Volumetric calibration in multi-space in large-volume machine based on measurement uncertainty analysis

Abstract

Machining accuracy is a significant index to evaluate characteristics of machine tools. Error compensation has been proved as an effective method for improving machining accuracy. Error measurement and identification are the prerequisite and basis for compensation. In order to detect volumetric error and identify geometric error accurately in large machines, a new method including workspace segment and multi-lateration measurement principle using laser tracker is proposed. Firstly, a synthetic test in large-volume machine containing three steps is presented. The Monte Carlo approach is implemented for evaluating the measurement uncertainty of volumetric calibration. Uncertainty sources in the calibration process, such as laser tracker sensors, thermal drift, and machine repeatability, are introduced. In order to decrease measurement uncertainty, the measurement station is optimized. Workspace is segmented, and volumetric error is calibrated in multi-station. Total least square method is implemented to transform coordinates among different measurement systems into reference frame accurately. Finally, the accuracy of algorithms for geometric error identification is verified by comparing the test results with that measured using a laser interferometer. And the volumetric calibration accuracy is also verified through comparing X-axis positioning error tested by the laser interferometer and laser track in multi-station.