Prof. Woodward, in his review of the influence of coercive force on iron losses at low frequencies [Proc. IEE, 1966, 113, (4), pp. 710-716], relates eddy-current losses to domain-wall movement, and domain-wall movement to coercive force, to establish that the hysteresis loss per cycle is a function of frequency. In this way he reaches the conclusion that the anomalous eddy-current loss is at least partly due to this frequency-dependent hysteresis phenomenon.Much of Prof. Woodward's support for this conclusion depends upon the argument that hysteresis loss is a minimum at low frequency, because only domain walls having the lowest coercive force will move. As the frequency is increased more walls are moving together, although the peak flux density is unchanged, and, consequently, a higher effective coercive force is operative so that hysteresis loss is increased.This explanation seems to be too hypothetical, because it fails to take adequate account of the all-important fact that hysteresis is an irreversible phenomenon arising from unstable domain-wall movement. Once a domain wall becomes unstable as a critical magnetising field is reached, an element of hysteresis loss in the form of a microscopic eddy-current effect results. The hysteresis loss is then increased if the magnetising field is increased further before the domain wall has regained stability in its new position. The field then accelerates the change and increases the unstable eddy-current loss, and therefore the hysteresis loss. In short, the hysteresis loss becomes frequency-dependent for any given wall movement. An increased number of wall movements is not the prerequisite; faster wall movement is sufficient.The frequency dependence of hysteresis loss was expressly investigated by the author* by studying the eddy-current anomaly in solid steel cores where domain size in relationship to the specimen dimensions is less critical, and low-frequency skin effects are induced.The limiting phase angle of the magnetising power at high skin effects is a good guide to distinguishing true eddy currents from hysteresis-type loss, because with high eddy-current skin effect the limiting phase angle becomes (TT/4) -(0/2) where fi = /x 0 exp (-j6) is the complex-permeability representation of the specimen, and 6 is a loss angle related only to hysteresis. The author's measurements indicated a steady increase in 6 with increasing frequency, and provided clear evidence that hysteresis loss was, in fact, increasing with frequency.This was further verified by measurements with a polarising field. The hysteresis loss and coercive force are increased by the action of a polarising field, and the theory as presented indicated that the frequency increase would be inversely proportional to the coercive force. Thus, the rate of increase of hysteresis loss with frequency should decrease with increase of polarising field. This was verified experimentally.To summarise, while agreeing with Prof. Woodward that hysteresis loss is frequency-dependent, I do not agree that it results ...
