Issue34

Y. Nakai et alii, Frattura ed Integrità Strutturale, 34 (2014) 246-254; DOI: 10.3221/IGF-ESIS.34.26 251 Rolling direction Rolling direction Crack 200 m  50 m  (a) Flaking (b) Enlarged view of flaking Figure 8 : SEM micrograph of surface at flaking ( p max = 5.39 GPa, N f = 7.67×10 6 cycles). Another horizontal shear-type crack was also formed from the same inclusion. Its formation site was shallower than that of the displayed vertical crack, but unfortunately it could not be shown in the same figure. Such parallel shear-type cracks were also previously observed in a specimen with an artificial hole, which simulated an inclusion [8]. At N = 7.67×10 6 cycles, flaking occurred as a result of the propagation of the shear-type crack. The shape of the flake was similar to that observed on the surface by SEM as shown in Fig. 8. D ISCUSSION n the specimen with a sulfur concentration of 0.020 mass%, a vertical crack formed from an inclusion whose length in the thickness direction was 30 μm, and a shear-type crack was formed after the vertical crack reached a length of 50 to 60 μm, which was similar to the initiation condition of the shear-type crack in the specimen with a sulfur concentration of 0.049 mass%. Since the number of cycles to crack initiation at the surface and flaking are higher for the former specimen, and the propagation of the vertical crack was followed by shear-type crack formation, the length of the inclusion must affect the formation and propagation of the vertical crack and subsequent formation of the shear-type crack. Therefore, the flaking life must be affected by the length of the inclusion. SEM observation of the surface showed that the depth of flaking was from 20 to 40 μm in both specimens, which is similar to the depth of the shear-type crack. The depth where the shear stress had the maximum value was 67 μm when p max was 5.39 GPa, meaning that the site of shear-type crack formation was not coincident with the position of maximum shear stress. The existence of a vertical crack may thus change the shear stress distribution. The flaking process in RCF observed in the present study is summarized in Fig. 10. (1) A crack that is perpendicular to the rolling surface and rolling direction forms from an inclusion that is close or adjacent to the rolling surface. (2) The crack propagates vertically in the depth (thickness) direction. (3) After the vertical crack propagates to a critical depth, a shear-type crack forms, which is parallel to the rolling surface. (4) The shear-type crack propagates to induce flaking. In the previously proposed mechanism of RCF, only a shear-type crack was considered in the discussion of the RCF process. In the present study, however, it was shown that the formation of the shear-type crack is induced by a vertical crack that forms before the shear-type crack. The formation of the vertical crack is affected by the shape and size of inclusions, which are determined by the sulfur concentration and heat treatment. C ONCLUSIONS n the present study, 4D observations of the formation and propagation of the rolling contact fatigue (RCF) tests were performed on a high-strength steel by combining a newly developed compact rolling contact fatigue test machine with synchrotron radiation computed laminography (SRCL). I I