Methods to reduce variation in cutting tool life tests

Abstract

Cutting tool life tests are expensive, due to long test duration, high machine hour rates and the number of repetitions required to achieve the desired significance of the mean value. The number of tests can be reduced by identifying the causes for the variation and eliminating them. The biggest source of variation in rough turning tests of titanium with commercially available cemented carbide inserts proves to be the cutting edge micro-geometry. Two different approaches for assessing the relevant geometry features are compared. The first method uses optical 3D microscopy. The second method involves force measurement in a very short preliminary cutting test and a force model for oblique cutting to extract data about the cutting edge. It is less time consuming and requires less investments than the first method, but delivers slightly worse results. Using the values from one of the preliminary tests, one can discard expected outlier inserts before beginning with the tool life test or correct tool life results afterward. The tool life of cutting edges with worse-than-average micro-geometry is corrected upwards based on the data of the preliminary test, whereas better-than-average cutting edges are corrected downwards. In the given case, this leads to a reduction of the standard deviation of the tool life by the factor of three.