Real-time cutting tool condition monitoring in milling by means of torsional vibration measurement of the spindle shaft

On-line cutting tool condition monitoring in various machining processes is an important aspect of adaptive and flexible manufacturing systems research. This research developed a new method for tool condition monitoring in milling via measurement of the torsional vibration of the spindle shaft which overcame many problems of the existing techniques. To measure the torsional vibration of the spindle shaft in milling, a laser torsional vibration measurement system was built, which has the advantages of being noncontacting, not limited by mechanical resonance, does not interfere with machining process, and is suitable for non-laboratory environments.;Details of the interaction between the tool wear and breakage and the torsional vibration of the spindle shaft were analyzed experimentally and mathematically. Results showed that the torsional vibration of the spindle shaft was sensitive to tool wear and breakage. It is shown that cutting force, torsional vibration of the spindle shaft, and damping increase with tool wear. Since frequency analysis of a signal can show the amplitude of each frequency component in the signal in terms of frequency power, the measured torsional vibration signals were frequency analyzed and the effects of tool wear on the frequency powers of forced and the natural vibration were examined. Increase in damping was due to an increase in the frictional damping as a result of an increase in the tool-workpiece contact area. This increase in frictional damping caused a reduction in the increase rate of the natural frequency power and hence the frequency power ratio of the forced and the natural vibration increased with tool wear. The estimation of tool wear was made possible through estimating the increase in the frictional damping using the frequency power ratio. The frequency power ratios obtained in various tests showed that frequency power ratio was a reliable factor for tool wear detection. In the tests for tool breakage, the major changes were observed in the forced vibration components. When broken tools were used, cutting forces increased and consequently amplitudes of forced vibration increased.;Based on the results a simple, fast, and accurate signal processing algorithm was developed, which enabled real-time identification of the cutting tool condition in milling. This monitoring system, therefore, can be a part of an adaptive milling system which makes untended operation of the system possible.