Mechanical fatigue is a well-understood ageing mechanism and the main contributing factors, including load profile (frequency, stress ratio), temperature, strain rate, stress or strain amplitude, and microstructure (grain size and orientation in rolled material), have been extensively investigated for the materials used in nuclear power plants (NPPs). Mechanical fatigue is a generic engineering issue and the findings, originating either from nuclear research activities or from within non-nuclear research, are well-integrated into the different national codes and standards (e.g. U.S. Nuclear Regulatory Commission (NRC) [NUREG-CR-6909], German Nuclear Safety Standards Commission (KTA) [KTA3201.2] [KTA3211.2] and French RCC-M) including fatigue assessment methodologies and fatigue information for steels and some other metallic materials (e.g. nickel alloys and zirconium alloys) used in NPPs.
In the context of plant design and lifetime assessment of components, this ageing mechanism is relevant because of the potential for cyclic loading. Further, two types of fatigue data are of relevance, namely endurance data and crack growth data.
Endurance data in the form of fatigue design curves (magnitude of cyclic stress against logarithmic scale of cycles to failure, S-N curves) have been compiled for different materials or material classes (Carbon Steels, Austenitic Stainless Steels), e.g. separate S-N curves for ferritic steels, AISI321, AISI347 and other austenitic stainless steels in KTA [KTA3201.2] [KTA3211.2]. Fatigue design curves are usually used in combination with estimation for recurrent or cyclic loads of components during the lifetime of a NPP to consider degradation caused by mechanical fatigue.
Crack growth data and therefore crack growth rates have been investigated for the different materials used in NPPs. Typical curves present the growth rate da/dN in mm/cycle versus ΔK (stress intensity factor). Crack growth data are necessary for the assessment of components if defects have been detected.
The accumulation of fatigue damage can be expressed with the Cumulative Usage Factor (CUF) if the number of applied load cycles for a component is known or can be estimated. The assessment of the CUF for components important to safety or critical locations is usually performed periodically during the lifetime of a NPP in the framework of ageing management. Depending on national regulations, countermeasures have to be taken or additional evaluations have to be undertaken if thresholds of the CUF are exceeded.
2.1.1 Significance for NPPs
As well as differentiating between endurance fatigue and fatigue crack growth, mechanical fatigue can be divided into high-cycle fatigue and low-cycle fatigue. High-cycle fatigue corresponds to a high number of cycles (typically > 104) with relatively low stress amplitude. Typical examples of high-cycle fatigue are flow-induced vibrations or active mechanical components, like pumps or valves. Vibrations are of special importance because they can accumulate fast during operation and hence cause fatigue damage early in plant life. On the other hand, low-cycle fatigue can occur with a relatively low number of cycles (or at low frequency), but with high stress amplitude which can be larger than the yield stress of the material. Typical cases of low-cycle fatigue are transients and start-up or shutdown phases of NPPs.
Different TECDOCs have been published by the International Atomic Energy Agency (IAEA) concerning the assessment and management of ageing of major nuclear power plant components which are important to safety. Operating experience had been taken into account and the significance of this ageing mechanism can be summarized as follows:
- Concerning pressurized water reactor (PWR) and boiling water reactor (BWR) pressure vessels, mechanical fatigue is not a significant ageing mechanism except for the closure studs, which can be replaced [IAEA05] [IAEA07];
- For CANDU pressure tubes and reactor assemblies mechanical fatigue is not a significant ageing mechanism [IAEA98] [IAEA01]. In general the number of assumed cycles from the original design requirements is higher than the experienced ones;
- In steam generators mechanical fatigue can be significant only for the U-bend region of tubes in case of an improper antivibration bar support in combination with high recirculation flows [IAEA11];
- For primary piping in PWRs mechanical fatigue is only mentioned to be significant for small-diameter piping [IAEA03]. For safety relevant piping in other NPP types a similar evaluation can be taken although not explicitly mentioned.
The significance of mechanical fatigue for components of a Westinghouse four-loop PWR design and a BWR-5 design was also evaluated in a Proactive Materials Degradation Assessment (PMDA) programme by an international panel that is documented in the final report [NUREG-CR-6923].
The significance can increase if other ageing mechanisms are effective at the same time. Synergisms of the mechanical fatigue mechanism with other ageing mechanisms, i.e. thermo-mechanical fatigue (TMF) and creep-fatigue, are discussed in this synthesis report. Fatigue under environmental influence is not in the scope of this synthesis report.
2.1.2 Topics of Research Projects
Within the research of Euratom FP4 to FP7, relatively few projects have taken place that deal with mechanical fatigue issues for materials used in operating European NPPs because on one hand the mechanism and its contributing factors are well-understood and on the other hand well-established networks of excellence like the Nuclear GENeration II & III Association (NUGENIA) exist that also initiate and support corresponding R&D projects and programmes. The projects from Euratom FP4 (NET) and FP5 (GRETE, VERLIFE) are related to the fields of nuclear fusion, non-destructive examination techniques and fracture/lifetime assessment, whereas projects from Euratom FP6 (RAPHAEL) and FP7 (ARCHER, GETMAT, HELIMNET, MATTER) are directed towards the development of new generation nuclear power plants and required materials. Results from other non-nuclear projects funded under the FP4-BRITE/EURAM 3 programme (FATIGUE-DESIGN), the FP5-GROWTH programme (FITNET, TMF-STANDARD) or by the European Coal and Steel Community (now RFCS), including 7210-MC/109, 7210-MA/131, 7210-MA/823, 7210-MA/951 and 7210-PR/303, are also relevant for nuclear application and are related to the fields of lifetime assessment, standardisation, fatigue improvement/design and multi-axial load.
The results of the Next European Torus (NET) project regarding fatigue experiments under irradiation are intended for nuclear fusion. Nevertheless, the tested materials – austenitic stainless steels of the type AISI 316 – are widely used in NPPs and some parameter combinations are also of relevance and interest for nuclear fission.
The objective of the GRETE project was lifetime management together with the capabilities and reliability of associated innovative inspection techniques [EUR22282] the latter one being important for fracture assessment. Non-destructive examination techniques to monitor fatigue damage of piping have been investigated within this project. The objective of the VERLIFE project was the lifetime assessment of components and piping especially in VVER nuclear power plants [VERLIFE2003].
The RAPHAEL project from Euratom FP6 and the ARCHER project from Euratom FP7 concerned high temperature reactors (HTRs) intended for operating temperatures 900 °C – 1000 °C and 600 °C respectively, which are in the first case even far beyond operating conditions of actually running advanced gas-cooled teactors (AGRs). Materials which have been investigated in the ARCHER project include austenitic stainless steel Alloy 800, also used for steam generator tubes in German PWRs, and AISI 316 which is used for re-heater pipes in AGRs. The focus of experiments for time and cycle dependent effects on material behaviour (creep and fatigue) was towards Alloy 800 steel at elevated temperatures above 600 °C, much higher than operating temperatures in PWRs, whereas tests and experiments with the austenitic stainless steel AISI 316 concerned corrosion issues, which is not the objective of this synthesis report.
Further projects from Euratom FP7 (GETMAT, HELIMNET, MATTER) were also intended for the development of new generation reactors operating at different conditions than actually running European NPPs (liquid metal cooling, supercritical water), which requires research in the field of new materials like ferritic/martensitic steels (e.g. P91) or oxide dispersion-strengthened (ODS) steels, which are not used in operating European NPPs [GETMAT2014] [HELIMNET2013] [MATTER2016]. Nevertheless within the MATTER project generic issues of creep-fatigue were handled, e.g. interaction between creep and fatigue, different rules describing the interaction of creep and fatigue, implementation of creep-fatigue in codes and standards. A creep-fatigue interaction model was developed and also evaluated for grades of AISI 316 to accurately predict time or cycles to failure, which is relevant for operating gas-cooled reactor (GCR) plants [MATTER2016] [Utili2012].
As part of the structural integrity assessment, fracture assessment methods as well as NDE (non-destructive examination) techniques and lifetime prediction are concerned. Mechanical fatigue is one of the ageing mechanisms which has to be considered for fracture assessments – either for assessments performed during the design phase or for postulated or detected cracks in components. In the case of postulated or detected cracks, a limitation of the maximum load during the lifetime or the prediction of the residual lifetime is of interest. Fracture assessment methods which take fatigue into account have been developed and standardised in the FITNET thematic network [Hadley2008] [Janosch2005] [JRC33994] [JRC37165] [Kocak2006] [Taylor2003]. The “fitness for service” method was developed not only for nuclear application but also for the non-nuclear field.
The TMF-STANDARD non-nuclear project addressed the issue of standardisation of strain-controlled TMF tests. Within the project, equipment and protocols for testing under simultaneous mechanical and thermal cyclic load were developed. A code of practice was developed addressing the selection of materials, the definition of operation conditions, the quality control, and the design and failure analysis of components under thermo-mechanical loading. Although the majority of pre-normative R&D and validation testing have been carried out on a nickel-based alloy using two types of TMF cycles, namely out-of-phase (Re= -1, f = 180°) and in-phase (Re = -1, f = 0°) TMF cycles it is not restricted to a certain class of materials, nor to a specific type of TMF cycle [EUR22281] [JRC40879] [JRC40880] [JRC42967].
Further topics dealt with in non-nuclear research projects and funded by the RFCS include fatigue design of welded stainless steel plate or tube and the provision of design guidance in the projects 7210-MA/131, 7210-MA/823 and 7210-MA/951 [EUR19972]; problems of multiaxial load and fatigue behaviour of welded high-strength components in the RFCS project 7210-MC/109 [EUR20050]; and manufacturing techniques including welding techniques and post-weld improvement techniques in the RFCS project 7210-PR/303 [EUR22809]. Although not directly related to nuclear industry the research results can be applied to operating European NPPs as the tested materials are similar or a generic methodology is used.