I used to design marine radios in the 1970's and 1980's. I have
plenty of experience doing it wrong.
Yes. As John Larkin mentioned, it's dependent on what manner of
signal you shove through the connections. The assumption with gold is
that you have a "dry" contact, which means no DC. That includes AC
signals sitting on top of a DC voltage. For example, one of my less
brilliant ideas was to combine the +15v power and signals on the
AN/SRD-22 VHF direction finder antenna. There was no immersion test
for antennas, but it barely passed the salt fog and salt spray tests
due to the DC. In order for it to work, we had to use 50 millionths
gold plating. Any less would just disappear. Inside the radio were
the usual (now ancient) 0.156" gold PCB edge connectors. The PCB
connectors did well, but they too had to be 50 millionths gold plated.
It was correctly decided to concentrate on sealing the radio and
antenna, rather than trying to make it corrosion proof. There just
wasn't anything better than gold.
You didn't specify which ASTM or MIL test you were running. In my
never humble opinion, the real problems are how long an immersion
test, and whether it is cyclic (dip-dry-dip-dry-dip-dry-etc). Short
term immersion is probably not a big challenge. Long term and cyclic,
where little things like changes in internal air pressure can pump
salt water into the unit under test, are far more difficult. Salt
crystals can also creep into cracks in the PCB, under paint, inside
devices, and then expand when heated. You may not have a problem
during immersion, but when the device is powered on and the crystals
expand, you may see some very strange problems.
Good luck.
