A Ball Bar Measurement Scheme with Single-Axis Rotation Modes for Position-Independent Geometric Error Identification in Dual Rotary Table Five-Axis Machine Tools

Position-independent geometric errors (PIGEs) of rotary axes are critical factors limiting the machining accuracy of five-axis machine tools. Aiming at the accurate identification of rotary-axis PIGEs for dual rotary table five-axis machine tools, this study proposes an improved double ball bar (BB) measurement scheme. This scheme features excellent decoupling performance, convenient operation and high efficiency: complete error decoupling is realized via independent single-axis rotary measurement; a standard fixed-length ball bar is adopted without any auxiliary fixtures to effectively reduce setup-induced errors; a non-iterative analytical algorithm based on circular eccentricity fitting is developed to accurately identify all geometric errors of rotary axes. Based on homogeneous coordinate transformation theory and the fundamental BB measurement principle, mathematical models that correlate rotary-axis PIGEs with BB length deviations are established for four designed measurement modes. By constraining only one single rotary axis to move while fully locking all other axes during each test, the proposed method enables BB measurement data to exclusively reflect the geometric errors of the tested rotary axis, thereby fundamentally eliminating geometric error coupling induced by multiple axes. Subsequently, the correlation between PIGEs and BB length variations is quantitatively analyzed via numerical simulation. On this basis, an analytical PIGE identification strategy is developed using circular eccentricity fitting of measured BB trajectory data. Taking a typical BC-type dual rotary table five-axis machine tool as the experimental platform, all eight rotary-axis PIGEs are successfully identified and compensated. Experimental results demonstrate that the maximum positional error is reduced from 144.53 μm to 7.72 μm, achieving an overall accuracy improvement rate of 79.48%. The proposed method enables high-precision PIGE decoupling and identification, effectively improves the machining precision of five-axis machine tools, and exhibits good applicability for dual rotary table machine tools, providing a reliable alternative for geometric error identification in five-axis machining systems.