NF.nroKKS cox 1. 11 mm; /jro Hn.icT.-txcF.s mi 



ustxl in !•"!>;. .'Ui and Za is ;ui indurtani'i- of I.O lu-nry. This inipt'dancr 

 curve has four parts, (wo in carii shed. It starts on tlu- resistance 

 axis at the intersection of the Z3=oc and ^3 = circles. As the fre- 

 quency increases from zero the first part of tiie cur\e is traced out in 

 Sheet II. At 2'i cycles the impedance is approximately HlO+iHO 

 The reactance com[)i)nent has a maximum of aboiil 2')() ohms at about 

 70 cycles, the resistance component has a maximum of about 720 

 ohms at alwut U»0 cycles, the reactance component has a minimum 

 of alxiut —110 ohms at about 300 cycles, and finally at about 480 

 cycles the curve reaches the inner boundary, whereupon it changes 

 to Sheet I. It remains in .Sheet I up to a frequency of about 910 

 cycles, the resistance component having a minimum and the react- 

 ance component a maximum, which may be read from the diagram. 

 The im|X'dance between 010 cycles and approximately 1,300 cycles 

 lies on Sheet II, and from 1,390 cycles to infinite frequency on Sheet I. 

 The resistance component has a total of three maxima and three 

 minima, and the reactance component three maxima anfl two minima, 

 following the cyclical order: /?-minimum, A'-maximum, /^-maximum, 

 A'-minimum. 



An interesting e.xercise is to obser\e the effect on the impedance 

 curve of changing the value of the inductance Z3. The curve inter- 

 sects the Zj-constant circles at the same frequencies in each case, 

 l)Ut the points of intersection are moved in a clockwise or counter- 

 clockwise sense as Z3 is increased or decreased. With each such 

 change parts of the impedance curve disap[)ear from one sheet and 

 reappear on the other. For instance, with a decrease of the inductance 

 Zj the first loop of the impedance curve on Sheet II shrinks, and with 

 sufficient decrease in inductance may become too small to plot, al- 

 though it does not disappear entireK'. 



It is evident that if Z; and Z,i are formed of reactance networks of 

 greater complication the impedance curve may be very involved. 

 But no matter how tortuous its path, it is restricted to the impedance 

 region, that is, to the ring-shaped region between the non-intersecting 

 boundary circles determined by the resistance network alone. 



My thanks are due to Dr. George A. Campbell for his stimulating, 

 continued interest, and to Mr. R. M. Foster for suggestions on every 

 phase of this work. 



