In recent years, the widespread adoption of CNC lathes has facilitated automation and unmanned operation in machining processes. However, tool wear during prolonged cutting operations remains a critical issue, leading to decreased machining accuracy and productivity. Conventional wear detection methods—such as visual inspection or image processing—require machine downtime, limiting the benefits of full automation. This study proposes a method for estimating cutting resistance using motor torque data obtained from the CNC servo system, eliminating the need for external dynamometers. The estimated cutting resistance was compared with actual measured values, showing high accuracy in the main and feed force directions. Furthermore, a camera unit installed above the machining mechanism was lowered during measurement to capture images of the tool, and image processing was used to measure flank wear width, validating the effectiveness of automatic wear measurement. Results also indicated that wear progression velocity increased with cutting length and responded more sensitively than wear width or cutting resistance at early stages. These findings suggest that combining torque-based resistance estimation with wear velocity monitoring can improve tool condition monitoring and enable timely tool replacement without interrupting machining operations.
HOMMA et al. (Wed,) studied this question.