Abstract
Nowadays, the use of technologies to increase productivity, reduce time, as well as reduce the possibilities of errors, has become indispensable. All processes have opportunities for improvement, and this can be done based on calculations that with the support of computational systems can be reduced considerably in time. In the heat treatment industry and more specifically in the electromagnetic induction heat treatment industry is no exception. Today we have numerous tools to optimize the design process of inductors used in heat treatment of metals. These tools can show us, in a virtual way, the results that we can obtain before having to manufacture the inductors, all this based on FEA (Finite Elements Analysis) simulations that performing calculations considering physical parameters approximate us to what we would have as a result. Computer based simulation programs for induction heating and resulting metallurgy are extremely useful in developing tooling and process for induction heating. Induction hardening simulation brings elements of inductor design, steel properties such as time-temperature-transformation curves, both thermal and magnetic properties at various temperatures and cooling rates based on the phase of the quench media on cooling. A common method in place hardening (static hardening) knows as single shot hardening. In this process, the inductor is designed with a top and bottom half loop connected by heating rails. The length of heating is determined by the length of the rails and percentage height of the width of the half loops. Accurately predicting the length of the heating pattern in this 3D modeling approach is computationally a heavy load on the modeling pre-requisites. Commonly the inductor is modeled and then tested with the actual results showing a different length than what was predicted. It is important to consider that like any system, these simulation tools are not infallible and have several factors that can affect the accuracy of the simulation results. This paper reaches into the analysis of why the predicted length may differ prom the test results discussing what factors constitute the largest variance from the predicted outcome. Inductor design and the reliance on set up will be discussed.