444 



THE BELL SYSTEM TECHNICAL JOURNAL, MARCH 1957 



O 



z 

 < 



X 



o 



0.03 



0.02 



5C 



+1 



0.01 



1000 



2000 



3000 



4000 



5000 



h",applied field in oersteds 



6000 



7000 



Fig. 8 — Evaluation of xm" ± k" from measurements of A(l/Q)± for low-loss 

 BTL ferrite, saturation magnetization 1300 oersted. 



Whereas it is possible to obtain good agreement between measured 

 and theoretical curves of Xm' ± k from Polder's relations (3), such an 

 agreement cannot be obtained for Xm" ± k" from a comparison with 

 (4). This discrepancy is assumed to be caused by the fact that Polder's 

 theory was developed for single ferrite crystals and did not take into 

 account the random orientation of crystal axes in polycrystalline ma- 

 terials, such as the ferrites used in these measurements. It is reasonable 

 to expect a broadening of the resonance line and a departure from the 

 Lorentzian shape in polycrystalline ferrites; hence, the expression for 

 the absorption line (4) no longer holds. 



Measurement of the electric susceptibility of ferrites did not present 

 any major difficulties, provided some care was used in the suspension of 

 the discs at the cavit}' center. 



Summing up these results it may be said that the disc method has 

 3'ielded satisfactory measurements of the intrinsic parameters of poly- 

 crystalline ferrites below and above resonance as well as in the un- 



