Experimental Validation of Wear-Noise Correlation In Different Temperature Conditions

Abstract

Friction and acoustic emission (AE) techniques have been used widely for machine condition monitoring, especially for fault detection and diagnosis, yet little work has been done on the development of wear and AE methods, even though wear is one of the main causes of faults. As widely reported, surface roughness and its changes have a close relationship with the wear status of bodies in sliding contact. In a wear process, the surface of a moving component evolves continuously, with the change starting at a micro-level and gradually progressing to a macro-level. Therefore, it is important to monitor and quantify the change in the surface roughness, which often has to be carried out after stopping the machine and examining the surface using offline analysis techniques. This thesis aims to study the correlation between wear and noise signal for surfaces in sliding contact under different temperature conditions and validate the results with an analytical model which has been done by another PhD student at Cranfield University. Tests were conducted on a tribometer using pin on disc technique with two different loads (10N and 20N) and three different temperatures ( ambient, 40°C and 60°C) . The discs used are made of mild steel, iron and aluminium where the surface roughness of tested discs were measured before and after the experiment. The noise signal was analysed using root mean square processing technique for each cycle to establish a relationship between the signals and wear volume. Later this relationship was compared with analytical model to check the validity of the model workability under different temperature conditions.