Characterisation of tool wear and failure

Abstract With machine tool becoming progressively more automated and integrated it is vital that stoppages should be minimised. One critical area that remains a problem is the accurate prediction of tool wear and tool failure. Due to the inconsistency of tool deterioration even under controlled conditions it is extremely uneconomical to use a statistically determined “safe” tool life. A more viable alternative is to monitor the tool during cutting and give a pre-warning just prior to tool failure. Numerous attempts have been made to design a suitable tool-wear sensor using parameters such as cutting force, temperature, acoustic emission, and so on. These approaches tend to treat the symptom. Another approach is to study the cause and the mechanisms leading to tool wear. Until now, most studies have assumed a two-dimensional loading. In the present paper a three-dimensional loading is attempted. Here, an experimental set-up using a split-tool is presented to measure the loading on the rake face during cutting. Using a finite-element analysis, the principal and von Mises stresses in the tool are examined to ascertain the points of highest stress. From this, it is possible to predict the mode and location of tool failure. The main conclusions that can be drawn are firstly, the number of areas subjected to tensile stress and the magnitude of the stress is smaller with a rounded nose tool than a sharp tool. Secondly, three regions along the cutting edge experience exceptional high stresses and are most susceptible to failure. Thirdly, tensile stresses are found only near the cutting edge. This refutes findings using two-dimensional loadings.