PSI - Issue 7

Kentaro Wada et al. / Procedia Structural Integrity 7 (2017) 391–398 K. Wada et al./ Structural Integrity Procedia 00 (2017) 000–000

392

2

1. Introduction It is well known that, during the rolling contact fatigue (RCF) process of bearing, a fatigue crack can grow via various modes, viz. , the opening-mode (mode I), the shear-mode (modes II and III), or any combination of modes I − III ( i.e. , mixed mode), under a complicated, tri-axial stress state (Hashimoto et al. , 2017; Komata et al. , 2012; Olver, 2005). With the aim of achieving the quantitative evaluation of the RCF strength as a crack problem, a number of researchers has been studying the behavior of shear-mode fatigue crack growth (FCG) in high strength steels. For example, Murakami et al. (2002) obtained the mode II FCG threshold of a bearing steel by using double-cantilever type specimens. Other researchers also developed original testing methods for measuring the shear-mode threshold of high strength steels (Fujii et al. , 2001; Nishizawa et al. , 2005; Otsuka et al. , 2004; Matsunaga et al. , 2009, 2011; Okazaki et al. , 2014, 2017). However, in order to describe the RCF failure comprehensively based on fracture mechanics, it is also important to elucidate the behavior of mode I cracks under compressive mean stresses - the main focus of this research. Although numerous studies have been conducted about the effects of the stress ratio, R , on FCG and fatigue limit, only very few have targeted the low stress ratio regime ( i.e. , R ≤ − 1). For example, Yamabe et al. (2007) reported that the fatigue limit of notched cast iron was in good agreement with a modified Goodman diagram in the stress ratio range of − 4 ≤ R ≤ 0. Kondo et al. (2003) also documented that the modified Goodman prediction worked well for the pre-cracked carbon steel, JIS-S25C ( HV = 141), within the range of − 6 ≤ R ≤ 0.6. However, to the authors’ knowledge, hardly any studies have examined the low- R effects on high-strength steels ( e.g. bearing steel) commonly used for the fabrication of rolling contact components. To describe the RCF process quantitatively, fatigue thresholds should be clarified, such as the threshold stress for crack initiation and the threshold stress intensity factor range, Δ K th . Therefore, in this study, tension-compression fatigue tests were carried out at various stress ratios on the semi-circularly-notched bearing steel, SAE52100, in order to elucidate the effect of compressive mean stress on fatigue thresholds.

Nomenclature a

Half-length of crack at surface Half-length of notch at surface

a init

Δ K eff Effective stress intensity factor range Δ K eff th Threshold effective stress intensity factor range Δ K th Threshold stress intensity factor range HV Vickers hardness K max th Threshold maximum stress intensity factor ω Compressive yielding zone size R Stress ratio σ 0.2 0.2% proof stress σ a Stress amplitude σ B Tensile strength σ cl Crack closing stress σ m Mean stress σ max Maximum stress σ max, th Threshold maximum stress σ mim Minimum stress