Kinematics Error Compensation for a Surface Measurement Probe on an Ultra-Precision Turning Machine

Duo Li, Xiangqian Jiang, Zhen Tong, Liam Blunt

Research output: Contribution to journalArticlepeer-review

23 Citations (Scopus)


In order to enhance the measurement availability for manufacturing applications, on-machine surface measurement (OMSM) is integrated onto the machine tools, which avoids the errors caused by re-positioning workpieces and utilizes the machine axes to extend the measuring range as well. However, due to the fact that measurement probe actuation is performed using the machine tool axes, the inherent kinematics error will inevitably induce additional deviations onto the OMSM results. This paper presents a systematic methodology of kinematics error modelling, measurement, and compensation for OMSM on an ultra-precision turning lathe. According to the measurement task, a selective kinematics error model is established with four primary error components in the sensitive measurement direction, based on multi-body theory and a homogeneous transformation matrix (HTM). In order to separate the artefact error from the measurement results, the selected error components are measured using the reversal method. The measured error value agrees well with the machine tool’s specification and a kinematics error map is generated for further compensation. To verify the effectiveness of the proposed kinematics error modelling, measurement, and compensation, an OMSM experiment of an optically flat mirror is carried out. The result indicates the OMSM is the superposition of the sample surface form error and the machine tool kinematics error. With the implementation of compensation, the accuracy of the characterized flatness error from the OMSM improves by 67%.
Original languageEnglish
Article number334
Number of pages15
Issue number7
Publication statusPublished - 2 Jul 2018


Dive into the research topics of 'Kinematics Error Compensation for a Surface Measurement Probe on an Ultra-Precision Turning Machine'. Together they form a unique fingerprint.

Cite this