Indirect tool-wear maps for tool condition monitoring in dry metal drilling operations

Abstract

To avoid damage to work and/or machine, real-time tool condition monitoring is necessary in automatic and sustainable manufacturing operations. In particular, metal machining with NC machine tools can benefit handsomely from the identification of dull tools in real-time so that they can be replaced. This requirement is especially true in dry (sustainable) drilling operations where heat buildup represents a major challenge.

In this work, quantitative maps of indirect tool-wear of chisel drills undergoing dry machining are charted based only on transducers reporting electrical current measurements from machine (spindle and feed drives) motors. Associated with the maps are qualitative descriptions of the various modes of tool-wear afflicting the drill tools. Based on these tool-wear maps, a novel wear criterion is developed that rely on the % increase in motor (spindle and feed drive motors) RMS current values and is dubbed the Current Rise Index (CRI). For verification, this index is found to positively track the corresponding increase in cutting forces. Utilizing this index, an implementation example is presented in this paper by which a real-time tool monitoring of chisel drills is achieved by inputting the CRI along with the cutting parameters to an Artificial Neural Network (ANN) which yielded good tool condition predictions.