Surface profile and acoustic emission as diagnostics of tool wear in face milling

Abstract

This thesis examines the relationship between progressive wear of cutting inserts during a face milling operation and the acoustic emission and surface profile generated by that process. Milling experiments were performed on a range of workpiece materials using both eight point and single point inseý arrangements contained in two cutters of different geometries. Surface profile measurements were made using a stylus profilometer at intervals during the experiments. Correlations between the wear state as measured by the length of the flank wear land (Vb) and the spatial frequency content of the surface profiles were established. Investigations into the variation of fractal dimension of a milled surface with Vb demonstrated that no correlation was observable between these quantities. Acoustic emission (AE) measurements were made using a non-contacting fibre-optic interferometer which allowed the rms of the AE signal and its mean frequency to be determined. Correlations between these parameters and Vb were established for a range of workpiece materials and cutter geometries. It was shown that neither AE measurements nor surface profile measurements in isolation could predict tool wear state in all situations. The advantages of fusing data from surface profile and AE sources via an artificial neural network in tool wear monitoring were demonstrated