A comparison of gear tooth bending fatigue lives from single tooth bending and rotating gear tests

Gear tooth bending fatigue failures are instantaneously catastrophic to gear drive power transmission systems. For this reason, gear designers must understand the limits of their design with respect to the desired application and service time. Fatigue testing on gear specimens has been preferred metric on which to base future designs. Single Tooth Bending test (STB) or a Rotating Gear (RG) test methodologies have been used for this purpose. STB type tests generally form the large majority of gear fatigue testing due to cost and availability but does not fully simulate the actual operating conditions of rotating gears in service. As RG evaluations are costly and time-consuming, it is desirable to quantify how a stress-life (SN) relationship regressed through STB testing compares to that produced in RG testing. In this study, both STB and RG test methodologies are employed to test the same specimen design. Matrices of fatigue tests are executed and statistical regression techniques are used to estimate bending fatigue lives as a function of stress for both sets of data. The resultant SN curves are compared to determine any differences in allowable stress. Techniques are then employed using single set data (STBF or RG individually) to demonstrate the calculation of correlation coefficients, which can approximate the total difference determined between the two data sets.