Crown gear coupling is a critical component of high-speed trains’ traction drive systems. Due to its structural characteristics, the coupling can operate under four alignment conditions: parallel alignment, angular misalignment, parallel offset, and compound misalignment, which lead to complex, nonlinear, and time-varying meshing behavior. Understanding how the nonlinear and time-varying characteristics of a crown gear coupling change under different operating postures is essential for developing a mathematical description of its time-varying meshing behavior. This is essential for the design and analysis of high-performance and high-reliability couplings. However, in this regard, in-depth investigations remain scarce. To address this issue, this study focuses on a crown gear coupling commonly used in high-speed trains. Based on its actual motion characteristics, a finite element modeling approach that considers multiple alignment conditions is proposed. The validity of this method is verified through bench tests. On this basis, the variation patterns of time-varying torsional stiffness and stress distribution under parallel alignment, angular misalignment, parallel misalignment, and compound misalignment are investigated. The results indicate that the variation patterns of time-varying torsional stiffness and stress distribution of the crown gear coupling differ significantly across operating conditions. Under ideal alignment, the torsional stiffness remains constant, and the maximum stress magnitude and its distribution are consistent across all teeth. Under parallel misalignment, the torsional stiffness of the crown gear coupling shows a sine- or cosine-like variation, while the maximum stress magnitude is similar for different teeth but occurs at different locations. Under angular and compound misalignment, the torsional stiffness and stress distribution show similar trends. The torsional stiffness exhibits periodic “M-” or “W-shaped” fluctuations, while the maximum stress and its distribution on the tooth surfaces show strong nonlinear differences.
Ma et al. (Tue,) studied this question.