Published Online: 5 June 2008
Page Count: 16
National Institute of Advanced Industrial Science and Technology, Nagoya,
(Received 2 May 2007; accepted 2 May 2008)
We have demonstrated a simple comparison method for rolling contact fatigue (RCF) performance of silicon nitrides using the balls-on-flat RCF test under stepwise loading. The RCF performance was evaluated by the mean effective mechanical input (Min)m which is the product of the mean effective load Pm, calculated from test loads and revolutions during the dwell time for each load, and the number of load cycles. The test materials were six kinds of commercially available silicon nitride, four of bearing-grade silicon nitrides (BG) and two general purpose ones (GP). The minimum test loads were 1 and 0.5 kN for BG and GP, respectively. The load was increased stepwise with each step 107 contact cycles unless failure of the specimen was detected by the acceleration pickup mounted on the loading arm of the testing machine. The balls and test specimen were immersed in mineral oil with kinematic viscosities of 8 mm2/s for BG and 68 mm2/s for GP. A clear distinction in (Min)m was made among the BG, and the GP have lower (Min)m compared to BG, despite small differences in conventional mechanical properties such as Vickers hardness, fracture toughness, and elastic modulus. The results of the balls-on-flat RCF test under constant load for one BG lend support to the hypothesis that the loading history in stepwise loading can be converted to the mean effective load Pm. It is also supported by the fact that classification of failure mechanisms in stepwise loading is similar to that for a constant load of 4 kN, which is close to the mean of Pm of the material. These results imply the viability of a standardized comparison method for RCF performance of bearing-grade silicon nitrides as well as screening for poor performance materials.
Paper ID: JAI101214