MEASUREMENT OF PHASE DISTORTION 



531 



The following method of measurement is given for those cases where 

 the method already described is not sufficiently accurate because of the 

 smallness of the impedance cycles for systems having large attenuations 

 or delays." It is assumed here that the characteristic impedance of 

 the system is the same for both directions. 



K = CHARACTERISTIC IMPEDANCE OF SYSTEM 



I'ig. 4 — Modified arrangement for special impedance measurements. 



(Open termination.) 



Impedance measurements are made as before with the following 

 changes as shown in Fig. 4: (1) X/2 is bridged across the measuring 

 trunk in place of K; (2) instead of having the output of the network 

 short-circuited or open, the output is bridged on the input. This case 

 corresponds somewhat to the one above in which the open termination 

 was used on the output of the system, with the exception that the 

 return current now traverses the system only once in its complete 

 trip from the bridge back to the bridge and, consequently, is attenuated 

 and delayed only one half as much as before. These impedance curves 

 are quite similar to those obtained above and may be interpreted in a 

 similar manner. In this case, the envelope delay of the system in 

 seconds at the frequency {J2 +/])/2 cycles is, approximately. 



T. 



1 



2(/2-/i)' 



where (/2 — /i) is the length in cycles per second of one half an im- 

 pedance cycle of an impedance curve. Fig. 5 shows the computed 

 impedance curves resulting from measurements made in this way on 

 the same 100-mile unit of phase corrector and also gives the correspond- 

 ing envelope delay-frequency characteristic of this network. 



The case just described corresponds to the former case with the open 

 end termination; that is, the results obtained are the same as those 

 which would be obtained from the former case by using only one half 



" D. K. Gannett, U. S. Patent 1,725,756. 



