Evaluation Method for Mean Stress Effect on Fatigue Limit of Non-Combustible Mg Alloy

Abstract

Structures are commonly loaded with a mean stress. Therefore, it is necessary to investigate the influence of the mean stress on the fatigue strength of the structural materials. Fatigue strength of a material, whose fatigue crack is initiated from an inclusion, strongly depends on the size of the inclusion. The fatigue strength of the material is quite variable. Therefore, if we test the effect of the mean stress on the fatigue strength of this material, the effect would be hidden in the scatter and the effect becomes uncertain. In this study, we propose an evaluation method for the mean stress effect of the inclusion-induced scattered fatigue strength using the non-combustible Mg alloy AMX602B (X=Ca) (Sakamoto et al., 1997; Chang et al., 1998; Akiyama et al., 2000). We discuss the equivalence of an artificial defect and an actual defect (inclusion). Figure 1 shows the S-N diagram for the smooth specimens of the non-combustible Mg alloy AMX602B (X=Ca) by the authors (Kitahara et al., 2005; Kitahara et al., 2006; Fujii et al., 2008; Masaki et al., 2008). The relationship between the load stress and the fatigue life of the smooth specimens significantly varies. Moreover, no non-propagating crack was observed in the unbroken specimens. The condition of the surface of the fatigue fracture origin is shown in Fig. 2. The fracture originated at a non-metallic inclusion. There have been several studies that investigated the effect of the mean stress on the fatigue strength of the conventional Mg alloy (Forrest, 1962; Heywood, 1962; Osgood, 1970; Ogarevic & Stephens, 1990; Akiyama et al., 2000). Forrest (Forrest, 1962) reported that the effect of the mean stress on the fatigue strength of a Mg alloy can be evaluated using the modified Goodman diagram. In contrast, Heywood (Heywood, 1962) and Osgood (Osgood, 1970) reported that the fatigue strength of a Mg alloy under a high mean stress became low and that the fatigue life evaluation using the modified Goodman diagram may not be conservative prediction. However, the reason why the fatigue strength under a high mean stress decreases has not been clarified. In this study, rotating-bending fatigue tests and tension-compression fatigue tests were carried out on specimens with an artificial defect (a small hole or a small crack). Especially, we examined why the fatigue strength under a high mean stress decreases and whether the fatigue strength at N = 107 under a mean stress can be applied to an estimation using the modified Goodman diagram. The fatigue testing of the small holed specimens and the small