A MAGNETIC CURVE TRACER 



31 



deflection is produced by an air core mutual inductance, Xi is found to 

 be 0.134 sec. This determines Xo as 0.180 sec, in disagreement with 

 the value 0.055 sec. calculated from Eq. 5. Johnson assumes in his 

 derivation that dB/dll is constant and hence that the shape of the 

 curve before He is reached has no effect on A/Zc- It is probable that 

 if the equation were changed to allow for dB/dll being a function //, 



-0 5-0 4 -0.3 -0.2 -0.1 0.1 0.2 0.3 0.4 0.5 



H 



Fig. 13— A hysteresis loop of permalloy containing 78.1 per cent nickel. 



that the difference could be accounted for. At any rate, this error is 

 negligible for all but specimens with exceptionally high dBidH or 

 great thickness, and the true coercive force can always be found by 

 taking two loops with different values of dHldt and extrapolating to 

 dllldt = 0. 



Another possible source of error is the passage of a large fraction of 

 the photo-electric cell current through the search coil, the field being 

 thereby altered. The maximum photo-electric cell current used is on 

 the order of 5(10)-^ amperes. Since the search coil is unlikely to have 

 more than about 400 turns per centimeter, this would make the 

 maximum error in // about 2.5(10)-* gauss, which is negligible for most 

 measurements. 



As a test of the accuracy of the instrument, a comparison was made 

 with curves made by ballistic galvanometer measurements. Fig. 13 

 shows a loop taken of the specimen which Bozorth used in some 

 previous measurements.-* ^ Both the coercive force and the maximum 

 induction taken by the two methods agreed to within less than one 



' R. M. Bozorth, Phys. Rev., 32, 124-132 (1928). 



