Nissan used a cheap elastomeric seal that goes around the feedthrough interconnet, in addition to another one that mechanically supports it. It often times disintegrates due to vibration, and long term exposure to fuel additive vapors which the S12 was never intended to run on.
It is possible that the center conductor might short to the cover plate, due to this elastomeric fitting, despite the top caps being covered with epoxy. If its inside the tank, no problem, sparks are no big deal there. It its external, and you have another one that is leaking fuel vapor, big bang is possible.
They also did not properly terminate the connections.... over time, the wires will corrode, and they may well break off inside the cover caps which were molded in place. Then again, the S12 is probably well past its design service life, so it probably was an acceptable compromise. Chances are, 20 years ago, vapor migration through molded on plastic was considered a minor problem, if it was even recognized outside of the military arena.
I machined new feedthroughs, made out of Delrin, and used nitrile orings for sealing. I also changed the terminations to prevent corrosion. Considering a new plate is only $150, its wasn\'t the most cost effective, but having had to replace these a few times, I decided to fix the problem once and for all.
I pulled my fuel pump notes, and made a pdf file. I intentionally left the design documents for the spacers, orings, and terminations off, as I don\'t want someone to make a mistake and have a fire. A minor error in dimensions, or material selection, could have serious consequences, and I want no part in that.
The document has also not been checked for accuracy, especially on the third page, as my intern put it back together for me, so I didn\'t get a chance to see if it was correct or not. I could just see him not following the directions, and it works.... I just don\'t know. However, you may find the operation specs and wiring to be helpful.
If you do still want to use epoxy, the marine stuff may or may not hold up to fuel vapors. The original, not the fast curing JB weld does. It also withstands gasoline in liquid form in one isolated long term test. It is not acceptable for pressurized sections. We could never get it to withstand more than 5 psi before leaking, unlike high dollar industrial epoxies. The other problem with an epoxy repaird, is its hard to properly prepare the surface, especially when you have a unknown plastic cap, and a likely corroded top cover. I don\'t recommend it.