1124 THE BELL SYSTEM TECHNICAL JOURNAL, SEPTEMBER 1957 



Fig. 15 — Corrosion pockets in galvanized steel armor wires of submarine cable 

 after 12 years of service. 



Of special interest is a particular type of corrosion which has occurred 

 in two separate cable samples — one from Alaskan waters, the other 

 from off the southern coast of Newfoundland. In one case the age of the 

 cable was 12 years, and in the other 36 years. The Alaskan sample was 

 located in an area characterized by water velocities of 5 to 9 knots. In 

 both instances, the outer jute and most of the flooding compound was | 

 gone. Corrosion, instead of starting and progressing on the outer sur- ■ 

 face of the wires, had started, and been confined largely to the sides of , 

 the wires. Usually there are corresponding areas of corrosion on two ! 

 adjacent wires to form "corrosion pockets." These pockets are illustrated 

 in Fig. 15. In the case of the 12 year old cable, corrosion caused failure 

 of the armor wires. In the case of the cable which was in service 36 years, 

 failure was reported to have occurred from chaffing on a rocky bottom. 

 Close examination suggests that failure may more reasonably be at- 

 tributed to sevei'e corrosion of the type just described. The exact cause 

 of the corrosion pattern is still to be determined. 



Sufficient samples have not been examined as yet to form a coordi- 

 nated picture with respect to the inner jute. In the case of cable samples 

 from the North Atlantic, the inner jute bedding w^as in good condition 

 in cables which had been in service for as long as 30 or 40 years. Samples 

 more than 40 years old showed the effect of deterioration. Although only 

 a limited number of samples have been examined from the Caribbean, 

 most of them from water about 50 feet deep, jute and cotton tape com- 

 ponents were in poor condition in certain spots. It was evident that 

 microbiological deterioration of the jute had occurred. In no case has 

 there been any evidence of deterioration of the insulation of the central 

 conductor of any of the cable samples. In the case of the older cable 

 samples the insulation was gutta percha, but in the most recent samples 

 it has been polyethylene. 



