The energy loss per cycle in a ferromagnetic material subjected to small alternating fields is sometimes separated into three parts; the first due to eddy current loss, the second to hysteresis presumed to follow Rayleigh's laws at these flux densities. The origin of the remainder is unknown and is the subject of much controversy; it has been variously termed ``magnetic viscosity,'' ``after effect'' and ``square law hysteresis.'' In studying energy loss in a ring of compressed iron dust, hysteresis loops have been measured ballistically by a new method with a relative error inBmas low as 0.01 percent. The range in maximum flux density is from 2 to 100 gauss. The loops are lenticular and very slender,Bmbeing about 1500 times the remanence for the smallest loop. The smaller loops are at flux densities considerably below those investigated by Rayleigh. His findings as to variation of area, of remanence, and of permeability with loop amplitude are confirmed and extended through the new range of flux densities, though the shape of the loops is not as simple as that he proposed. The energy loss per cycle is proportional toBm3while the remanence is proportional toBm2. The ballistic measurements have been compared with a.c. bridge measurements on the same specimen. The loss of unknown origin is not included in the hysteresis loss measured ballistically. Comparison is made between the third harmonic induced voltage computed by a Fourier series analysis of the ballistic loops and the harmonic actually generated by the specimen. Agreement is found between the measured and the computed values. Possible explanations of the discrepancy between the ballistic observed energy losses and a.c. findings are discussed.
展开▼
