After recovery it was decided to test termination 4A to 

 destruction. We constructed a tensile test machine using 

 scrap I-beams and a hydraulic ram. The termination was 

 pulled in 500 lb, tension increments, and held for at least 

 4 hours at each level. X-rays were taken at the start, after 

 the 8000 lb. proof test level, and after destruction. The 

 8000 lb. proof test level was applied at both room tempera- 

 ture and at 40° F. Failure occurred at 15, 000 lbs. tension (Fig. 4-3) 

 since the armor was designed for 15, 000 lbs. ultimate strength, 

 the test proved conclusively the adequacy of the eppxy-potted 

 termination for static loads. A cyclic load test is considered 

 necessary to support development of a long-lived system, and 

 will be conducted if funds become available. 



Also subsequent to recovery, the manufacturer attempted an 

 8000 lbs. proof test of Shot 2B-3A. In this test, termination 

 3A failed at 6800 lbs. Station 3 had been proof tensioned to 

 6400 lbs. during loading on the barge in November! The mode 

 of failure was identical to that experienced at termination IB 

 i. e.. the armor strands slipped out of the epoxy plug due to 

 inadequate potting or improper preparation of the armor strands. 

 Fig. 4-2 illustrates this failure mode. During subsequent 

 proof testing, the armor slipped out of termination 3B at 6500 

 lbs! 



Clearly the failures during installation and subsequent 

 proof tests could be traced to inadequate quality control. The 

 manufacturer took steps to remedy the problem, and we de- 

 vised a procedure to inspect and proof test all signal cable 

 shots at delivery. It was decided to pull the complete length 

 of all shots to prevent damage to the cable during testing and 

 to provide confidence that no serious damage had been sustained 

 during deployment, retrieval, or shipping; or because of sub- 

 sequent corrosion. The procedure and results obtained are 

 given in Table 4-3. The test "facility" used for pulling the long 

 shots of cable consisted of two Schlumberger well- logging 

 winch trucks at either end of a straight stretch of road. Some 

 testing was done on a stretch of new, unopened highway in 

 Houston, and the balance was conducted on a dirt farm road 

 Northwest and outside the city limits of Houston. Photographs 

 are shown in Fig. 4-4. 



4, 1. 3 Testing of Standard Hardware and Special Fittings 



In this section we shall review the testing of standard hard- 

 ware such as shackles, chain, and synthetic rope, and special 

 fittings such as couplings, bridle, head frame, etc. The 

 standard articles are produced under somewhat uncertain 

 quality-control procedures for a competitive market and are 

 generally recommended for use with safety margins larger 

 than we can permit. Consequently we have adopted a 



220 



