Abstract
This paper presents the development of a strategy to improve current root stress predictions at the edges of spur and helical gear teeth. Industry demands for higher power density gearboxes require maximum utilization of the load carrying capability of the gear. This requires an accurate understanding of the load distributions and the resulting root stresses. Current analytical methods for predicting root stresses include accurate but time-consuming three-dimensional finite element analysis and some special purpose gear programs. The special analysis programs are computationally faster but rely heavily on empirical mathematical formulations which do not easily lend themselves to the inclusion of edge effects. This effect is seen in the experimental and finite element results due to the difference in stiffness between the edges of helical gear teeth and increases with helix angle. Stress predictions from a state-of-the-art empirical program are compared with experimental and finite element results and a correction suggested for this discrepancy.