Reliability prediction and demonstration for wear failure mechanism

This paper presents a methodology of integrating wear models in a system level model to assess the risk of failure in a field application, through endurance testing at the system level. Correlation of the wear prediction by the system-integrated wear model with actual measurement was performed and used to estimate the field operation reliability. Results are shown for sliding wear as well as impact wear phenomenon in this paper. The methodology adopted is to develop a system level dynamic model of its operation to capture the forces at the wear interfaces which is then integrated with Archard’s wear model. This model is then run using the system level sensor data captured during endurance testing to replicate the wear interface dynamics and predict the wear phenomenon. Since this wear cannot be measured during system operation, the effect of wear which results in a system level performance change is also modelled. This enables to track the actual system level performance change during endurance and correlate with the model estimates. In the case of sliding wear, wear modeling and prediction was done for a friction material, and the mean wear predicted was not different from the model predicted values at 95% confidence under a field application duty cycle. For impact wear failure mechanism, the wear volume was calculated using CMM measurement of the wear surface and the model predicted wear volume is within 76% accuracy of actual wear volume. Areas in the model which can modified to improve the accuracy of the model are mentioned in this paper.