PSI - Issue 7

Taizo Makino et al. / Procedia Structural Integrity 7 (2017) 468–475 Taizo Makino et Al./ Structural Integrity Procedia 00 (2017) 000–000

470

3

rotating an inclined axis from the X-ray beam direction. A 37 keV monochromatic X-ray beam was employed. The experimental setup and the 3D reconstruction procedures are given by Nakai et al. (2016). 2.3. Successive observation of damage process in one specimen by SRCL Fig.1 presents the change of 3D-image of cracks and inclusions in one specimen during the RCF test (Nakai et al., 2016 and 2017). We can see from (a) that many stringer inclusions are randomly distributed in the internal region of the specimen. A crack, referred to as a vertical crack in this paper, was initiated around a relatively long inclusion, in which the upper end was close to the rolling contact surface. The length in the depth direction of the inclusion was 55 μm. After that, the vertical crack was propagated in the depth direction with an increasing number of cycles as shown in (b). Next, another crack, referred to as a horizontal crack in this paper, was initiated at the face of the vertical crack as shown in (c), and propagated in a parallel direction to the surface as shown in (d). Finally, flaking failure occurred. Fig.2 (a) is the damage process of the case in which the artificial defect originates RCF damage investigated by the authors (Makino et al., 2014). Fig.2 (b) is 3D image obtained by rotating Fig.1. (a) is similar to (b), that is to say, both cases, in which the artificial defect and inclusion originate RCF damage, follow almost the same damage process. It is concluded that the RCF damage process emanating from inclusions leads to vertical crack initiation, its propagation and horizontal crack initiation, its propagation, and flaking failure as in the case from artificial defects.

Fig.1. Change of 3D-image of cracks and inclusions during rolling contact fatigue test (Nakai et al., 2016).