Tuning Flank Waviness for Minimized Mesh Force Variation
摘要
End-of-line testing of high-end transmissions, such as those used in electric vehicles, identifies units where structural or airborne noise levels exceed target values. This results in high rejection costs, creating a strong initiative to reduce system excitation from gear mesh forces of non-constant nature. Traditional evaluation efforts, using manufactured gears subjected to geometrical measurement and analysis, are costly and produce findings late in the manufacturing process, causing project delays or excessive rejection rates and costs. The approach presented aims at prediction of excitation and response during the design phase by considering the designed gear macro and micro geometry. It also aims at assessment of the performance of manufactured gears by creating digital twins. Such digital twins may be subjected to geometrical analysis, non-loaded contact analysis, loaded contact analysis, and forced response analysis in a gear design and analysis software. The results may be compared to those obtained for the designed gear provided the design was undertaken with the same software. Findings will demonstrate that the control of amplitude, wavelength, direction, and phase shift of manufacturing errors influence the spectral content of the excitation and the resulting housing excitation. These findings emphasize the need to differentiate geometrical effects on single gears, measured by tactile or optical means, from pair or mesh interaction and system-level behavior. The ultimate objective is to advance gear quality control beyond purely geometrical thresholds, such as surface roughness or form deviations listed in ISO 1328, towards performance indicators like transmission error, contact stress, or micropitting risk. Engineering analysis based on measured or predicted manufacturing errors will allow for the assessment of gear performance rather than compliance with geometrical tolerances, making the decision to accept or reject manufactured gears more robust and application-oriented, ultimately reducing waste and cost.