Comparative analysis among Steinmetz-based analytical models applied to inductors using iron powder material considering DC bias

Abstract

In this work, core loss in inductors with DC bias is analyzed, with different flux densities, frequencies and duty cycles, while comparing Steinmetz-based models (OSE, MSE, IGSE and WSE) to experimental results. The losses estimated on a finite element analysis (FEA) software simulation are included in the comparison. Experimental results indicate the most accurate model according to waveform using iron powder (High Flux) material. As contribution, two main insights are given. The first is that the influence of DC Bias in High Flux material has been found to be negligible, which has not been shown. This means that Steinmetz-based models which do not consider DC Bias may be used to estimate losses in the material. The second insight is in regard to the usage of different Steinmetz-based analytical models. It is demonstrated that the actual Steinmetz coefficients that are used influence the results greatly, much more so than the difference among the models themselves. To demonstrate this fact, it is shown that the original Steinmetz equation, the simplest analytical model, is sufficient to describe core losses, given proper coefficient adjustment.