PHASE DISTORTION AND PHASE DISTORTION CORRECTION 207 



Fig. 1 on comparing curve (2) with curve (1). Curve (2) is computed 

 from the formula ^ 



A{t)=~ 



= 1 r 



sin twdw, 



(17) 



where q:(co) is the real component of the transfer admittance of cable 

 and network combined (equation (14)). 



1.0 

 0.9 

 0.8 

 0.7 

 0.6 

 0.5 

 0.4 

 0.3 

 0.2 

 01 



SOLID LINE CURVE^ 

 REPRESENTS RELA- 

 TIVE AMPLITUDE OF AR-" 

 RIVAL CURRENT ON 500 

 MILE CABLE WITHOUT TERM- 

 INAL PHASE DISTORTION COR- 

 RECTIVE NETWORK. 



DOTTED LINE CURVE REPRESENTS RELA- 

 TIVE AMPLITUDE OF ARRIVAL CURRENT ON 

 500 MILE CABLE WITH TERMINAL NETWORK TO 

 CORRECT PHASE DISTORTION OVER RANGE 0-25 

 CYCLES PER SECOND. 



5 10 15 20 25 



FREQUENCY IN CYCLES PER SECOND 



Fig. 9 — Amplitude variation on long telegraph cable 



The curves of Fig. 9 show that this network affords some attenuation 

 equalization as well as very good phase correction. Amplitude and 

 phase correction, as we have seen, are analytically independent 

 processes. Nevertheless, some arrangements may, theoretically, be 

 designed to correct amplitude and phase simultaneously. A method 

 for so designing a network similar to the one under discussion at 

 present has been developed by O. J. Zobel.^° In such cases, however, 

 in order to obtain physically desirable values in practical applications, 

 it has usually been found necessary to design the network for one 

 purpose, thereby automatically obtaining some improvement in the 

 other respect, as in the present instance. 



The maximum phase displacement obtainable with one section of 



1° This is discussed in a forthcoming paper by O. J. Zobel. 



