Process monitoring to assist the workpiece surface quality in machining

Abstract

The paper reports on research which attempts to correlate the quality of the machined surface after broaching and the output signals obtained from multiple sensors, namely acoustic emission, vibration, and cutting forces. The quality of the machined surface was estimated in terms of geometrical accuracy, burr formation, chatter marks and surface anomalies. Cutting conditions were varied based on an orthogonal array with cutting speed, coolant conditions, and tool settings as factors. Each orthogonal array was repeated at three levels of the tool wear. The results show that the cutting force signals are sensitive enough to detect the geometrical deviation of the machined profile, burr formation and to a lesser extent the chatter marks. The vibration signals were found sensitive to detect the chatter marks while the acoustic emission signal proved to be efficient for the detection of small surface anomalies such as pluckings, laps, and smeared material. However, up to now, no clear distinction between the different types of the surface anomalies could be made using the analysis of the acoustic emission signal. Time and frequency domain analysis of the output signals were carried out in order to develop appropriate techniques for qualitative/quantitative evaluation of the machined surface quality. It was found that each sensory signal is rather limited to a narrow field of application where certain surface features are detectable. The limitations of the employed sensory signals/analysis methodologies used to assess the workpiece surface quality, and their applicability in the industrial machining conditions are also discussed.