164 THE BELL SYSTEM TECHNICAL JOURNAL, JANUARY 1957 



period of time. These factors suggest as an objective for submerged re- 

 peaters that the tubes should not be responsible for a system failure for 

 many years, possibly twenty, after the laying of the cable. Such very 

 long life requirements make necessary special design features, care in the 

 selection and processing of materials that are used in the tubes, unusual 

 procedures in fabrication, detailed testing and long aging of the tubes, 

 and the application of unique methods in the final selection of individual 

 tubes for use in the submerged repeaters. 



As indicated in the foreword and discussed at length in companion 

 papers, the British Post Office developed the section of the cable system 

 between Clarenville, Newfoundland, and Sidney Mines, Nova Scotia. 

 This part of the transatlantic system uses the 6P12 tube which was devel- 

 oped at the General Post (3ffice (G.P.O.) Dollis Hill Research Station. 

 The submerged portion of this system contains 84 tubes in 14 repeaters. 

 Bell Telephone Laboratories developed the part of the system between 

 Clarenville, Newfoundland, and Oban, Scotland. This section requires 

 102 repeaters, including 30() tubes, of a type known as the 175HQ. 

 Although a common objective in the development of each of the two 

 sections has been to obtain very long life, the tube designs are quite dif- 

 ferent. 



The Bell System decided on the use of a repeater housing that could be 

 treated as an integral part of the cable to facilitate laying in deep Avater. 

 The housing is little larger than the cable and is sufficiently flexible to be 

 passed over and around the necessary sheaves and drums. In such a 

 housing the space for repeater components is necessarily restricted. This 

 space restriction, combined with the general philosophy that the num- 

 ber of components should be held to an absolute minimum and that each 

 component should be designed to have the simplest possible structural 

 features, has resulted in the choice of a three-stage, three-tube repeater. 

 In this design, each tube carries the entire responsibility for the con- 

 tinuity of service. 



The Post Office Research Laboratories, prior to the development of 

 the transatlantic cable system, had concentrated their efforts on shorter 

 systems for shallow water. The placing of the repeaters on the bottom 

 did not present the serious problems of deep-sea laying, so more liberal 

 dimensions could be allowed for the repeater circuit. A three-stage ampli- 

 fier was developed which consisted of two strings of three tubes each, 

 parallel connected, with common feedback. The circuit was so designed 

 that almost any kind of tube failure in one side of the amplifier caused 

 very Httle degradation of circuit performance. This philosophy of having 



