There are two primary types of in-service damage; inlet rolled joint fuel bundle bearing pad fretting and debris fretting. Inlet rolled joint fretting affects only the 13 fuel bundle channel design used in the Bruce and Darlington reactors. With this design there is interaction between the pressure tube at the inlet rolled joint burnish mark and the fuel bundle bearing pads. Inspections have generated sufficient information to characterize the severity and distribution of the inlet fretting in these units. The combination of research data and data obtained from removed tubes make it possible to disposition all tubes with such flaws for continued service.

Inadequate cleanup after construction and commissioning and isolated operational incidents have led to debris entering the Heat Transport System. Debris, which is carried around the circuit by the coolant and then trapped in the fuel bundles, or between the bundles and the tubes, can result in wear of both the fuel sheaths and the pressure tubes. In many reactors, full length volumetric inspections indicate that the level of debris fretting is unit dependent. Because of the random nature of this fretting mechanism, it is difficult to predict the location and severity of potential fretting. To ensure that debris fretting that may exist in a particular reactor core will not result in an unacceptably high probability of tubes being susceptible to DHC, the following programmes are used to complement the limited volumetric inspections.

• Deuterium monitoring programme. This is important because debris frets in the body of the tube are not considered to be an integrity concern if the hydrogen equivalent concentrations in the pressure tubes remain below the terminal solid solubility limit at normal operating temperatures at the flaw tip.

• Additional volumetric inspections to assess the distribution of debris fret geometries in a core.

• Probabilistic core assessments to establish the probability of initiating DHC from this mechanism.

• Pressure — Temperature limits to avoid full pressurization of the pressure tubes at conditions when DHC can occur, i. e., when the hydrogen concentration exceeds the terminal solid solubility limit and to ensure that there are adequate margins against fracture at all operating temperatures.

• Fuel failure monitoring. Debris can cause wear of the fuel sheaths and can be an indicator of wear of the pressure tube.

At most multi-unit CANDU NPPs, the PLiM programme is implemented as part of the annual business planning, work management and asset management processes. There is a yearly review of condition assessments to identify actions for the work plan (maintenance and inspection, 1 year time-frame), business plan (1 to 5 year time-frame), and the strategic plan (> 5 year time­frame). Fuel channel, feeder and steam generator life cycle plans are reviewed annually to identify required inspections and input to the generation plan.

III.1. PLANNING FOR LONG TERM OPERATION

The philosophy at the multi-unit CANDU NPPs has been to preserve the option for LTO, through ageing management of major components. The LTO decision will be made as required, given the lead-time for procurement of major equipment, and potential overlap of units in refurbishment.

Design life for the plant and major components was 30 years. Service life, during which components operate safety and reliably, may exceed the design life. Alternatively components may be replaced at 30 years to provide a longer service life for the plant as a whole. Therefore, unless there are critical components that cannot be physically replaced or refurbished at their end of life, the optimum life for the plant may be based on economics rather than technical issues.

The most expensive component that requires replacement in order to extend plant life is the pressure tubes. Pressure tube replacement or large scale fuel channel replacement (LSFCR) is required at about 30 years, due to ageing degradation similar to what was described in Section 3 for single unit HWRs. LSFCR was accomplished successfully at Pickering A in the 1990s, and there have been developments since then in reducing the cost and time required.

The second most expensive component to replace for LTO of the multi-unit CANDUs is the steam generators. The current target life for steam generators at the older sites (Pickering, Bruce) is to reach pressure tube end of life, so that replacement of steam generators and pressure tubes could be accomplished in the same outage. These two plants have Inconel 600 as the steam generator tube material. Worldwide experience with Inconel 600 steam generator tubes has been that a high percentage of them were subject to stress corrosion cracking so severe that replacement around 20 years service life was required.

The multi-unit CANDU plants have all implemented steam generator life cycle management plans to inspect, monitor and mitigate steam generator degradation to achieve the design life. However, steam generator replacement will be required for life extension, except at Darlington. Darlington is the newest multi-unit CANDU, and has Inconel 800 SG tubes that are not expected to require replacement for LTO.

Bulk feeder replacement is another significant activity that may be required for life extension, depending on the effectiveness of current and planned activities to mitigate feeder wall thinning due to Flow Accelerated Corrosion. Replacing feeders during the LSFCR may actually reduce the duration of the LSFCR by improving access to the fuel channels.

Annual PLiM plans are produced as part of overall utility long term strategic planning (OPG, input from Bruce Power is needed). These life cycle plans contain alternatives for long term operation, including shutdown at design life, or retube and extend life. The main elements of the LCPs are listed below:

• Extended life is to the end-of-life of the replacement pressure tubes (additional 30 years);

• Cost of retube, replacement of SG, feeder replacement and other refurbishment is estimated and input to asset evaluation for each plant;

• Benchmarks are used for change in maintenance costs with age; and

• Environmental, safety and emerging regulatory and licensing issues will need to be considered in any life extension decision.

Alternative operating scenarios are analyzed, using discounted cash flow, to determine the alternative that creates the maximum value for the corporation. Uncertainties in the cost of licensing issues, refurbishment (balance of plant, retube costs), electricity cost, capacity factor, etc. are captured through a sensitivity analysis.

Equipment lifetime estimates and repair/replacement costs are required as input to the life extension decision. Equipment obsolescence is to be addressed by re-engineering and spare parts strategies (control, instrumentation equipment obsolescence and computer system upgrades). The cost of other balance of plant refurbishment activities as described below also needs to be factored into the LTO decision.

• In cooperation with CANDU utilities, developed a comprehensive CANDU plant life management programme, including strategy, processes, and procedures.

• Completed many ageing assessments of CSSCs and assisting CANDU utilities with implementation, including training.

• The CANDU PLiM methodology has been successfully applied to the life extension project planning for the CANDU 6 plants at the Centre Nucleaire Gentilly 2 in Quebec, and for the Point Lepreau Generating Station in New Brunswick.

• Encourageing younger CANDU plants to start PLiM early in life (such as at Cernavoda Unit 1 and Qinshan Units 1 & 2)

• Maintaining Technology Watch (emerging issues) and operational experience feedback databases

• In addition to COG-funded programmes, performing additional R&D programmes in various areas important to CANDU PLiM.

• Continuing to evolve a highly integrated and comprehensive approach to various PLiM activities, with supporting processes and databases. The methodology foresees the application of PLiM to all stages of the plant life (from design to extended service operation).

• Applying PLiM technology and processes to its own nuclear facilities, such as to the life extension of the NRU research reactor.

• Experience from PLiM programmes on operational reactors is fed back into the new CANDU products. The PLiM technology and knowledge is being converted into enhancements of new reactor designs (such as ACR). For instance, AECL’s PLiM group has been assisting the ACR design team perform detailed maintenance assessments during the design phase.

Canadian Nuclear Safety Commission (CNSC)

The CNSC activities as related specifically to PLiM are in the main text of this TECDOC. Conclusions

In Canada, plant life management continues to receive considerable attention. Efforts are underway to maximize the technical and economic benefits from PLiM work and the benefits derived from the detailed understanding of ageing behaviour at the plants. There are on-going improvements to the technical assessments of SSC ageing and the processes used to perform these assessments. There are also improvements in the linkage to current operational performance and to the plant programmes (that are the primary means to manage ageing), to the business planning/asset management and to the supporting R&D programme. AECL is assisting CANDU owner/operators with their PLiM programmes; using PLiM on its own research reactor; and applying the experience gained to new CANDU reactor designs. Canadian CANDU owner/operators are all taking active steps to assess and understand the ageing of their plants. Prognosis for extended service operation is very positive.

In case limited resources were available, an alternative approach can be adopted after enough operation records have been acquired. The process, as described below, is based on an expert panel approach and may make use of industry knowledge base or maintenance templates (if available). Close collaboration of various departments within the power plant is essential to the success of this method. Following are the main steps:

(1) Identify key equipment and components in critical systems (contributors to systems functional failures) and create a database

(2) Group key components by type and technical characteristics and identify applicable preventive maintenance tasks, using manufacturer recommendations and current station practice.

(3) For each type of equipment, assemble a panel of engineering, maintenance and operation experts to carry out a comprehensive and systematic review of current preventive maintenance comparing it to manufacturer, industry knowledge base and determine the most appropriate maintenance surveillance and inspection tasks and their frequency.

(4) Collect and summarize the expert panel recommendations for preventive maintenance programme enhancement (new tasks, revised and/or proposed for cancellation) and issue final reports (customized preventive maintenance templates for each type of equipment and component)

(5) Review existing maintenance work orders and procedures accordingly.

CANDU reactors have experienced two types of feeder degradation:

• Pipe wall thinning due to flow accelerated corrosion (FAC) and

• Cracking.

FAC wall thinning has been seen at most stations while cracking has only been observed in a few situations. The thinning rate of the feeder pipes has been shown to be dependent on water chemistry, particularly the pH and the electrochemical potential. The mechanistic understanding of feeder cracking is still limited. To date, inter-granular cracks have been observed on both the inner and outer surface of the first and second bends on the outlet feeder and on one repaired weld. Inner surface cracks are postulated to be caused by stress corrosion cracking. Figure 10 shows the schematic diagram of feeder pipe.

Header ^

Weld

Channel

Fig. 10. Schematic diagram offeed pipe.

Outer surface cracks are currently believed to be caused by low temperature creep cracking, assisted by hydrogen ingress due to feeder thinning. The single crack on the repaired weld is currently believed to be an interrelation of both mechanisms. It is believed that all cracks are caused by unrelieved residual stresses induced during manufacturing or welding, plus other factors such as material susceptibility and chemical environment.

Due to the currently insufficient understanding of feeder cracking mechanisms, feeder pipes with known cracks cannot continue operation and must therefore be repaired or replaced. Current practice is such that, if an inspection identifies a crack, the cracked piping section is removed and replaced with new pipe. Upon crack discovery, the inspection scope is expanded to the sites deemed to have a similar risk of cracking. Substantial inspection for cracking has been performed at most plants. In some cases, 100% of bends considered susceptible were inspected. Since the discovery of the crack in a repaired weld in 2003, the risk and number of repaired welds, the safety case for continuing operation with any such potentially affected feeders, and surveillance methods and their implementation have been assessed.

Feeder Ageing Management Programmes have been developed and are updated periodically to account for inspection findings and subsequent assessments. This programme includes inspection plans at each planned outage and R&D plans. In addition, improved feeder inspection tools have been developed to inspect inaccessible sites. More reliable leak detection systems are also being considered.

The feeder pipes are of carbon steel material conforming to ASME material specification SA106 Gr B in RAPS, MAPS and NAPS units and to SA333 Gr 6 in KAPS units onwards. This material as PHT pressure boundary material has been tested extensively and is found to be a good option for LBB application. During en-masse coolant channel replacement (EMCCR) of RAPS-2 in 1998, extensive in-service inspection (ISI) of the feeders was carried out. Thinning was observed on the feeder elbows immediately after Grayloc joints particularly on outlet feeders.

Mapping of the thickness indicated that the thinning was generally local. All inlet and outlet feeder elbows were inspected. The study of reduction of wall thinning and assessment of structural integrity of these feeder elbows was carried out. The phenomenon called “flow assisted corrosion” (FAC) was understood to be the cause of local thinning of the feeder elbows. As the thinning is predominantly on extrados of the elbow, the thinning is classified as localized thinning, which needs to have different treatment compared to the normal pipe as far as structural integrity is concerned.

To assess the structural integrity of these elbows of RAPS-2, ASME Code Case N-480 was used which defines the approach. A streamlined procedure and acceptance criteria for these elbows was evolved and which is being used to assess the wall thickness reduction in feeder elbows of Indian PHWRs. Some of the elbows in RAPS-2, which were having balance life less than 10 years, were weld buildup to increase the thickness.

In MAPS-1 also the thinning of feeder elbows was observed, though the thinning was to a lesser extent than that of RAPS-2. However from the consideration of long term corrective measure and enhancement of feeder life commensurate with the life of replaced coolant channels, part length, from Grayloc end including Grayloc hub and elbow(s) of all feeders were replaced successfully during 2005 EMCCR campaign. The new elbows are of higher thickness, sch.160. The material of new elbows & pipes is alloyed with 0.2 % minimum Chromium. It is expected that the refurbished feeders now will have the life of about 25-30 years.

On the similar lines as in MAPS-1, it is planned to replace part length of all feeders in NAPS — 1 also during EMCCR campaign in the year 2006.

The ISI programme for Indian PHWRs includes the thickness measurement of feeder elbows. However, in light of RAPS-2 experience, more refined procedure and newly evolved technique have been developed for UT thickness measurement and volumetric examination of elbows on sample basis for crack detection. The criteria of selecting feeders for periodic ISI is also revised based on operational experience.

In operating stations, strict chemistry control of primary fluid is required to have reduced effect of FAC. It is also required to take wider base data for inspected feeders based on previous assessment. Repeatability of thickness measurement for the feeders with balance life less than 5 years is required for periodic assessment. Such feeders are subjected to more frequent assessment. For some feeder elbows, weld builtup can be carried out to increase the balance life as done in RAPS-2. It is required to balance between economy and man-rem budget, while deciding to discard the thinned elbow and reinstall with the new one. Replacement of critically thinned elbows can be planned in long shutdown such as EMCCR.

In future reactors (TAPP-3 & 4/ RAPP — 5& 6 /Kaiga 3 & 4), the material SA333 Gr.6 with minimum Cr of 0.2% as an alloying element has been used. This coupled with higher size elbows can reduce FAC effect to great extent and expected to give adequate design life for the feeders.

1.1.1. Introduction of approaches for PLiM

There are two conceptual approaches that can be applied concerning a utilities’ plan for continued operation. One is based on periodic safety review (PSR) [24], another on license renewal application (LRA) [25]. The USA practice and regulations follow LRA concept, while most of the European countries and Japan use PSR for obtaining permission for continued operation and to support eventual arguments for long term operation. In some Member States, these two different concepts and regulatory approaches have been adapted (e. g. Spain, Hungary, and the Republic of Korea, see table 1).

Longer life and increased output may be both attractive objectives for HWR owner/operators. Both objectives could potentially be achieved, with a variety of measures. It is important to note that these changes must not degrade existing safety margins or result in unacceptable environmental impacts. Some of these measures impact the CANDU nuclear steam plant (NSP) that others that apply more to the balance of plant (BOP). The PLiM programme has an important role, both with the LTO and to power uprate, described below.

CANDU power uprates can involve fuel improvements (more heat generation), operational changes that mainly involve equipment improvements (for greater cycle efficiency) and operational changes that mainly involve improvements to plant maintenance, surveillance and inspection (for greater durations between outages). Usually combinations of these changes are considered. The technical and business case for implementing these changes must take an integrated approach and assess the impact of component life and performance, as well as the economic return-on-investment. It also involves the regulator if the changes impact on licensed conditions.

The assessment of changed conditions on component life is typically where the PLiM programme for these components is important and useful. Systematic ageing assessment processes in combination with any previous PLiM assessments are used to assess the revised operational conditions for the component of interest. The specific components that are assessed vary and depend upon the type of power uprate being considered. Typically it could involve the fuel channels, feeders, steam generators, turbine-generators, boiler feed pumps, moisture separator reheaters and the condensers as well as a variety of electrical equipment.

Some of the proposed changes could potentially impact system operating margins. These impacts must be assessed. Any ageing issues related to these systems (or important components within these systems) must also be considered. An example would be boiler feed pump vibration. If current performance (prior to power uprating) is marginal, then the proposed power uprated conditions must be carefully assessed. For example, one might first think that this would make the current situation worse but it might actually improve performance (for instance if the changed condition moved pump operation to a better point of the pump performance curve).

Assessing the power uprate changes is not just technical but also involves a detailed economic assessment. Usually the proposed changes are grouped into several packages. Then the net present value (NPV) of each package (say Option A, Option B, etc) is evaluated and compared to a base case. The economic calculations include parameters that affect the electricity generation value. Some of these are electricity price, fuel cost, Operations, Maintenance and Administration cost (OM&A) and capital cost. The recommended option is the one that provides the greatest improvement in net present value (NPV) over the remaining plant life, compared to the base case, provided the associated technical risks are acceptable.

Effective ageing management requires a team approach in the application of a systematic ageing management process, Safety Reports Series No. 15, IAEA, Vienna (1999), which is an adaptation of Deming’s ‘PLAN-DO-CHECK-ACT’ cycle to ageing management of an SSC.

A comprehensive understanding of a component, its ageing degradation and the effects of this degradation on the component’s ability to perform its design functions are the BASIS and a prerequisite for a systematic ageing management process. This understanding is derived from: knowledge of the design basis; the design and fabrication data (including material properties and specified service conditions); the operation and maintenance history (including commissioning and surveillance); inspection results; and generic operating experience and research results.

The PLAN activity in the ageing management process is aimed at maximizing the effectiveness of ageing management through the coordination of all programmes and activities that relate to managing the ageing of a component. It includes the identification and documentation of applicable regulatory requirements, operating limits and design assumptions, relevant programmes and activities and their respective roles in the ageing management process, as well as a description of the mechanisms used for programme coordination and continuous improvement. The DO activity of the ageing management process is aimed at minimizing expected component degradation through the operation/use of the component in accordance with operating procedures and limits. The goal of the CHECK activity in the ageing management process is the timely detection and characterization of any significant degradation through component inspection and monitoring and the assessment of observed degradation to determine the type and timing of any corrective actions. The ACT activity in the process is aimed at the timely mitigation/correction of component degradation through appropriate maintenance and design modifications, including component repair and replacement.

The closed loop of the generic ageing management process indicates the need for continuous improvement of a component specific ageing management programme based on the current understanding of component ageing and on the results of self-assessment and peer reviews. Such an ageing management programme is a mixture of component specific ageing management actions designed to minimize, detect and mitigate ageing degradation before component safety margins are compromised.

This mixture reflects the level of understanding of component ageing, the available technology, the regulatory/licensing requirements, and considerations and objectives relating to plant life management. A low level of understanding of ageing of an important component requires careful monitoring because of high uncertainty in predicting the rate of degradation. The feedback of experience is essential in order to provide for ongoing improvement in the understanding of component ageing and in the effectiveness of the ageing management programme. The identification of unanticipated ageing phenomena and the development of appropriate ageing management actions for the benefit of all nuclear power plants depends, in particular, on the timely feedback of operating experience.

As of an example of ageing assessment of CANDU CSSCs, detail procedure and methodology applied to the ageing assessment of fuel channel are presented. Although ageing status depends on the design condition and O&M history of each plant, fuel channel is usually the most highlighted component in aged CANDU plants in terms of safety, integrity, performance, and maintenance [5]. To assess ageing of fuel channel they reviewed lots of plan information and evaluated its ageing in respect to the degradation mechanisms.

A. II.4.1. DESIGN REQUIREMENTS

Design requirements and performance expectations/criteria of components along with degradation allowances covering the service life of the pressure tube are reviewed first. This includes design temperature and pressure, design mechanical loads, pressure tube initial dimensions, corrosion and wear allowances, pressure tube deformation allowances, fracture toughness, pressure tube material, end fitting material, etc. Design documents such as FSAR, design manual, technical specification are reviewed to understand the design requirements. The fuel channel assemblies are designed to satisfy its intended functional requirements for 210,000 effective full power hours (EFPH) of CANDU6 reactor operation (i. e., 30 years at a capacity factor of 80%). Table A. II.2 shows the allowance of corrosion/wear and deformation of a pressure tube.