RELIABILITY ANALYSIS OF SLIDING COMPONENTS

Abstract

This report presents a reliability model for sliding components based upon a physical law governing the plasticity­dominated wear and the mild-oxidational wear processes : the Archard's wear theory (1953). Wear tests using a pin-on-disk configuration are conducted for 3 different pin materials to obtain the required data: the Archard wear factor. The pin materials used are pure iron, low alloy steel (En24) and sintered low alloy steel (GK6C) respectively. This wear factor is perceived as a stochastic variable to reflect the intrinsic variability of the wear processes. By assuming that failure is defined by the removal of a critical wear volume, the time to failure is then formulated as the first passage time of the underlying stochastic process. The corresponding reliability function can then be expressed as a function of the statistics of the Archard wear factor, the material properties and the sliding conditions. The lower confidence bound of the reliability function is also given by minimizing the reliability function where its parameters are allowed to vary within their respective confidence intervals. Finally, an alternative formulation of the reliability function is presented. This is evolved from the misconception that the Archard wear factor has been treated as a random variable. The resulting error, relative to the proposed model, is also computed for comparison purpose.