Eddy-Current Losses Calculated From a Domain Model and Observed Densities of Mobile Walls

Abstract

Recent measurements involving the superposition of small audio‐frequency field increments on silicon iron sinusoidally magnetized at low frequencies indicate that the number of mobile domain walls varied during the cycle. This experimental technique has been considerably refined to provide a better measure of the number of active domain walls. Based on a simple domain model, a calculation of the number of active walls can be made from the experimental data. The principal objective of this study is an explanation of the anomalies in the loss‐frequency curve of grain‐oriented silicon iron magnetized to high flux densities at low frequency. The data obtained from the superposed incremental magnetization experiment can be directly applied to a calculation of the eddy‐current losses with a simple domain model. Since no better information is available, it is necessary to take the speeds of the walls to be equal to complete the calculation. With this assumption and the audio‐frequency data, the losses are then calculated point‐by‐point over a cycle of magnetization. Comparing the calculations to loss measurements, 60 percent of the anomalous eddy‐current loss and some of the downward curvature of the measured loss‐frequency curve are accounted for in the 0–30 cps range. Considering the assumptions made, this result gives plausibility to the domain model and the interpretation of the audio frequency magnetization data.