44 



BELL SYSTEM TECHNICAL JOURNAL 



the effect of the convergent variation of the resistive and reactive 

 components of the line impedance, that is, the resistance increases 

 and the reactance decreases with increase in length of line. This lag 

 is greater for the higher frequencies. The total variation for sixty 

 miles of cable measured at twenty cycles is practically ninety degrees 

 which means that for a given field current the sensitivity using twenty 

 cycles must approach zero with some length of line between zero and 

 sixty miles. This condition is illustrated in Fig. 13, where the sen- 



^ 6 



■^ 3 



g 







10 



20 



30 



40 50 60 



Length of Line in Miles. 



Fig. 13 — Sensitivity to unbalance for the bridge network for impedance measure- 

 ments on lengths of non-loaded 19-gauge cable. These curves represent sensitivities 

 for frequencies of testing potential of four, eight and twenty cycles. 



sitivity for twenty cycles is a maximum at twenty to thirty miles, 

 but falls rapidly toward zero at the longer lengths of line. With the 

 lower frequencies and the setting of field current used, the general 

 effect is an increase in sensitivity as the length of line increases, and a 

 decrease in sensitivity with decrease in frequency. This decrease is 

 due to the decrease in reactance with increase in length of line (Fig. 10). 

 It would appear that provision should be made for shifting the phase 

 of the field current through ninety degrees, its two positions being 

 respectively in phase with the bridge potential and ninety degrees 

 leading. Thus the two components, r and Xb, could be balanced 

 independently except for the shift in phase with different line lengths. 

 This phase shift is small with a frequency of testing potential of four 

 cycles. 



