Key parameters for the reliable prediction of machine failure using wear particle analysis

Abstract

A spectrometric method was used for the determination of the chemical composition and concentration of wear particles generated in the contact surface of a worm wheel under different operating conditions. Due to different shapes and sizes, the distribution of wear particles in oil is not uniform. Because of the difficulties in obtaining a homogenous representative oil sample, the statistical factors that affect the accuracy of the results of wear metal analysis must be taken into consideration. For reliable prediction of incipient machine element failure, based on the results of concentration of wear particles in oil, analysis of variance and additional information on particle size and shape are needed.