AFCEN Fatigue Calculations Benchmark: Implementation of the RCC-M Rules in Probationary Phase for Environmentally Assisted Fatigue (EAF) Assessment on a Simple Test Case

In the framework of fatigue assessment of Nuclear Power Plant (NPP) components, Environmentally Assisted Fatigue (EAF) is nowadays receiving an increased level of attention by regulatory bodies, code committees and utilities worldwide. This concerns particularly the Long Term Operation (LTO) of NPPs, where EAF may impact significantly the stress reports that are revised for Periodic Safety Reviews (PSRs), but also nuclear new builds in some cases. In this context, several guidance documents were developed and issued, the most well-known example being the NUREG/CR-6909 report [1] prepared in the US, which prescribes inter alia an update of the Cumulative Usage Factor (CUF) through an environmental correction factor “F_en” that accounts for the detrimental effect of Light-Water Reactor (LWR) coolant environments on fatigue life and besides the water chemistry mainly depends on the temperature and the strain rate. Several nuclear codes have already incorporated this type of approach, including the AFCEN RCC-M code since its 2016 edition [2][3] through the introduction of two code cases or “Rules in Probationary Phase” (RPPs), entitled “RPP-2” and “RPP-3”. RPP-2 consists of an update of the design fatigue curve for austenitic and cast duplex stainless steels as well as Nickel based alloys, and is also a prerequisite for RPP-3 which provides guidelines for incorporating an F_en factor in fatigue usage factor calculations. RPP-3 describes a different method to consider EAF in fatigue analysis, that applies to austenitic and cast duplex stainless steel locations subjected to a Pressurized Water Reactor (PWR) primary circuit environment. It is an alternative to the straightforward application of the NUREG/CR-6909 methodology, where the F_en factor is alleviated by a factor of 3. This allowance, also known as the “F_{en-integrated}” approach, is possible because of an over-conservative quantification of the effect of surface finish under a PWR environment, which is accounted for through the design fatigue curve. This has been demonstrated on the basis of numerous fatigue tests led in a PWR primary circuit environment, on small scale fatigue specimens with a rough surface finish [4][5][6][7].

While RPPs 2 and 3 have been applied in several stress report calculations for the fourth decennial inspection of the 900 MWe French PWRs fleet, these rules are still a novelty and could be further improved in their practical implementation. From this perspective, AFCEN has then launched a benchmark exercise at the end of 2019, involving several nuclear stakeholders in Europe and China. This benchmark consisted in solving a sample problem really close to one already used in an earlier EPRI benchmark [8] which aimed at the implementation of the ASME code case N-792 [9]. The geometry studied represents a vessel nozzle with an attached piping, and the structure is exemplarily subjected to three thermal and mechanical transients. The sample problem has been solved through Finite Element Analysis (FEA) calculations with various Finite Element (FE) software packages. The benchmark was divided into two stages. The first step consisted in achieving thermal and mechanical FE calculations, which were kept as simple as possible in such a way as to avoid discrepancies between participants’ results. On the basis of the stresses obtained, the remaining part of the first phase consisted in calculating a CUF in air according to the RCC-M methodology. The second step was related to the calculation of a CUF in PWR primary circuit environment using the RPP-3, and more particularly to the calculation of F_en factors for the transient combinations identified during the previous stage. Since there is no mandatory rule (only guidance) in RPP-3 for calculating an equivalent strain rate, and since strain rate calculation constitutes a crucial step in the EAF concept, this is the area where discrepancies could be expected between participants’ results. Based on these results, improvements could be proposed to the content of RPP-3, or additional guidelines could be added.