Real-time Tool Prefailure Detection in Conventional and High-Speed Milling Applications

Tool chipping and breakage are sudden tool failure phenomena that are stochastic and occur in milliseconds. They can damage the machined part surface by inducing residual tensile stresses and altering the surface finish. Therefore, reliable manufacturing requires accurate prediction of cutting tool conditions and control of the cutting process. The objective of this paper is to describe a real-time tool condition monitoring system that predicts and prevents tool chipping or breakage in milling operations. A novel, multi-sensor, signal processing approach for online, prefailure detection is adopted. It identifies the unstable crack propagation during the prefailure phase using feedback signals, independent of the cutting parameters and workpiece material. Acoustic emission signals were processed in real-time to detect tool prefailure. Conventional and high-speed milling, experimental testing was carried out using different cutting parameters, tools and workpiece materials to validate the generalization capability of the proposed approach. Correlation between chipping size and prefailure phase features was developed for decision-making where a threshold function was introduced to avoid any tool chipping above the accepted levels. The developed system features a 2-way communication controller that can overwrite cutting parameters to safeguard the machined part. Results confirmed prediction accuracy. The time required for signal processing, decision making and communication with the machine controller allowed the stopping of machining before part damage.