Volume 1, Issue 1
Optimization of an Induction Hardening Process for a Steel Gear Component
Computer simulation of heat treatment processes has improved significantly over the past two decades, relating to both the material models and the database accuracy. Simulations are being used more aggressively in part and process design rather than just as a trouble shooting agent, meaning heat treat simulation is maturing as an accepted technology. In this paper, an induction heating and spray quenching process of a steel gear is optimized using the commercial heat treatment software DANTE. As a hardening process, induction hardening of steel parts is gaining popularity due to: (1) the process is consistent on a part-to-part basis, making it easier for quality control; (2) the process is more environmentally friendly than oil or polymer immersion quenching because a water spray is the typical cooling agent; and (3) fatigue life can be improved due to higher and deeper residual compression in the hardened surface. Two steel grades, AISI 5120 and AISI 5130, may be used in a thin-walled spur gear. The simplified heat treatment process includes vacuum carburization, controlled cooling to ambient, induction heating, and spray quenching. During induction heating, the internal heat generated by eddy currents is used as direct input to drive the thermal model. A sensitivity based optimization method is combined with heat treatment simulation to optimize the delay and spray quenching practice after induction heating. The objective function is defined to minimize the distortion of the gear tooth while satisfying the residual stress, gear surface temperature, and final microstructure requirements. The results of a two-dimensional (2D) plane strain single tooth model are evaluated during the optimization process, the quenching practice variables are adjusted, and the quenching model is re-run until the optimization function is satisfied. The residual stresses predicted after induction hardening are then mapped to a three-dimensional (3D) whole gear model as the initial stress state of a tooth loading model for predicting gear stresses. The significant effect on fatigue behavior due to residual stresses from heat treatment is addressed.