Bending fatigue design of case hardened gears based on test specimens

Different design methods against bending fatigue are reported in ISO 6336 standard. The standard suggests primarily the method based on reference test gears and provides the relevant fatigue curves. Additionally, the standard suggests the use of specimens (instead of gears) to generate the reference fatigue curves, but it also advices that specimen-based methods can be used when gears are not available and that specimens are particularly useful for comparing fatigue performances of gear materials relative to one another. The purpose of the present paper is to evaluate the accuracy of the specimen-based methods mentioned in the ISO standard when applied to design gears against bending fatigue. Experimental data were generated by means of pulsator fatigue tests on case-hardened gears used in off-highway vehicles. Afterwards, experimental results were compared with theoretical estimations according to the approaches based on reference test gears (as suggested by the ISO standard) and test specimens. Concerning the latter approach, the relevant fatigue design curves were generated by testing smooth as well as notched specimens made of the same case-hardened gear steel. It was found that the specimens-based methods are as accurate as the reference gears-based method, provided that the material notch sensitivity factor is properly calibrated on the experimental results obtained from specimens