PSI - Issue 19

P. Cussac et al. / Procedia Structural Integrity 19 (2019) 463–471 P. Cussac / Structural Integrity Procedia 00 (2019) 000 – 000

464

2

is one of the issues raised by these new requirements. Surface anomalies considered in this context may result from manufacturing or maintenance operations. Indeed, surface defects, a few tenths of a millimeter deep, can be generated when unexpected events such as fall or friction of tools happen, or during operations required and controlled, such as handling of large components. Despite a positive field experience, the issue of nuclear components fatigue resistance in the presence of surface imperfections arises today. Conventional approaches are based on representative surface conditions of the used parts. However, some imperfections may be more severe than the surface conditions considered. The question of a potential reduction in fatigue life in the presence of these surface imperfections can therefore legitimately arise. The fatigue resistance of metal components in the presence of surface anomalies has been already studied but mainly in the context of fatigue with stresses under the yield strength. Generally, surface imperfections are prime sites of initiation and can affect fatigue behavior [Doremus (2015)], [Gourdin (2017)]. However, the damage that may occur on the components considered in this study is more a result of the low-cycle fatigue that is associated with generalized plasticity. The work presented here aims at studying the impact of surface imperfections on the fatigue behavior of nuclear components and at evaluating the influence of characteristic parameters under such conditions of low-cycle fatigue. A 304L type austenitic stainless steel used for manufacturing the primary circuit piping of French nuclear power plants is studied here on the basis of a uniaxial fatigue test campaign conducted in total strain. In addition, during the fatigue tests, the initiation and propagation of cracks have been followed by the direct current potential drop method (DCPD) as well as ink markings.

Nomenclature a 0

crack depth and semi minor axis of the ellipse

b 0 D 0

semi major axis of the ellipse

sample diameter da/dN 0 fatigue crack propagation rate DCPD 0 Direct Current Potential Drop N 5

lifetime associated to a 5% drop of the maximum stabilized stress

V 0 V 0

difference of electrical potential

difference of electrical potential at the first cycle

2. Material and experimental methods

The 304L austenitic stainless steel specimens, with a cylindrical shape, were taken from a rolled sheet produced by Creusot Loire Industrie [de Baglion (2011)], [Poulain (2015)]. Their diameter is 9 mm, with a 13.5 mm useful gauge length (Fig. 1). The longitudinal axis of the specimens is parallel to the rolling direction. In order to overcome the roughness due to the machining process and to eliminate any scratches, each specimen systematically undergoes several polishing steps until a so-called "mirror-polished" state is obtained.

Fig. 1. Specimen Geometry