Induction hardening, a promising approach for selective hardening of metal parts, is widely used for surface hardening, where a hard surface is required alongside a tough core. Regarding the complexity of this process, parts’ geometry deeply affects the temperature distribution and hardness profile accordingly. In this study, two magnetic flux concentrators are introduced to our induction machine set in order to control the magnetic flux and consequently hardness profile (case depth) of spur gears. The performance of magnetic flux concentrators is examined by the effect of machine parameters on the case depth and the edge effect of AISI 4340 steel-made spur gear. Design of experiments based on Taguchi method is primarily used to optimize the number of experimental trials. Then, the hardness profiles of heat-treated gears at the tip and root of gears are measured by microindentation hardness tests. The results are analyzed using analysis of variance (ANOVA) and response surface methodology (RSM) to determine the main effect of process parameters, also the best combination of process parameters that maximizes the case depth and minimizes the undesirable feature of edge effect. Finally, the predicted case depth models versus process parameters are developed based on linear regression method. To this end, four predictive models of case depth at tip and root in the edge plane and middle plane of spur gears are generated. Results imply that maximum case depth with minimum edge effect at root and tip is achieved by setting up the highest machine power, longest heating time, and minimum axial gap between concentrators and the spur gear. This study provides a good exploration of case depth in presence of magnetic flux concentrators under various process parameters and gives a reliable guideline towards edge effect during induction hardening process.

感应淬火是用于金属零件选择性淬火的一种有前途的方法，被广泛用于表面淬火，在这种表面淬火中，需要在坚硬的铁芯旁边加上硬质表面。关于此过程的复杂性，零件的几何形状会相应地影响温度分布和硬度分布。在这项研究中，两个磁通量集中器被引入到我们的感应电机组中，以控制正齿轮的磁通量，从而控制正齿轮的硬度分布（表面深度）。通过机器参数对机壳深度的影响以及AISI 4340钢制正齿轮的边缘效应来检验磁通集中器的性能。基于Taguchi方法的实验设计主要用于优化实验次数。然后，通过微压痕硬度测试测量齿轮末端和根部的热处理齿轮的硬度分布。使用方差分析（ANOVA）和响应面方法（RSM）对结果进行分析，以确定过程参数的主要效果，以及过程参数的最佳组合，可以最大程度地增加案例深度并最小化边缘效应的不良特征。最后，基于线性回归方法建立了预测的案例深度模型与过程参数的关系。为此，生成了四个正齿轮端面和中部端面和齿根深度的预测模型。结果表明，通过设置最高的机器功率，最长的加热时间，可以实现最大的外壳深度，并且在根部和尖端处具有最小的边缘效应，集中器和正齿轮之间的最小轴向间隙。这项研究很好地探索了在各种工艺参数下存在磁通量集中器的情况下的深度，并为感应淬火过程中的边缘效应提供了可靠的指南。