ELECTRON TUBES FOR A TRANSATLANTIC TELEPHONE CABLE 167 



value was later lowered when other operating conditions were changed. 

 Subsequent results showed that in this range the voltage effects on 

 thermionic life were relatively negligible. 



The third assumption, that low cathode current density fa\'ored longer 

 thermionic life, affected the tube design by suggesting the use of a large 

 coated cathode area. This implied the use of relatively high cathode 

 power. It was decided early in the planning of the repeater that the 

 voltage drop across the three heaters operated in series would be used 

 to supply part or all of the operating plate and screen potentials. For a 

 60-volt plate and screen supply, the heater voltage could be as high as 

 20 volts. A quarter of an ampere was considered a reasonable cable cur- 

 rent consistent with voltage limitations at the cable terminals. Thus 5.0 

 watts were available for each cathode. With this power, a coated area of 

 2.7 square centimeters was provided. The "\'alue of the cathode current, 

 the cathode area, and the interelectrode spacings define the transcon- 

 ductance. Very liberal interelectrode spacings were provided consistent 

 with reasonable tube performance. The original design called for a spac- 

 ing of 0.040 inch between control grid and cathode. This value was later 

 reduced to 0.024 inch, and a satisfactory design was produced which 

 gave 1,000 micromhos or one milliampere per volt at a cathode current 

 drain of approximately 2.0 milliamperes. The resulting current density 

 of approximately 0.7 milliampere per square centimeter is in sharp con- 

 trast with values such as 50 milliamperes per square centimeter used 

 currently in tubes designed for the more conventional communication 

 uses. Subsequent data, discussed later, indicate that for current densities 

 of a few milliamperes per square centimeter, the exact value is not 

 critical in its effect on thermionic life. 



Subsequent Production Programs 



The development of the tube was pursued on an active basis through 

 the years leading up to World War II. During the war development ac- 

 tivity essentially stopped. It was only possible to keep the life tests in 

 operation. After the war the development of the tube was completed and 

 a small production line was set up in Bell Telephone Laboratories under 

 the direct supervision of the tube development engineers to make and 

 select tubes for a cable between Key West, Florida, and Havana, Cuba. 



This cable turned out to be a "field trial" for the transatlantic cable 

 which was to come later. A total of G submerged repeaters containing 18 

 tubes were laid and the cable was put in operation in June, 1950. The 

 cable has been in operation since this date without tube failure, and 



