No abstract

No abstract

No abstract

A B S T R A C T Fatigue tests up to 10 8 cycles were carried out for two spring steels (Heats A and D1) and one valve spring steel (Heat F) with tensile strength, σ B , of 1720, 1725 and 1764 MPa, respectively. The size and composition of inclusions in Heats Dl and F were controlled. The surface-type fracture occurred at shorter lives below 10 6 cycles, while the fish-eyetype fracture occurred at longer lives. The fatigue limit, σ W , at 10 8 cycles was 640 MPa for Heats A and D1 and 700 MPa for Heat F. Al 2 O 3 inclusions for Heat A and both TiN inclusions and matrix cracks, i.e. internal facets, for Heat F were observed at the fish-eye-type fracture sites, while only matrix cracks were observed for Heat Dl. ODA, i.e. optically dark area, which is considered to be related to hydrogen effects, were formed around Al 2 O 3 and TiN inclusions. Fatigue tests were also conducted after specimens were heated up to 573 K in high vacuum of 2 × 10 -6 Pa. The heat treatment eliminated matrix cracks for Heat D1 and the fatigue limit at 10 8 cycles recovered to the estimated value of 920 MPa from the equation σ w = 0.53 σ B for the surface fracture. These results suggest that inclusions control and hydrogen influence the gigacycle fatigue properties for these high strength steels. In addition, it is expected that the creation of a martensite structure with a high resistance to hydrogen effects in the inclusion-controlled steel could achieve the higher fatigue limit estimated for the surface-type fracture.

Gigacycle fatigue tests were conducted for several versions of JIS-SCM440 low-alloy and JIS-SUJ2 bearing steels using 20-kHz ultrasonic fatigue testing to elucidate the relationship of the inclusion size and type to fish-eye fracture properties. The total number of tested specimens was over 200. Most of the specimens revealed Al 2 O 3 or (Cr, Fe) 3 C inclusion-originating types of fisheye fractures, while TiN inclusions and the matrix also caused fish-eye fractures in some specimens. Based on these fatigue test results, 10 9 -cycle fatigue limits were estimated according to inclusion size by resorting the obtained data points according to their inclusion sizes at the fisheye fracture origin. The estimated fatigue limits revealed saturation when the inclusion sizes were smaller than 15 lm, while those fatigue limits depended on the inclusion sizes to the -1/6th power in the case of inclusions above 15 lm in size. The saturation of the 10 9 -cycle fatigue limits was considered to be caused by the effects of the optically dark areas (ODAs). Moreover, the fatigue limits also depended on inclusion type. In comparing the Al 2 O 3 and (Cr, Fe) 3 C inclusions, the key features causing the difference in the fatigue limits were likely to be bonding between the inclusion and the matrix, i.e., the (Cr, Fe) 3 C inclusions were tightly bonded to the matrix, unlike the Al 2 O 3 inclusions, although both inclusions were of the hard type.