FREQUENCY AND TIME-DOMAIN ANALYSES OF SENSOR SIGNALS IN DRILLING .1. CORRELATION WITH DRILL WEAR

Results of an investigation on a multisensor approach to drill wear monitoring are reported. Four sensors, namely, thrust, torque, and strains in two orthogonal directions of the machine table, were used in this study. A strategy involving the use of a combination of accelerated testing at high cutting speed to obtain different states of;drill wear and testing at the recommended speed to obtain data pertinent to conventional drilling conditions was adopted. The wear on the outer corner of the drill bit was used as the performance index. The sensor signals were investigated in both the time and frequency domains. In the time domain, no significant difference was observed between the signals for a sharp drill and a worn drill, and the correlation between the sensor signals and drill wear was found to be rather poor. Consequently, the power spectral densities of the sensor signals were calculated to obtain information in the frequency domain. The power spectral densities of thrust, torque, and strain in the X-direction showed good correlation with drill wear.