 ·Table of Contents ·Nuclear Industry | From Technical Specification to Examination results: The Story of Qualification for Belgian RPV InspectionD. Couplet,Tractebel Energy Engineering, Av. Ariane, 7 1200 Brussels ; BELGIUM T. Françoise, Intercontrôle, 13, Rue du Capricorne - SILIC 433 94583 Rungis ; FRANCE Contact |
The whole qualification process, performed in an industrial context, has shown the importance of various aspects : a review of the construction files and of previous ISI results; a detailed definition of the inspection objectives (defects to be detected and level of characterization); a systematic verification of the NDE procedure performance, for each objective or situation; the feedback from site experience on procedure and qualification optimization; and a multi-disciplinary approach.
The paper presents how the initial inspection requirements and the methodology of qualification were specified, by combining the rules of the ASME XI code and the European Methodology for Qualification. It shows that the NDE procedures have been systematically improved during the qualification process and how the procedure performance and limitations have been determined and demonstrated by an adequate combination of tests on blocks and technical justification.
To conclude, the paper also states the lessons learnt from the first on-site experience of Reactor Pressure Vessel ISI with qualified NDE procedures.
In 1997, the Belgian utility (Electrabel/Tractebel) decided to launch a call for tender for In-service Inspection of Reactor Pressure Vessels. The contract covers all 7 Belgian units, during the present 10-years inspection intervals. The scope is the inspection of all items of the Reactor Pressure Vessel, as required by the edition 92 of the ASME XI Code, which applies in Belgium.
When developing the strategy and the specification for this inspection and for the qualification of the Non Destructive Examination system, Tractebel considered two main inputs :
Belgian requirements leave the possibility of applying any NDE system and qualification process, but set some criteria and rules derived from ASME XI and ENIQ documents. ISI and qualification objectives are defined by reference to ASME requirements and based on past experience, on some physical reasoning and on fracture mechanics analysis. Qualification dossier and technical justification, as defined by the European Methodology, are required for NDE procedures. A practical trial must be performed in most cases. It must be demonstrated to be at least equivalent to appendix VIII "performance demonstration" in giving evidence that the technique is fit for the purpose.
Besides the NDE aspects (ISI and qualification objectives), many other requirements were defined in the technical specification : on site organization, industrial efficiency and profitability, duration of the inspection,...The best solution had to take all these aspects into consideration.
A contract was signed with Intercontrôle, and was the starting point of a close and continuous collaboration to implement the requirements of the specification.
Though the ASME Code remains the regulatory reference, many adaptations were brought to implement the qualification process. With the permanent goal of achieving the most effective technical solution, in relation to the safety and industrial objectives set in the technical specification, some flexibility was left in the process. However, all particular requirements or deviations were justified and documented, and were approved by the Safety Authorities; the whole qualification process was also controlled and followed up by the Safety Authorities.
This way of working leads to an approach and a practical implementation that allows to reconcile the safety concerns and the economical requirements of the Nuclear Power Plants.
Definition of ISI objectives
The ISI objectives clarify and quantify the reason for examination : what must be detected ? where ? how must it be characterized ?
The ISI objectives constitute an essential input to the development of the NDE procedure and to the qualification process. The European Methodology for Qualification puts emphasis on a correct definition of ISI objectives, as this will orient the efforts to be developed as well as the value of the inspection results.
The ASME Code does not explicitly define ISI objectives. The inspection is expected to detect any type of discontinuity, in a well defined volume of material. However, there are some precise criteria to accept or not the detected defects (acceptance criteria). Only the unacceptable defects must be treated further : follow-up by successive examinations and justification by analytical evaluation.
The acceptable defects are considered to have no impact on the structural integrity of the vessel. They should theoretically also be detected, but we may accept that this detection is not 100% reliable (what would be impossible) since there is no risk associated with the presence of these defects.
The basic rule that defines ISI objective is then "all defects or groups of defects that would be unacceptable according to ASME XI acceptance standards must be detected and correctly characterized (position, size)".
This general requirement was however reviewed by a group of experts in various disciplines : structural integrity, degradation mechanism, welding procedures, fabrication process, equipment supply.
This review is based on a systematic analysis, using different assessment tools (matrices, categories, qualitative evaluation of danger and probability of appearance,...). All type of defects (planar and volumetric fabrication defects, thermal and mechanical fatigue cracks, corrosion induced cracks) are envisaged in all items of the vessel (items from the ASME XI examination program). Each type of defect is classified according to whether it has already appeared, it could appear (according to international experience) or is highly unlikely to appear. The potential impact of the defect on safety (vessel integrity) is qualitatively assessed. From there, the requirements are defined concerning each type of defect, for each item : the defect does not need to be detected or it should be detected or it must be detected with a qualified NDE technique.
This review led, for some items, to limitations of the types of defect to search for, depending on some characteristics of the defect, such as its orientation or location.
For example, there is no need to search for axial defects in studs and ligaments between threaded holes, whereas it is required in nuts.
Through this approach, it has been possible to establish realistic ISI objectives resulting from adequate safety considerations and leading to economically acceptable solutions for examination and qualification.
The ISI objectives were defined, but there were no requirements about the NDE techniques to use for the examination. The choice was left to the contractor to develop and propose the more adequate technique. The verification of conformity of the NDE system to the requirements of the technical specification (are the ISI objectives reached by the proposed system ?) is achieved through the qualification.
Besides the requirements in terms of detection of defects, the specification imposes criteria for on-site efficiency : duration of the inspection, stability and flexibility of the system with respect to environmental conditions (radiation, temperature,...), quality of the reporting, etc. This is necessary to ensure that the inspection results will be exploitable in an industrial context, and that the efforts, including their financial impact, are optimized in function of the priorities concerning safety issues.
Methodology for qualification
The effectiveness of inspection in improving safety is related to the performance of the applied NDE techniques and the correct knowledge of this performance.
Assessing this performance is the very purpose of the qualification.
ASME XI gives in appendix VIII instructions on how to conduct a performance demonstration : qualification objectives are defined, assessment criteria and requirements for the qualification process are given. The performance demonstration is an efficient way to assess a NDE system. On the other hand, being very prescriptive, appendix VIII does not always fulfil all Belgian requirements. In some cases, strict application of appendix VIII might be very expensive; in other cases, the requirements are considered not adequate to Belgian ISI objectives, or technically insufficient, and some qualifications are not covered by appendix VIII. Therefore, the utility has searched for complements or alternatives in the European Methodology.
In accordance with this European Methodology, the Belgian utility maintains generally the performance demonstration or practical trials as an important part of the qualification process. However, the use of technical justification becomes also a requirement to assess the validity of the performance demonstration or of the practical trials, to complete or extend the qualification and to optimize the economical aspects of the process. For most NDE systems, a complete qualification dossier, as described in European Methodology, is required.
Types of defects.
In order to define the qualification requirements, the review of defects by the group of experts is used again. For every item of the RPV, the types of defect to be considered for qualification are defined.
The existence of this type of defect is considered as possible, though it has not necessary appeared (or has been detected) at this time. The presence of these defects increases significantly the risk to structural integrity. The size of these defects is such that they would exceed the acceptance standards (IWB-3000) of ASME XI for ISI (or Pre-Service Inspection for manufacturing defects).
The qualification process shall focus on these defects : evidence must be given that the NDE method will detect and correctly size them. Some examples of defects are :
Moreover, the NDE method shall also search for other unspecified defects that could have an influence on structural integrity, as required by ISI objectives (preventive action). For these unspecified defects, a formal practical exercise (in order to demonstrate the performance of the technique) cannot be conducted, and the general ability of the method to detect these defects will be explained by physical reasoning.
Qualification requirements.
Different levels of qualification are considered. These depend on the requirements of the ASME Code for qualification (or absence of requirements), on the applied NDE technique, and on the safety significance of the examined item.
The following types of requirements are defined :
When a qualification dossier is required, The ENIQ Methodology and ENIQ Recommended Practices will be used as guidelines.
Any qualification dossier must also contain a "qualification procedure" that explains how the final objective of qualification will be met : what information, experiments, tests, studies, simulations are available or will be obtained and how; what practical trials (if performance demonstration is needed) have been or will be organized and according to which rules.
Results of earlier practical trials may be used, if they are well documented, as support to demonstrate the performance of the technique. The results must however be used in relation to the specific objectives of Belgian inspections.
The technical justification shall identify all essential parameters, and for each of them, describe the difference between tests (and trials) and ISI situation, and analyze its influence. Use of mathematical modeling is appropriate to cover all values of the essential parameters.
This approach helps to reduce the number (and formalism) of practical trials (perhaps limited to the use of past results), but requires an important work of reasoning and systematic justification of announced performance. This is more rigorous than only statistical results on test blocks, because the NDE procedure performance is assessed for all ISI objectives and conditions that could be encountered at sites, as defined in the technical specification.
Specific requirements
The examination of threads in flange is required by the specification in order to detect fatigue cracks. But qualification requirements do not exist for this item in appendix VIII of Section XI of the ASME Code. Nevertheless, in order to define qualification objectives, it was chosen to follow Supplement 8 - qualification requirements for bolts and studs - requirements, and to adapt it to the concerned area (type 1 of requirement for qualification dossier, as defined above).
So it was decided to demonstrate that the examination procedure enables to detect circumferentially oriented notches located at the thread root with maximum depths and reflective areas as specified in Table 1 below :
| Bolt or stud size | Depth (in.) | Reflective area (sq. in.) |
| Greater than 4 inches diameter | 0.157 | 0.059 |
| 2 inches diameter and greater, but not over 4 inches diameter | 0.107 | 0.027 |
| Table 1: Maximum Notch Dimension (from supplement 8) | ||
Description of NDE Technique
The "flange tool" of the MIS (Machine for in-service Inspection) performs examination of threads in flange. One focused transducer placed on this tool scans the required area from the flange surface.
All reflectors which produce a response equal to or greater than -21 dB of the maximum of the Distance-Amplitude-Curve (DAC) are recorded. According to the depth, a threshold of notation is applied.
Description of the qualification process
Because of various designs of the seven pressure reactors in Belgium, threads in flange areas have many different geometrical characteristics. It was unthinkable to manufacture as many blocks as configurations. We decided to manufacture two simple blocks that cover the whole range of configurations. Main characteristics of first block are stud hole of 155 mm diameter and thread of 3.175 mm, main characteristics of second block are stud hole of 180 mm diameter and thread of 4 mm.
One of the first tasks of the qualification was to produce technical justification to study all configurations and all important parameters.
The technical justification contained :
| COMPONENT | |||||
| Parameter | Definition | Origin of information | Nominal value | Tolerance | Justification |
| Stud hole | Diameter of stud hole | As built drawings | Nuclear Power Plant (NPP) A : Æ= 180 mm NPP B : Æ= 155 mm NPP C : Æ= 153 mm | - | Diameter of qualification blocks : 155 mm and 180 mm. |
| Mating surface | Influence of mating surface | As built drawings | 13 mm or 25 mm or 38 mm | - | 1rst step : No mating surface is machined on qualification blocks. 2nd step : Mating surface is machined on qualification blocks. |
| DEFECT TO BE DETECTED | |||||
| Parameter | Definition | Origin of information | Nominal value | Tolerance | Justification |
| Type of defect | Type of defect to be detected | Belgian requirements | Fatigue cracks | - | Defects in qualification blocks are notches as specified in supplement 8 of appendix VIII. |
| Orientation | Orientation of defects | Belgian requirements | Perpendicular to axis | 0° | Notches in qualification blocks are perpendicular to axis. |
| Location | Location of defects | Belgian requirements | Thread root | - | Notches in qualification blocks are located at thread root. |
| ENVIRONMENT | |||||
| Parameter | Definition | Origin of information | Nominal value | Tolerance | Justification |
| Water temperature | Temperature of water and steel during examination | Belgian specifications | 20o | 20°C (r)60°C | This parameter is justified by modeling. (see Table 3 hereafter) Conclusion : Temperature has little influence on defects detection level. |
| Table 2: Example of Treatment of Essential Parameters | |||||
| Temperature | Depth of defects (mm.) | ||||
| 20 | 65 | 110 | 195 | 330 | |
| T = 20°C Amplitude of detection (dB) : | +1 | +7 | +6 | -3 | -11 |
| T = 60°C Amplitude of detection (dB): | +1 | +8 | +7 | -1 | -9 |
| Table 3: Results of Modeling Water Temperature Influence | |||||
During the first step of qualification tests on blocks, the presence of cladding over the mating surface, the level difference between mating surface and flange surface and the bore hole diameter were not considered. A systematic review of essential parameters was then performed within the technical justification. It confirmed the influence of these factors and thus, the necessity to take them into account at this point of the qualification.
These parameters are very difficult to treat by modeling and original blocks did not reproduce them. In the second step of qualification, one of the two blocks was modified in order to add a bore hole and a mating surface. These modifications were quite simple and new qualification tests were performed. They showed that initial procedure used in the first qualification tests needed only very simple adaptations to pass the other qualification tests.
Evaluation
The qualification of a procedure requires rigorous work, but experience shows that results will be improved if some parameters can be adapted (with justified reasons) during the qualification period in order to obtain an optimized procedure.
Therefore, one must expect to discover new information during the inspection, and use this feedback for final adaptation of the procedure. The integration of feedback from field experience is indeed a strong recommendation of the European Methodology for Qualification.
On-site results
In 1999, some of the new qualified procedures were applied for examination of the Reactor Pressure Vessels of Doel and Tihange.
These optimized procedures are successful regarding industrial efficiency. The inspection time is reduced by about 20% in comparison with previous procedures (conform to edition 80 of ASME XI). Inspection costs are reduced.
The detection capability has also been enhanced. Some of the defects specified as ISI objectives present a challenge for NDE. In order to qualify for these defects, the procedure sensitivity had to be relatively high. This means that a lot of other defects, easier to search, are also detected, including defects with no impact on structural integrity (such as small volumetric fabrication defects).
Many indications were reported in some welds; most of them had not been detected with previous NDE techniques. For a correct interpretation of these results it was necessary to review the fabrication files. The existence of small acceptable fabrication defects (such as porosities clusters) was clearly shown by the results from radiographic examination. Mathematical modeling was used to simulate the signal obtained from such acceptable fabrication defects, and these signals could be correlated with the results of the 1999 inspection. Simulations and information contained in the technical justification of the procedure could explain why these indications were not detected with the previous NDE procedures. Without this work, all indications noted for the first time in 1999 would have been declared "in-service induced".
The presence of acceptable volumetric fabrication defects has no consequence on the structural integrity of the vessel. It must be considered as a "normal state" of the weld, considering the technical limitation of the welding technologies.
However, the ASME code (section XI - IWA 3300) requires to follow some rules for characterization. All reported indications must be characterized as "planar" flaws, by considering a projection in case of volumetric defects. Discontinuous indications must be grouped when close to each other and considered as one single planar flaw for evaluation. A subsurface indication close to the surface shall also be considered as a surface flaw. Limitations on the aspect ratio impose in some cases to consider flaws much longer than they really are. All these conservative rules can be applied easily when only a few major defects are reported; when the results of the examination are more detailed and complete, due to the required high sensitivity of the NDE procedure, the strict application of these flaw characterization rules leads to formally consider very large defects that are not acceptable according to "acceptance standards" of ASME and difficult to justify. It must be stressed that the so considered defects are an artificial construction, sometimes quite far from the reality.
Feedback on procedure and on qualification
The field experience confirmed the high sensitivity of the technique, and its ability to detect the defects aimed at. However, the tools used for evaluation of results (NDE analysis and ASME rules) did not allow an easy interpretation of the information. There was therefore a necessity to adapt these tools or find alternatives.
Based on the better knowledge of the weld actual conditions, the ISI and qualification requirements were refined. The detection of defects significant for structural integrity is still required, but these defects are defined more accurately: orientation, size in function of the position in the thickness of the weld and of the aspect ratio. Typical "reference defects" are given as input for the adaptation and complementary qualification of the procedure. The acceptable size of the defects is, in most cases, calculated through a "defect tolerance analysis". This determination is performed according to the same principles that have led to the development of the ASME "acceptance standards", but the calculation and results are adapted to the specific case of the Belgian RPVs (materials, stresses, etc...).
Moreover, some new requirements are defined for classification of the detected defects. The NDE procedure must be able to identify and eliminate innocuous indications (such as volumetric fabrication defects, acceptable according to ASME III construction code). For these indications, the ASME characterization rules will not be followed, giving a clearer picture of the indication of interest to assess structural integrity.
The NDE analysis must also report separately defects whose presence in the weld cannot be ignored, but which have no direct impact on structural integrity.
Each class of defects is precisely defined, in function of the characteristics of the defects : size, planar or volumetric, position,... For each class, there are different requirements regarding qualification of the procedure, detection and sizing capabilities, reporting, grouping of the defects and analytical evaluation.
The NDE procedures are then adapted to fulfil these objectives. The acquisition and the setting of the sensitivity are not modified when the qualification showed their necessity for detection of the "worst case defect". Different analysis tools are developed to determine more accurately the characteristics of the indications (distinction between planar and volumetric, proximity between defects or to the surface, etc.). Sorting criteria, based on clear characteristics, are designed to establish a correct and quick classification of all detected indications.
The aim is to identify rapidly the defects significant to structural integrity, if any, and to give a useful characterization of the state of the inspected item. This information is then forwarded on-line to structural integrity experts, who will be able to define the operational conditions of the reactor in function of the integrity of the vessel. All these operations must be performed in a very short time, as the results are needed before restarting the reactor after refueling.
The qualification of the modified NDE procedure is performed through technical justification, as proposed in ENIQ Recommended Practices. New simulation and analysis of available results from the initial qualification are used to validate the modifications.
In parallel with NDE procedure optimization, a strategy is also studied to use alternatives to the ASME rules of grouping indications and aspect ratio limitation. Calculation tools, based on finite element analysis, are developed to provide a defect justification tailored to each specific situation, avoiding the conservatism of a general solution.
Lessons learnt
Through this experience, we have learnt that NDE procedure development and qualification is a necessary but long process, which must include feedback from industrial application to achieve optimization of the technique.
The better and the more complete is the input information concerning the component, the more efficient will be the procedure. Unfortunately, a perfect knowledge of the component is not possible; this information often comes from the examination itself.
The initial definition of ISI objectives must try to cover all potential situations, within reasonable effort. But again, only the field experience can confirm the relevance of the objectives and the potential need for adaptation. Setting from the beginning too sophisticated objectives may lead to unnecessary costs in procedure development and qualification. However, the ISI and qualification objectives must not only consider what must be detected (for safety considerations), but also set some requirements for false call or for a correct treatment of indications from non-significant defects. This is essential to obtain information usable for the assessment of RPV integrity.
Finally, it was found necessary to adapt the evaluation standard to the performance of the NDE technique. Analytical evaluation of defects and qualification of the NDE procedure are linked and must be approached in a global strategy, by a multi-disciplinary team.
From the experience we have gained through this work, as main conclusion we wish to stress again the key features of an inspection with qualified NDE technique :
| © AIPnD , created by NDT.net | |Home| |Top| |