The following is a paper by H. Aspden published by the Institution of Electrical Engineers as Monograph No. 165M (January 1956) and later in Proc. I.E.E. vol. 103C at pp. 272-278 (1956).
MAGNETIC TIME-LAG EFFECTS IN SOLID STEEL CORES
Abstract: The discrepancy between the theoretical eddy-current effects and the actual eddy-current effects in thick steel cores is investigated. By using thick cores the magnetic inhomogeneity arising from ferromagnetic domain configuration is of little consequence and the discrepancy may be attributed to an intrinsic time-lag effect in the magnetization process. The time-lag is measured and explained in terms of a localized eddy-current action associated with the dissipation of hysteresis loss. It was found that the time-lag depended upon frequency and was less important at high frequencies.
Commentary: The experimental research reported in this paper aimed at explaining the mysterious extra loss that occurs in steel when subjected to alternating magnetization. This particular paper concerned the problem in solid steel rods magnetized at kilocycle frequencies as distinct from that in laminated steels at power frequencies. One important distinction between these two cases is that the physical size of magnetic domains and consequent magnetic inhomoheneity should be of lesser relevance in the magnetization of solid cores, thereby allowing one to see if the eddy-current loss anomaly could be eliminated in the latter situation and confined to the sheet steel laminations where magnetic domain inhomogeneity is prevalent.
In the event it was found that in the solid steel cores there is indeed an anomalous loss resembling an eddy-current effect, but it is attributable to what one might term the dynamic component of the hysteresis loss. It corresponds to a time-lag effect of the order of 10 microseconds and arises from the increase in magnetizing field during the period taken by the jumps in the movement of the magnetic domain walls. However, this degree of time-lag cannot explain the eddy-current loss anomaly that occurs at the lower power frequency of 50 Hz, where time-lags of the order of 100 microseconds and more are needed to match the eddy-current loss observed.
Accordingly, this paper, though establishing a case for added loss attributable to magnetic time-lag effects has to be considered in the context of the separate reports which were based on the same Ph.D. research programme.
Press [1956b] and [1957a] for further information.