PSI - Issue 19

Masahiro Takanashi et al. / Procedia Structural Integrity 19 (2019) 275–283 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

276

2

Keywords: Fatigue crack growth; Large-scale speicmen; Carbon steel; Low-alloy steel; Fracture surface; Beach mark

1. Introduction In Japan, design fatigue curves for nuclear power plants are stipulated in the JSME S NC1 Codes for Nuclear Generation Facilities – Rules on Design and Construction for Nuclear Power Plants – (2016). The design fatigue curves originated from those of the ASME Boiler and Pressure Vessel Code, Section III, which was established in 1960’s. The original design fatigue curves of the ASME were developed in such a way that a factor of 2 on stress and a factor of 20 on life were applied to the best - fit curves obtained from fatigue tests by the small - sized specimens. These factors have been considered not as safety margins but rather as adjustment factors to apply the small - sized specimen data to actual reactor components. However, the conservatism of the factors of 2 and 20 has been discussed for a long time. In the United States, the Nuclear Regulatory Commission issued Regulatory Guide 1.207(2007), technically based on NUREG/CR - 6909(2007), and stipulated a different design curve with a factor of 2 on stress and 12 on life for austenitic stainless steel from those of the ASME. The ASME followed this trend and incorporated the design fatigue curve into Section III, 2009 addenda. There has been the same need to rationalize the design fatigue curve in Japan. To address this issue, the Design Fatigue Curve (DFC) Phase 1 and Phase 2 subcommittees, organized under the Atomic Energy Research Committee of the Japan Welding Engineering Society, have proposed new best - fit fatigue curves, design fatigue curves, and fatigue analysis methods for carbon, low - alloy, and austenitic stainless steels (Asada et al., 2018). For the verification of the proposed design fatigue curves, a Japanese utility collaborative project was launched. As a part of this project, the authors conducted reversed four - point bending fatigue tests for large - scale specimens of carbon steel and low - alloy steel plates, and drew the conclusion that approximately 3.0–4.0 - mm - deep crack initiation lives in large - scale specimens were equivalent to the fatigue lives obtained from small - sized specimens (Takanashi et al., 2018, 2019). Because the fatigue life determined by the best - fit curve corresponded approximately to only 3.0–4.0 - mm - deep crack initiation life, the crack depth was expected to be much smaller for the design fatigue life. It would be unreasonable and uneconomical to define a fatigue life of a large component in an actual plant by such a small crack initiation. A quite natural idea is to design or maintain an actual plant in view of the crack growth; such a concept has been reflected to a flaw tolerance method as described in the ASME Boiler and Pressure Vessel Code, Section XI, Appendix L. However, as reported in the previous studies (Takanashi et al., 2018, 2019), the cracks observed in the large - scale specimens grew with coalescence. Since a crack growth law is usually determined using a notched specimen, like a CT specimen, no coalescence occurs during testing. The applicability of such a crack growth law to a smooth specimen is uncertain. In this study, a crack growth analysis of a large - scale specimen was conducted, and the results were compared with the crack size (as evidenced by introducing beach marks) observed on the fracture surface.

Nomenclature a

crack depth (mm)

c half crack length (mm) d a /d N crack growth rate (m/cycle) C material constant in Paris ’ law ε a strain amplitude (%) m material constant in Paris ’ law Δ K

stress intensity factor range ( MPa m )

N bfc

fatigue life obtained from the best-fit curve proposed by the DFC subcommittee

N f

failure life