246 THE BELL SYSTEM TECHNICAL JOURNAL, JANUARY 1957 



In the finalized plans^ for the Newfoundland-Nova Scotia cable it 

 was required to carry 60 circuits, so that some redesign of the 36-circuit 

 'prototype' repeater became essential. It was accepted, however, as a 

 guiding principle throughout the redesign that there should be no de- 

 parture from previous practice without serious consideration and ade- 

 quate justification. It was obviously not an occasion to experiment with 

 new ideas. 



Post Office and Bell Telephone Laboratories experiences were pooled 

 for the project, and the whole technical resources of both organizations 

 were freely available at all times for consultative purposes. Detailed 

 manufacturing and testing specifications were exchanged and approved, 

 and each party was free to inspect the other's production methods. This 

 mutual interchange was undoubtedly highly beneficial, and in the British 

 case it resulted in a still more rigorous control of manufacturing and in- 

 spection methods. 



PLANNING 



General 



Preliminary design calculations indicated that it should be possible 

 to increase the circuit-carrying capacity of the Anglo-Norwegian proto- 

 type repeater from 36 to 60 circuits, and tests on a model confirmed 

 that, with band frequencies of 20-264 and 312-552 kc, a 60.0 db gam 

 at 552 kc could be realized with satisfactory margins against noise, dis- 

 tortion and overload. This gain fixed the repeater spacing at about 20.0 

 nautical miles so that on the selected route two repeaters would be re- 

 quired on the land section between Clarenville and Terrenceville and 

 14 in the Terrenceville-Sydney Mines sea section. It was noted that the 

 land repeaters might have to work with an ambient temperature 12°C 

 higher than in the sea repeaters. 



Each repeater would need to be energized with a direct current of 

 316 ma at 124 volts so that the total route voltage would be about 2,300 

 volts. This voltage would be quite acceptable to the repeaters, but for 

 normal operation it was proposed^ to feed from both terminals simul- 

 taneously, thereby halving the maximum voltage to ground.^ 



The precise localization of any faulty or aging repeater would be of 

 paramount importance. It was decided to retain the two supervisory 

 methods which on the prototype had worked satisfactorily in this re- 

 spect. These consisted of a pulse-distortion equipment requiring no addi- 

 tional components in the repeater and a loop-gain monitoring set in- 

 volving a special unit in the repeater and the allocation of a 4-kc band 

 (260-264 kc) for its operation. 



