WIDE-BAND OPEN-WIRE PROGRAM SYSTEM 



377 



to the circuit, one being measured and the other alternately cut off 

 and on to determine whether one frequency adversely afTected the 

 transmission of the other or produced undesirable sum and difference 

 products. Such distortion effects were found to be small. Measure- 

 ments were made to determine whether the overall transmission varied 

 with the load applied. With a testing power which was varied in 

 magnitude from 50 milliwatts to .1 milliwatt, the transmission varied 

 slightly more than 1 db, that is, with the heavy load the circuit loss 

 was somewhat more than 1 db greater than at the light load. 



A noise and crosstalk survey was made on these program circuits 

 and on message circuits on the same pole lead. Observations were 

 made at the terminals of the message circuits while a program was 

 being transmitted on the program circuits to determine the amount of 



1 CIRCUIT NO. 2, SAN FRANCISCO TO CHICAGO, WEST 



2 CIRCUIT NO. 1, CHICAGO TO SAN FRANCISCO, EAST 



3 LOOP, SAN FRANCISCO TO CHICAGO TO SAN FRAN 

 2 WEST TO EAST, 1 EAST TO WEST 



CIRCUIT NO. 2 2689 MILES 29 AMPLIFIERS 



CIRCUIT NO. I 2395 MILES 23 AMPLIFIERS 



LOOP 5084 MILES 52 AMPLIFIERS 



MINUS VALUE INDICATES TRANSMISSION IS DOWN 



FROM lOOO-CYCLE VALUE 



TO EAST 

 TO WEST 

 CISCO, 



:;:2^ 



30 



100 



500 1000 



FREQUENCY IN CYCLES PER SECOND. 



10,000 



Fig. 24 — Average transmission frequency characteristics. 



interference introduced into the message circuits from the program 

 circuits, and, conversely, observations were made on the program 

 circuits while various paralleling message circuits were in use, and the 

 resulting interference was recorded. 



The noise or crosstalk volume on the program circuits was measured 

 by means of a volume indicator, which had inserted between it and the 

 circuit at the point of measurement a network having a loss-frequency 

 characteristic such that the various frequencies affecting the meter 

 reading were attenuated or weighted in much the same way that the 

 ear weights the different frequencies. Crosstalk volume and noise on 

 the message circuit were measured with an indicating meter in much 

 the same manner except that the network used here had an attenuation 

 frequency characteristic corresponding very nearly to that of the ear 

 and an average telephone set. The network used on the program 



