Sideband analysis of cracked planetary gear train considering output shaft radial assembly error

In a planetary gear train (PGT), the accuracy of gear tooth crack fault monitoring is affected by many external factors, such as the radial assembly error of the output shaft. In this paper, a threedimensional loaded tooth contact analysis (3DLTCA) model is proposed considering the variation of the center distances and tooth root crack fault of the planet gear. The time-varying meshing stiffness (TVMS) caused by the crack fault of the planet gear and the radial assembly error of the output shaft has been revealed. By introducing the TVMS as the vibration excitation source into a coupling PGT-shaft dynamic model, the dynamic response under the joint influence of assembly error and the crack fault is calculated. By comparing with the experimental results, the distribution law of sidebands caused by the radial assembly error of the output shaft and the crack fault of the planet gear is obtained. Based on the distribution law of sidebands, a signal decomposition method is proposed. Sidebands caused by the assembly error and the crack fault are separated. Through the validation of the condition indicators (CIs), it is found that the proposed signal decomposition method based on the distribution law of sidebands can improve the accuracy of diagnosing the crack fault of the planet gear under the influence of the radial assembly error of the output shaft.

Automated summary

This paper proposes a three-dimensional loaded tooth contact analysis model considering the variation of the center distances and tooth root crack fault in a planetary gear train (PGT). The time-varying meshing stiffness caused by the crack fault of the planet gear and the radial assembly error of the output shaft is revealed. By introducing the time-varying meshing stiffness as the vibration excitation source into a coupling PGT-shaft dynamic model, the dynamic response under the joint influence of assembly error and crack fault is calculated. A signal decomposition method based on the distribution law of sidebands is proposed, which separates the sidebands caused by the assembly error and the crack fault. The proposed signal decomposition method improves the accuracy of diagnosing the crack fault of the planet gear under the influence of the radial assembly error of the output shaft, as validated by condition indicators (CIs).