Adaptive real-time model on thermal error of ball screw feed drive systems of CNC machine tools

Abstract

The positioning error of ball screw feed drive systems is mostly caused by thermal deformation of the ball screw shaft in machine tool feed systems. This paper presents an adaptive real-time model (ARTM) for predicting the temperature transient distribution and thermal error distribution of the ball screw shaft. FEM integrated with the Monte Carlo method is applied to determining the heat generation rates of two bearings, moving nut and two guides in the ball screw feed system. Then based on the data of the FEM calculations, an exponential function of the feed velocity and time is used to describe variation of the temperature difference between the measured surface point and the center of the kinematical pair with time. Finally, a numerical forecasting algorithm is developed to predict the thermal characteristics of the screw shaft and its effectiveness is verified by the experiments.