prototype connectors. The greatest problem 

 encountered in the disassembly was the effort 

 involved in dissolving the polyurethane potting 

 material. After nearly four weeks of dissolving the 

 polyurethane, the internal components were available 

 for inspection. 



It was later discovered that the polyurethane 

 potting can be quickly removed by first baking the 

 assembly at 400-500°F for 3 hours. The poly- 

 urethane is thermally decomposed to an easily 

 crushed solid that is readily removed. However, this 

 process damages all components that are not metallic. 



Analysis of the internal components of the 

 experimental connectors presented the following 

 results: 



1. Dummy Piston. As shown in Figure 7, the 

 dummy pistons were severely deformed due to cold 

 flow of the glass-filled Teflon material from which 

 they were machined. This deformation accounted for 

 the stiffness of operation of the pistons in the 

 assembled connector. It is believed that this deforma- 

 tion was caused by compression or shrinkage of the 

 surrounding polyurethane material. 



2. Penetration Pins. Figure 8 shows tracking 

 paths between the high-voltage cable terminal and the 

 threaded fitting that is at ground potential. This is 

 positive evidence that the polyurethane did not 

 always seal to the DAP and other materials that make 

 up these penetrators. 



3. P r e s s u r e - E q u a 1 i z i n g Pistons 

 (Compensators). The four pressure-equalizing pistons 

 were displaced by springs that were located on the 

 seawater side of the piston. These springs were 

 severely corroded, and corrosion products con- 

 taminated the seawater in the compensator cylinder. 

 The corrosion products and other foreign material 

 that collected in the cylinder caused the pistons to 

 jam. 



It was also found that the female contact ring 

 was located too close to the side of the urethane- 

 filled cavity. The spacing was less than 1/8 inch, 

 which was a very marginal thickness for the dielectric 

 strength of urethane. The problem was compounded 

 by the fact that an oil flushing duct was drilled 

 through the housing and urethane direcdy into the 

 contact area. This provided a trap for contamination 



and a very short tracking path for high voltage 

 leakage currents. Previous high voltage pulse tests of 

 the assembled connector had indicated that break- 

 down did in fact occur at this point. 



To correct the deficiencies found during the 

 testing and disassembly of the experimental 

 connector, one wet connector set was extensively 

 modified as a prototype. As shown in Figure 9 the 

 modifications include new and more reliable electrical 

 pins, pin-wiping seals, dummy pistons, pressure- 

 equalizing system, electrical insulation, and a 

 Dyna-Grip termination. The new design can now be 

 quickly assembled or repaired at sea because no 

 potting compound is used. 



The modified male pin, female contact 

 assembly, and dummy piston are shown in Figure 10. 

 The male pin contact is gold-plated copper and is 

 insulated with molded high-quality, glass-filled epoxy. 

 The pin is 0.613 inch in diameter and has an overall 

 penetration of 2.375 inches. The convex tip of the 

 copper contact is machined to align with the dummy 

 piston in the female half. The female contact 

 assembly consists of a 0.375-inch copper rod, a crimp 

 type wire terminal and a gold-plated copper female 

 contact (taken from the experimental model), and is 

 insulated with a machined sleeve of acetal material 

 (DuPont Delrin). The female pin insulator is machined 

 to provide proper contact alignment, a hexagonal 

 portion for reliable assembly with a socket wrench 

 and a cylindrical surface that is turned to the same 

 diameter as the cable insulation to aid in taping and 

 insulating the terminal. The dummy piston is a copy 

 of that used in the experimental model except it is 

 made of Delrin and has a concave surface that aligns 

 and nests with the leading edge of the male pin. 



Several types of pin-wiping seals were tested 

 during these modifications. The pin-wiping seal 

 shown in Figure 10 is a conventional, automotive- 

 type lip-seal. Several similar seals were tested for 

 pin-wiping performance using the test apparatus 

 shown in Figure 11. The apparatus was submerged in 

 seawater and a male pin was used to depress the 

 dummy piston simulating an underwater connector 

 mating. The mating sequence was repeated 50 or 

 more times for each type of seal. Monitoring of the 

 dielectric strength of the insulating oil was the 

 indication of the relative performance of the seals. In 

 general the single 0-ring would reliably allow the 



11 



