Rolling Contact Fatigue Tests of Ceramics by Various Methods: Comparison of Suitability to the Evaluation of Silicon Nitrides

Abstract Results of rolling contact fatigue (RCF) tests by balls-on-rod (BOR) and two-roller methods have been compared to those by the balls-on-flat (BOF) method by using three different silicon nitride grades. Two were bearing grade (BG) and one was a general purpose (GP) grade. For a GP material, there was not much difference between the results of BOR and those of BOF, and the observations of a spalled region by BOR suggested similar spalling behavior to BOF. Meanwhile, for BG materials, most tests were terminated without spalling, but grooves, whose likely cause is inferred to be insufficient lubrication, were formed. As for the lifetime in BOR, however, the results for BG materials were much longer than those of the GP, demonstrating that the BOR result is qualitatively consistent with that of BOF, where instead most specimens failed in RCF by spalling or peeling. In the two-roller method, tests of a limited number of specimens were operated at high slip ratios of 10 % during longer cycles under low contact stress, and wear was predominant, suggesting the potential of providing the information on wear behavior rather than RCF. The results suggest that the two-roller method could differentiate the BG materials in terms of wear behavior. Considering these results, both the BOF and the BOR are applicable to distinguish BG and GP materials in terms of fatigue resistance. In particular, the BOF is judged to be more suitable for comparison of the RCF behavior of BG materials, which are commonly tested under higher contact stresses.