TELEPHONE TRANSMISSION OVER LONG CABLE CIRCUITS 77 



Before leaving the subject of "echoes" it is believed that it will be 

 of interest to point out some of the important characteristics of two- 

 way repeatered circuits which result from these effects. 



1. The minimum permissible net equivalent (total loss minus total 

 repeater gain in one direction) of a four-wire circuit of a given 

 length depends only on the velocity of propagation and the 

 balance conditions at the terminals of the circuit. When con- 

 ditions are such that the balance conditions cannot be improved, 

 increasing the velocity of propagation will enable a lower net 

 equivalent to be obtained. 



2. In the case of a two-wire circuit with reasonably smooth lines, 

 the exact location of the repeaters and the gains at which indi- 

 vidual repeaters are worked have little effect on the overall 

 result so far as echo effects are concerned. This follows from 

 the fact that the echo paths from end to end of such a circuit 

 are usually of more importance than the shorter echo paths. 

 Evidently, moving the individual repeaters about or altering 

 their gains has no effect on the longest paths, provided the 

 total gain in each direction is kept constant. 



3. In the case of a two-wire circuit of a given length, the velocity 

 of propagation and smoothness of the lines are of most impor- 

 tance in limiting the possible net equivalent, the line attenua- 

 tion being of secondary importance. 



For example, in the case of the transcontinental (New York- 

 San Francisco) open-wire line, the original circuit was loaded. 

 (Although this paper deals particularly with repeaters on cable 

 circuits this example was selected because it so well illustrates 

 this point.) The velocity of propagation was such that voice 

 currents required about 0.07 second to travel from one end of 

 the circuit to the other. The total line equivalent was equal 

 to about 50 miles of standard cable. By applying repeaters 

 to this circuit it was possible to obtain a working net equivalent 

 of about 21 miles. 



The unloading of the circuit increased the velocity so that the 

 time of transmission was reduced to 0.02 second, about 0.3 

 of the time required when the circuit was loaded. The attenua- 

 tion was increased so that the total line equivalent without 

 repeaters was equal to about 120 miles of standard cable, a 

 little more than twice the equivalent of the loaded circuit. By 

 applying repeaters of an improved type to this circuit so as to 

 keep the quality good in spite of the increased attenuation and 



