AGMA 16FTM20 PDF

In order to increase the power density of gears, a high level of information concerning the load carrying capacity is necessary. Calculation methods for the flank and tooth root load carrying capacity are well-established in the industry and are an important tool for the gear design engineer. The existing methods cover various types of gears, such as cylindrical, bevel, or beveloid gears. Calculating the flank and tooth root load carrying capacity for various gear types relies either on analytical formulae (Hertzian theory, fixed beam) or on results of FE-based tooth contact analysis software.

The existing calculation methods for the tooth root load carrying capacity derive the material strength either from fatigue limit tables, that are based on test rig results, or from the calculation of local material data (e.g. based on hardness, residual stress, and oxidation) by means of empirical formulae. The research of the influence of material defects, such as pores or inclusions, in the context of weakest-link models, has shown that the material fatigue depends on the distribution of defect size within the material. Models for the consideration of the defect size on the tooth root strength, such as the model according to Murakami, have not been applied in fatigue models for gears yet and are focused on in this report.

Therefore, the objective of this work is to introduce a method for the calculation of the tooth root load carrying capacity for gears, under consideration of the influence of the defect size on the endurance fatigue strength of the tooth root. The theoretical basis of this method is presented in this paper as well as the validation in running tests of helical and beveloid gears with different material batches, regarding the size distribution of inclusions. The torque level for a 50 percent failure probability of the gears is evaluated on the test rig and then compared to the results of the simulation. The simulative method allows for a performance of the staircase method that is usually performed physically in the back-to-back tests for endurance strength, as the statistical influence of the material properties is considered in the calculation model. The comparison between simulation and tests shows a high level of accordance.