A Comparative Study of Wear Debris and Vibration-Based Gear Damage Detection Methods Applied to Mild Wear in a Spur Gear System

Abstract

Mild wear in gear transmission systems accounts for 41% of the gear flank surface failure mechanisms. Vibration detection parameters and wear debris methods are widely used for the condition monitoring of gear transmission systems experiencing mild wear. In this study an attempt has been made to predict the effectiveness of wear debris damage detection and vibration parameters in estimating the mild wear progressive failure which is a precursor of breakdown. The vibration parameters root mean square (RMS), kurtosis, crest factor, energy ratio (ER), FM0, and correlation coefficient of residual signal (CCR) are obtained from the time-averaged vibration data for two sets of run-to-failure experiments for the EN24 steel spur gear system. Each of these experiments are performed at a constant speed (1200 RPM) and torque (40 Nm). Wear debris monitoring is performed online. The ER and FM0 have been found to consistently increase after 140 h, indicating initiation of progressive failure. The RMS and CCR values are found to be inconsistent in indicating initiation of failure. The kurtosis and crest factor parameters possessed constant behavior with respect to time. The number of wear debris particles changes at 110 h and 160 h respectively, showing the initiating and limiting values before the occurrence of severe wear for the first set of the experiment; similarly, 110 h and 140 h for the second set of experiment. Overall, it is found that the wear debris method could effectively predict the failure mechanism much earlier in comparison to parallelly obtained vibration indicators ER and FM0.