IRON UNDER PERIODIC MAGNETIZING FORCES. 93 



7.— EXPEKIMENTS ON THE BEHAVIOUE OF IRON UNDER PERIODIC 

 MAGNETIZING FORCES. 



By PROFESS'OR T. R. LYLE, M.A., Sc.D., and J. A, GRAY, B.Sc. 



I. This research was undertaken for the purpose, principally, o£ 

 findiug out whether any simple relation as regards relative amplitude 

 or phase holds either between the upper harmonics of an induction 

 oscillation in laminated iron and its fundamental — or (as we call it) 

 its first harmonic — or between its successive upper harmonics, when 

 the magnetizing force is approximately sinusoidal ; it was hoped that 

 some hint might thus be obtained concerning the link by which these 

 harmonics, all of odd order, are bound together, and, hence, some con- 

 ception rendered possible as to the nature of the oscillating system in 

 an ultimate magnetic particle when excited by periodic magnetizing 

 forces. 



Thus, if the sinusoidal magnetizing force — 



H = H, sin o)f. 



— be applied to a laminated iron ring in the usual way, it is known 

 that the induction B produced contains the full series of odd har- 

 monics and may be written — 



B = B, sin (cot — e;) + B, sin 3 {U — 6^ + B, sin 5 (o)^ — ^,) + &c. 

 We wished to find out whether any simple relation connected B , Bj, 

 &c., (9,, 0^, 6., &c., either to B^ or to one another or to the period of 

 the oscillations. 



In this we were not successful, but we think that the results we 

 obtained will be of value, especially to those interested in magnetic 

 research. 



In addition to the above purpose we wished to test, in a more 

 accurate manner than has hitherto been done, the correctness of some 

 conclusions given in a former paper* by one of us, relating to — (a) 

 what was called kinetic hysteresis ; (b) the formula proposed in the 

 same paper for the total iron loss I as a function of the frequency n, 

 and what was called the effective induction 'iB, namely — 



where ^ is a number about 1"5 — I'G ; (c) the modifications produced 

 on the characteristics of a B wave by change of wave form of the 

 magnetizing force H. 



2. In the paper already quoted, it was found that when H was 

 approximately sinusoidal B,/B,, B./B, were, after B, had attained a 

 certain value, practically linear in B, ; but it was pointed out that the 

 values of Bj, B,, &c., obtained, were probably much modified by a 

 peculiar action in the iron ring by which some of the energy trans- 

 mitted to it by the magnetizing force H sin lot was reflected back to 

 the magnetizing circuit in the form of currents, whose frequencies 

 were 3, .5, 7 times respectively that of the exciting current. 



* Variations of Magnetic H.ysteresis with Frequency. Phil. Mag., Jan., 1905. 



