On-machine measurement cycle for the adaptive thermal error compensation of linear axes

Abstract

Thermal compensation strategies are relevant for precision 5-axis machine tools, because thermal errors are among the major causes for geometric inaccuracies of machined workpieces. Therefore, many different compensation strategies have been developed to consider thermal errors in the numerical control of machine tools. However, automated and efficient thermal compensation strategies for 5-axis machine tools are still an outstanding challenge in industry. The thermal adaptive learning control (TALC) overcomes this issue by combining on-machine measurements and Auto Regressive models with eXogenous inputs (ARX). This paper presents an automated on-machine measurement cycle based on a touch trigger probe and an artefact. This on-machine measurement cycle enables the application of the TALC to the thermal errors of linear axes. The measurement cycle identifies the thermal error of a linear and a rotary axis and allows the separation of the artefact-related errors. The developed artefact is integrated into the machine tool table, to realize an efficient on-machine measurement without restricting the working space. The results of the developed on-machine measurement cycle are validated using grid encoder measurements of the same thermal load cases. This comparison verifies a good consistency of both measurement methods. Furthermore, the TALC is applied to a linear axis of a 5-axis machine tool. The results show that the TALC method significantly reduces the positioning error of the considered linear axes.