This paper was produced for the 2019 NAFEMS World Congress in Quebec Canada

Resource Abstract

The gearbox is one of the most critical parts in modern wind turbine system. With the developments of the wind industry, the power of wind turbines has grown rapidly in the last ten years. The gears in the driveline have large modules and are subject to heavy-duty cycles of the various wind situations. The wind turbine is designed for 20 years of life onshore and 25 years offshore. Many failures have occurred in wind farms due to premature mechanical and electrical faults. Among them, the large, heavy-duty gear damage is the costliest failure due to the replacement time and cost. There is an industry demand to find out the root causes of failure and to improve their life in the field.



The typical failures of large wind gears are found to be contact fatigue damages. It is different from gears with small modules such as automotive gears, where the contact failure is normally a result of pitting fatigue. The large wind turbine gears break mostly on the subcase fatigue.



In this paper, two typical contact failure phenomena associated with wind turbine gears have been investigated. One is the contact fatigue failure initiated from the gear tip, the other is the contact fatigue failure initiated from the sub-case cracks.



A 3D dynamic gear contact model using a typical 2MW wind turbine helical gear parameter has been established and analysed. The model considers the multi-tooth contact; load sharing under the elastic deformation; the impacting of the tooth meshing entry. The contact stress and interference for different tip reliefs have been studied. The results have been tallying with the observation from those failed from the field.



Many parameters could contribute to the subcase fatigue failure, e.g. the overload, the material defects, the heat treatment hardness profile from the tooth surface to the core, and ultimately, the maximum contact sheer stress below the contact surface. Small and large gears are all designed with ISO or AGMA standard, but have different failure modes. The contact stress using the theoretical Hertzian contact method, therefore, has been calculated and compared with that from the finite element contact model considering more accurate stiffness and load sharing. In order to get the stress as close as possible to the actually value, a refined multi-tooth contact 2D model has been meshed and analysed, the results have been presented and compared with the classic Hertzian contact stress. It has been found that the location of the maximum sheer stress beneath the surface are identical, but the maximum sheer stress has 8.4% difference. This will have a 1.175 factor difference in terms of the gear contact safety factor. This prompts further extensive contact stress study on large heavy-duty gears for improving gear design.