1362 THE BELL SYSTEM TECHNICAL JOURNAL, NOVEMBER 1953 



51 

 H 4 



^-3 



< -4 



a. 



MAXIMUM VALUE OF//' 

 APPROXIMATELY 30-40 



0^ 



'EFF 



Fig. 17 — Exaggerated scale plot of the real part of the effective permeability 

 versus effective dc magnetic field. 



an increase in d must be accompanied by a decrease in m+ to maintain 

 resonance. For convenience we reproduce in Fig, 17 the behavior of the 

 real part of the effective perhieability as a function of field. We see that 

 above resonance the permeability starts at a very high value and then 

 decreases as we move toward higher fields. Thus as we increase the 

 diameter, d, we would expect the "cavity" resonance to move to the 

 right, and since in doing so we go further from the ferromagnetic reso- 

 nance absorption we expect the height of the peak to increase due to the 

 fact that the loss tangent of the permeability is smaller, and hence the 

 effective Q of the "cavity resonance" is larger. Finally we observe that 

 the broadening of the peak which occurs in the larger diameters is due 

 simply to the fact that the slope of the /x+ curve decreases as we move 

 to the right so that a larger change in dc field is required to take m+ 

 through a given variation. 



A similar movement of the subsidiary hump can be effected through a 

 change in the dielectric constant of the medium surrounding the ferrite 

 element. However, altering the total length of the ferrite in steps of 

 0.030" until the total change was greater than 0.250" produced no no- 

 ticeable effect upon either the relative heights or positions of the two 

 peaks. 



