Process monitoring and chipping fault detection in drilling

Abstract

ABSTRACT Drilling is one of the most important and fundamental machining process and widely used in automobile, aircraft, home appliance and electronic industries. Chipping is among the important process faults occur commonly in drilling. Chipping has serious adverse effects on the surface finish and dimensional accuracy of the workpiece. Furthermore, chipping may cause tool failure which may lead to part scrappage and/or damage to CNC machines. Therefore, reliable monitoring of chipping fault would be indispensable to benefit from the machining equipment in an optimal way including automated production. In this project, chipping is induced in various amounts and types to different drills supplied by Arcelik A.S. Extensive numbers of experiments were performed by utilizing the chipped drills under different machining conditions in a CNC machine tool. Machining forces were obtained from the experiments performed in order to establish monitoring features for different chipping types / amounts in time domain, in frequency domain using Fast Fourier Transform (FFT) and in time-frequency domain using Discrete Wavelet Transform (DWT). Among the chipping types, tip chipping is easily identified in time domain whereas, shank chipping and arbitrary chipping are better monitored in time-frequency domain.