculated by: 8-2 8-2/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
In = I
where:
1 (1 + r )n
In I n r is the present value of a cost I in year n is the cost in year n calculated based on the chosen price level (normally the current price) is the number of years until the costs falls due is the discount rate decided by the management authority
The present value calculation is thus carried out in fixed prices (those of the initial year) with a chosen price level and a chosen discount rate. By fixed prices is understood the initial years prices. Inflation and development of wages, taxes, etc. should not be incorporated in the calculations. For an economic evaluation of alternative solutions is the most important parameters: Repair Maintenance strategies equal content The service life of the structural components The time frame for the calculations Time for repair and maintenance Residual value Discount rate
8.2.1
Repair Strategies
Special Inspection: In order to cover the relevant range of strategies, 2 to 4 (in special cases fewer or more) fundamentally different strategies are investigated as part of the special inspection. The strategies normally fall within the following groups:
Total replacement of the whole bridge or the bridge components in question. Thorough (major) repair of the relevant bridge components. Interim repair, and after some time a thorough repair/replacement. Doing nothing. 8-3 8-3/311
Example:
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
A slab/girder reinforced concrete bridge with a thin slab suffers from overload. Structural cracks and spalling of the lower concrete cover have developed in a few deck panels. If nothing is done, one or more panels are expected to fail within 5 years. Relevant strategies may be: A: B: Replacement of the slab, or possibly of the whole superstructure. Strengthening of the slab by pouring a new reinforced top layer on top of the existing slab, with anchors into the existing concrete. Injection of coarse cracks and local replacement of spalled concrete. Replacement of spalled concrete. After approximately 5 years the damage is expected to have reoccurred to the same extent. At that time the repair is repeated. After additionally 5 years the damage is expected to be so serious that the slab has to be replaced. Doing nothing. After 5 years the slab is expected to fail, and it is replaced or repaired.
C:
D:
Strategy A and D may seem very close to each other. But in reality, they are not: In A, a replacement of the slab is planned in advance, and it is possible to make the replacement with very little disturbance to the railway line. In D, we let the slab fail. When it has failed, the bridge must be closed to traffic while investigations and rehabilitation design is carried out. Thus, the bridge may be closed for several months, causing very high inconvenience for the users of the railway. In reality, the bridge must be closed well in advance of any possible failure of the bridge or bridge component. It is not acceptable to allow traffic on the bridge when there is a known risk of failure. Therefore strategy D may require the bridge to be closed well in advance of any possible failure which taken uncertainties into account may be 5-10 years. Extended Principal Inspection: Comprised in the extended principal inspection are general considerations regarding future maintenance activities. This includes a description of the need for major rehabilitation jobs and further inspections. 8.2.2 Service Life The service life of the bridge components in question is estimated for each maintenance strategy. Service life estimations is based on experience of the different maintenance methods used in the maintenance strategies. Estimation of the service life should be based on considerations where ordinary preventive and corrective maintenance is carried out on the component. 8-4 8-4/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Safety considerations can reduce the service life relative to that estimated on the basis of the selected maintenance strategy due to outdated components. Determination of the optimal repair time is associated with the evaluation of the development of damage and their influence on durability and safety of the components. One has to evaluate how fast the component is deteriorating and when the function requirements are no longer fulfilled, i.e. the end of its service life. This evaluation is often complex because it includes an evaluation of the continuing deterioration and the damage time-dependent development. When the possible damage development is evaluated it is important to describe the most possible development. Additional safety factors applied when determinating the damage development on the safe side should not be used when the economic optimal time for repair should be found. It can be wrong (and costly) to repair too early as it is too repair to late. However when there is a risk to the saefy of the bridge and railway users, the design codes safety limit should be used. 8.2.3 Time Frame The time frame is laid down on the basis of the service life of the main repair work necessary to carry out. The same time frame should always be used for the different maintenance strategies to ensure that they are economic comparable. The time frame should be chosen so long that cost that become due after the time frame has only little or no influence on the accumulated net present value. Normally is chosen a time frame of 25 years but it may be longer if the discount rate is low. 8.2.4 Time of Repair Repair time for the different strategies are based on experience. By postponement of a repair work normally the damage extent is increased and will consequently result in an increase in repair costs later. The repair time is therefore based on economic optimal service life of the different components. In that way a minimum present value is reached for each repair strategy. By stipulation of the repair time for each part of the maintenance strategy it should be taken in consideration the general costs such as traffic management and by that collect the different repair works in different time phases. To help choosing the optimal time for execution of a repair service life models and present value calculations may be used. General, where the optimal time for execution is found, the present value of the maintenance strategy will increase if the works is carried out in advance or postponed considering the optimal repair time. 8-5 8-5/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
This means that the economic optimal time for repair may differ depending on the discount rate used in the present value calculations. Due to budgetary limitations it may be necessary to postpone the works. This means that the present value of the maintenance strategy increases due to increases in the extent of repair work (increases in damage etc.). The costs due to increase in repair work should overdue the economic advantage by postponement. 8.2.5 Residual Value As consequence of using the same time frame for the different maintenance strategies it will often be a residual value of a repair work which service life is not reached within the time frame. This residual value should be incorporated in the strategies. The present accumulated value should contain the residual value with opposite sign. 8.2.6 Discount Rate The discount rate should be determined by the management authority and be based on the societies possibility for return of investments. The societies return of investments is depended on the interest rate and the inflation in the economy. The discount rate is normally the interest rate minus the inflation. If the discount rate is high, the pay pack time of investments should be low (i.e. throw away and buy new). Opposite if the discount rate is low it will pay back to invest in components that will have long service life, which means durable components and carry out proper maintenance. The above considerations points in the direction of using different discount rates which may be exemplified by investment in computes (i.e. installations) where the discount rate normally had to be high compared to concrete structures where it had to be low (concrete structures is expected to last for a long period). However to keep things simple normally one discount rate is used covering both installations and structural components. 8.2.7 Sensitivity Analysis The sensitivity analysis should prove the changes in present value due to changes in the different factors that are involved in the different maintenance strategies. A sensitivity analyse should show the increase in the costs due to postponement of the works. Only the parameters that have a significant effect on the different strategies need to be investigated. Parameters that have the same effect will increase or decrease the present value of all the strategies by the same factor. 8-6 8-6/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Be aware that it often pays to postpone a strategy if the damage costs develops with a lower rate than the used discount rate. 8.2.8 Optimum solution special inspection Within each strategy stated in the special inspection report an analysis is carried out in order to determine in detail which works to carry out and at what times. This is called finding the optimum 'solution' within the individual strategy. The relevant repair methods must be considered incl. the extent of the repair. (In the above example, should replacement of the slab include repair of girders?) The repair works must be described in so much detail that the cost estimate can be sufficiently accurate. When determining when to carry out repair works, the development of damage and the discount rate are the dominating factors. If you postpone a repair, normally the extent of damage and thus the repair cost will increase. However, if the annual increase in repair cost is less than the discount rate, it could be profitable to postpone the repair. When performing an economic analysis the bridge owner has to provide you with the value of the discount rate to take into account. If you continue postponing a repair work, the extent of damage (and thus the cost) will in most cases increase slowly and linearly until a point where the cost rises dramatically. This is because at some point the problem can not be solved by the proposed repair method, and a more extensive and expensive method has to be used. (E.g. if damage to the superstructure is allowed to develop, it may at some point be necessary to use interim supports when carrying out the repair). It is very often profitable to postpone repair works until just before this kind of 'jumps' in the repair costs. The present value method is used to determine which solution is optimum within each strategy. (The solution with the lowest present worth is the optimum). When comparing strategies, it is important that all strategies cover the same components of the bridge. If for example strategy 1 comprises repair of the superstructure while strategy 2 is a replacement of the whole bridge, both strategies must include all costs regarding repair and maintenance of the whole bridge. Otherwise the strategies are not comparable.
8-7 8-7/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
9.
9.1
9.2
Text Section
In the following the chapters that the extended principal inspection report must include are described: 9.2.1 Cover Page The cover page of the extended principal inspection report must comprise the following information:
Identification of the bridge owner (e.g. Central Railway) Identification of the bridge (Bridge-ID according to the bridge management system if a management system is used - and bridge name). 'Extended Principal Inspection of ... (the components in question)', e.g. 'Extended Principal Inspection of bearings and girders'. Date of the extended principal inspection. Name of the company performing the extended principal inspection.
9.2.2 Front Page The front page of the extended principal inspection report must comprise the following information:
Identification of the bridge owner (e.g. Central Railway) Identification of the bridge (Bridge-ID according to the bridge management system if a management system is used - and bridge name). 'Extended Principal Inspection of ... (the components in question)', e.g. 'Extended Principal Inspection of bearings and girders'. Date of the extended principal inspection. Identification of the extended principal inspection engineer(s) (name and company). 9-1 9-1/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
9.2.3 Summary The summary must contain all relevant information from the other chapters in a short form. This chapter must include a comprehensive overview of the registrations and conclusions on the damage to the bridge. It must comprise description of the extent of registrations, conclusions on cause and extent of damage, and recommendation for rehabilitations and further activities. However, the summary should not be more than 1-2 pages in length. 9.2.4 Motivation of the extended principal inspection This chapter describes why and by whom the inspection is initiated. It tells which bridge components are the objects of NDT-inspections and which visible damage has been registered. 9.2.5 Background documents This chapter lists the background material that has been available for the inspection, such as: Inventory report and previous principal inspection report. Reports from previous extended principal inspections or special inspections on the same bridge or from similar bridges with similar damage. 'As built' drawings. Materials specifications for steel, concrete, masonry etc. Structural and hydraulics calculations if relevant.
9.2.6 Registrations This chapter describes the registrations from the inspection. For each of the test methods used, the extent and location is described, and a summary of the results is given. The detailed record of all registrations is enclosed in the appendices. 9.2.7 Evaluation of registrations In this chapter the inspection engineer describes the probable deterioration mechanisms and causes of damage based on the registrations. The chapter must also include an estimate of the actual damage of the bridge components investigated. The damage mechanism should be described in detail. This means that in cases of corrosion, 'saline soil' is not sufficient as explanation. You must also explain where the water comes from, how the chlorides have reached the reinforcement, etc. 9-2 9-2/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
For the bridge components of NDT-measurements the extent of damage based on the NDT-measurements is described. Thus, this section includes a summary of the interpretations of the test results from all the NDT-methods used in the inspection and the registrations from the visual inspection. The chapter also includes the condition rating of the components inspected. A brief motivation for each component rating is given based on the previous sections. E.g. bridge deck, piers etc.. Based on the registrations from this inspected a condition rating of each bridge component has been made. The condition rating is a number of 1 to 6 and is based on the following guidelines: 1: 2: 3: 4: 5: 6: 0: A condition which warrants rebuilding / rehabilitation immediately. A condition which requires rebuilding / rehabilitation on a programmed basis. A condition which requires major / special repairs. A condition which requires routine maintenance. A sound condition. Not applicable. Not inspected.
9.2.8 General considerations regarding future maintenance activities This chapter describes the inspection engineer's recommendation of future activities. The need for major rehabilitation jobs and further inspections is included in this chapter. The description does not include budgets for the activities. If there is any doubt of the carrying capacity of the bridge recommendation of calculations must be included in this chapter. Based on the condition rating, the results from the NDT-investigations and the damage type, extent and cause on the selected bridge components a recommendation of an economic analyse must be made in order to select the optimal / best maintenance strategy for the bridge.
9.3
Appendices
The extended principal inspection report comprises those relevant of the following appendices. 9.3.1 A: Background Material This appendix includes copies of the inventory report of the bridge, the previous principal inspection report (if any), previous extended principal inspection reports and special inspection reports regarding the bridge components chosen for NDTinvestigations.
9-3 9-3/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
9.3.2 B: Selected Drawings This appendix includes selected drawings of the bridge itself and the bridge components of which the NDT-investigations are carried out. The selected drawings may be very useful to describe details of the design of the bridge. 9.3.3 C: Visual Inspection This appendix includes the registrations from the visual inspection of all the bridge components included in the inspection. General orientation of the bridge and the bridge components under investigation, numbering of components and damage pattern are most conveniently shown on sketches. An overview sketch of the whole structure is often suited to register the extent of damage (which columns have spalling of cover, which girders have shear and flexural cracks, etc.). The appendix should include photo pages for the photographs taken during the inspection. The photographs should always include the following:
Overview photographs showing the approach and surface, and the elevation of the bridge. Photographs showing the general design of the bridge components under investigation. Photographs describing the damage to the structure. Overview as well as closeup photographs. Photogrpahs showing details of the registrations, e.g. exposed reinforcement, corrosion of steel surfaces, deteriorated stone of masonry structures, etc..
9.3.4 D: NDT-method no. 1 Depending on the complexity of the NDT-method a general description of the principles of the method is described in this appendix. This appendix includes the registrations from one of the NDT-methods used in this extended principal inspection. The appendix should include sketches of the areas of measurements and of the measuring grid if used e.g. for HCP, Impact-Echo, Impulse Response (sMASH) etc. The appendix should also include relevant photographs related to the NDT-investigation (of break-ups etc.). Always note the dimensions of the bridge component in question. (Diameter of column; width, depth, spacing and length of girders, etc.). Mapping of HCP (Half-cell Potential measurements) readings is shown on sketches. Mapping of sMASH (impulse response measurements) readings is shown on sketches.
9-4 9-4/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Mapping of Impact-Echo readings is shown on sketches. 9.3.5 E - ?: NDT-method no. 2 - ? This appendix includes registrations from another NDT-method. The content of this appendix is similar to the content mentioned in section 9.3.4. 9.3.6 F.. Other In some cases it is appropriate to include other appendices than the 'standard' ones. These are numbered 'F', 'G'.....
9-5 9-5/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
10.
10.1
10.2
Text Section
In the following the chapters that the special inspection report must include are described: 10.2.1 Cover Page The cover page of the special inspection report must comprise the following information:
Identification of the bridge owner (e.g. Central Railway) Identification of the bridge (Bridge-ID according to the bridge management system if a management system is used - and bridge name). 'Special inspection of ... (the components in question)', e.g. 'Special inspection of bearings and girders'. Date of the special inspection. Name of the company performing the special inspection.
10.2.2 Front Page The front page of the special inspection report must comprise the following information:
Identification of the bridge owner (e.g. Central Railway) Identification of the bridge (Bridge-ID according to the bridge management system if a management system is used - and bridge name). 'Special inspection of ... (the components in question)', e.g. 'Special inspection of bearings and girders'. Date of the special inspection. Identification of the special inspection engineer(s) (name and company).
10-1 10-1/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
10.2.3 Summary The summary must contain all relevant information from the other chapters in a short form. This chapter must include a comprehensive overview of the registrations and conclusions on the damage to the bridge. It must comprise description of the extent of registrations, conclusions on cause and extent of damage, and the proposed repair strategy including cost estimate and time schedule. However, the summary should not be more than 2-3 pages in length. 10.2.4 Motivation of the special inspection This chapter describes why and by whom the inspection is initiated. It tells which bridge components are the objects of the inspection and which visible damage has been registered. 10.2.5 Background documents This section lists the background material that has been available for the inspection, such as:
Inventory report and previous principal inspection report. Reports from previous extended principal inspections or special inspections on the same bridge or from similar bridges with similar damage. 'As built' drawings. Materials specifications for steel, concrete and masonry. Structural and hydraulics calculations if relevant.
10.2.6 Registrations This chapter describes the registrations. On the basis of the visual inspection and prior knowledge the structure may be divided into homogeneous areas. A homogenous area is defined as an area where the parameters affecting the deterioration and the deterioration itself of the structure exhibits only a random variation. For each of the homogeneous areas a damage hypothesis is prepared. These hypothesis are described in this section. For each of the test methods used, the extent and location is described, and a summary of the results is given. The detailed record of all registrations is enclosed in the appendices. 10.2.7 Evaluation of registrations This section includes an interpretation of the test results from the NDTinvestigations. E.g. for chloride-measurements: do the measurements show risk of chloride initiated corrosion of the reinforcement are the values of the chloride content larger than the critical chloride content in the depth of reinforcement. 10-2 10-2/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
In this chapter the special inspection engineer describes the probable deterioration mechanisms and causes of damage, based on the registrations. The chapter must also include an estimate of the actual damage of the bridge components investigated. It should also include a description of the expected development of damage if no action is taken. The damage mechanism should be described in detail. This means that in cases of corrosion, 'saline soil' is not sufficient as explanation. You must also explain where the water comes from, how the chlorides have reached the reinforcement, etc. It is also important to explain the differences in damage appearance: Why are some columns damaged while others are undamaged, why is only the centre girder cracked, etc. 10.2.8 Repair strategies This chapter describes the relevant repair strategies for the bridge. The description of each strategy should comprise:
A general description of the 'idea' of the strategy, e.g. 'Replacement of the whole bridge', 'Interim repair, followed by major rehabilitation after 10 years'. List of all activities with year and cost estimate, e.g.: Activity Interim repair of bridge deck Replacement of bridge deck Year 2005 2015 Cost 5 mio. Rs. 45 mio. Rs.
If the contents and extent of the activities are not obvious, they should be detailed, e.g: 'Interim repair of deck comprises repair of honeycombs at 8 locations, and cement mortar injection of approximately 50 meters of cracks'. 'Replacement of deck comprises replacement of the deck slab on the whole bridge, including expansion joints, edge beams, ballast and tracks. The existing girders are re-used'. Description of possible disturbance to the traffic. 'Present value' of the strategy, calculated following the 'present value method'. Remember that 'doing nothing' may very well be one of the possible strategies. This strategy must be examined as well. This strategy will have no repair or maintenance costs, but it may imply severe inconvenience for the users of the railway. 10-3 10-3/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
10.2.9 Recommendation of follow-up activities This chapter describes the special inspection engineer's recommendation of future activities. Normally the recommendation will be to carry out the repair strategy with the lowest present value, as this should be the optimum thing to do. However, in some cases the recommendation may be to carry out further, more detailed investigations, or to monitor the development of damage for some time before making final conclusions on the optimal repair of individual bridge components or the bridge as a whole. If there is any doubt of the carrying capacity of the bridge recommendation of calculations must be included in this chapter.
10.3
Appendices
The special inspection report comprises those relevant of the following appendices. 10.3.1 A: Background Material This appendix includes copies of the inventory of the bridge, the previous principal inspection report, previous extended principal inspection reports and special inspection reports regarding the bridge components of the special inspection. 10.3.2 B: Selected Drawings This appendix includes selected drawings of the bridge itself and of the bridge components for the special inspection. 10.3.3 C: Visual Inspection This appendix includes the registrations from the visual inspection of the bridge components included in the special inspection. General orientation of the bridge and the bridge components under investigation, numbering of components and damage pattern are most conveniently shown on sketches. An overview sketch of the whole structure is often suited to register the extent of damage (which columns have spalling of cover, which girders have shear and flexural cracks, etc.). The appendix should include photo pages for the photos taken during the inspection. The photos should always include the following:
Overview photos showing the approach and surface, and the elevation of the bridge. Photos showing the general design of the bridge components under investigation. Photos describing the damage to the structure. Overview as well as close-up photos. 10-4 10-4/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
Photos showing details of the registrations, e.g. exposed reinforcement, corrosion of steel surfaces, deteriorated stone of masonry structures, etc..
10.3.4 D: NDT-method No. 1 Depending on the complexity of the NDT-method a general description of the principles of the method is described in this appendix. This appendix includes the registrations from one of the NDT-methods used in the special inspection. The appendix should include sketches of the areas of measurements and of the measuring grid if used e.g. for HCP, Impact-Echo, Impulse Response (sMASH) etc. The appendix should also include relevant photos related to the NDT-investigation (of break-ups etc.). Always note the dimensions of the bridge component in question. (Diameter of column; width, depth, spacing and length of girders, etc.). Mapping of HCP (Half Cell Potential measurements) readings is shown on sketches. Mapping of sMASH (impulse response measurements) readings is shown on sketches. Mapping of Impact-Echo readings is shown on sketches. 10.3.5 E - ?: NDT-method No. 2 - ? This appendix includes registrations from another NDT-method. The content of this appendix is similar to the content mentioned in section 10.3.4. 10.3.6 F: Economic analysis This appendix may contain present value calculations of the strategies under investigation. 10.3.7 G.. Other In some cases it is appropriate to include other appendices than the 'standard' ones. These are numbered 'H', 'I'.....
10-5 10-5/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
11.
References
[1] CEB: Durable Concrete Structures. Design Guide. Second Edition, reprint 1997. Geiker, M.: Durability of concrete chloride induced corrosion. Course 59203, Danish Technical University, 2000. Larsen, E. S.: Service Life Prediction of Cementitious Materials. SBI Report 221, 1992. Larsen, E. S.: Inspection of Structures Special Inspection Compendium for support and Inspiration (in Danish). VEJ-EU, 1997. Mattsson, E.: Electro chemistry and corrosion (in Swedish). 2end edition, Corrosion Institute, Stockholm 1984. Nielsen, A.; Eeg, R. and Sorensen, H.: Building Materials Metal (in Danish) Polyteknisk Forlag 1998. Sowden A.M. (1990). The Maintenance of Brick and Stone Masonry structures. ISBN: 0-419-14930-9. E. & F.N. Spon. First Edition Federal Lands Highway Program. Surface Nondestructive Test (NDT) Methods http://www.cflhd.gov/agm/engApplications/BridgeSystemSubstructure/211Sur faceNDTMethods.htm Departments of the Army. Seismic Design Guidelines for Upgrading Existing Buildings, appendix E. http://www.usace.army.mil/inet/usacedocs/armytm/tm5-809-10-2/app-e.pdf
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10] Ron Grieve. Non-destructive Testing of Concrete and Masonry Buildings. The construction Specifier, October 2005
11-1 11-1/311
Ref. 5721063-07_L002_Ver4_NDT_manual.doc
APPENDIX A
Handout of Slides from Classroom Training in NDT-Methods
5721063-07_L010-VerA_NDT_manual_App_frontpages.doc
APPENDIX A1
Introduction to the Class Room Training in NDT and UWI
5721063-07_L010-VerA_NDT_manual_App_frontpages.doc
NDT-Course
Agenda
0. Welcome by Central Railway, LTR and Ramboll 1. Presentation of the lecturers by AKN 2. Presentation of the participants by AKN 3. Presentation of the Ramboll Group by AKN 4. Lecture program for NDT and UWI by LTP 5. General introduction to deterioration mechanisms by LTP 6. General introduction to systematic operation and maintenance by LTP 7. Special inspection by LTP 8. Structural assessment Case by LTP
Non Destructive Testing and Underwater Inspection - 8 February, 2006 2
Slide
India 2005/2006
NDT-Course