The Measurement of Magnetic Hysteresis. 



349 



Now let the earth inductor be inverted, and so produce a change of 

 induction P, and let the primary current at the time be C, then 



Ko/ = qC | cdt = 2 CT/S. 



If 0i, 2 be the two throws which occur when C changes from C to 

 - C and from - C to C , and if be the throw due to the earth 

 inductor, then 6/<f> = w/w' and thus for a complete cycle, 



Thus the sum of the two throws 6\ and 2 is a measure of the 

 energy dissipated in hysteresis in a complete cycle. When the factor 

 C PN/A«(p has been determined, measurements of hysteresis can be 

 made as rapidly as measurements of induction with a ballistic galvano- 

 meter. 



§ 2. In developing a more complete theory the authors employ the 

 equations 



E = RC + ^(NZAB + L'C + Mc), 



= Sc + ^(wAB + MC + Lc). 

 at v f 



With the aid of the principle of the conservation of energy, these 

 equations lead to the result 



= U-X-Y. 



Here cr is the specific resistance of the specimen, and Q a numerical 

 constant depending upon the geometrical form of the section, having 

 the value 1/Stt or 0*03979 for a circle and 0*03512 for a square. 



The term U is determined by the dynamometer throws. The term 

 X is the energy dissipated in eddy currents in the specimen during the 

 two semi-cycles, and Y is roughly the energy spent in heating the 

 secondary circuit. 



It is shown that Y, when appreciable, can be determined by making 

 two observations for U with two different values for S. In the 

 authors' experiments Y was nearly always negligible. When a 

 suitable key is employed to reverse the current, X + Y can be 

 determined by making two observations for U with two different 

 resistances of the primary circuit, the E.M.F. being at the same time 

 so altered as to produce the same maximum current Co in each case. 



