Me vs NiCad battery damage on PC boards

[Tim Shoppa]

A lot of nice radio/test/computer equipment from the late 70's through 
the 80's and even the early 90's had the "feature" of battery-backed-up 
configuration settings. Often this was done thorugh NiCads soldered to 
the PC board (often in those little plastic modules). Over the years, 
the NiCads have usually leaked out their gunk (if not liquid, then 
fumes?) which has gone on to attack the PC boards and 
components/sockets around it. All the originally shiny metal now has a 
layer of white or white/green deposits over it. 

My primitive chemical understanding is that something-hydroxide 
(calcium hydroxide, sodium hydroxide?) from the battery gets out (if 
not as a liquid, then as a vapor) which is a base. This corrodes the 
metal surfaces around the battery. 

In my most recent escapade, I took such a board and put it in some 
dilute acetic acid (think "vinegar" except I was using some photo 
chemicals and I probably had the concentration stronger than the 
grocery store vinegar). The deposits turn even greener (in some cases 
bluer but maybe that was the indicator in the stop bath). Using a 
plastic brush I scrub away these deposits, but this does not result in 
a shiny clean PC board. 

Again, my primitive chemical understanding: the acid combines with the 
base whatevery-hydroxide and what we're left with salts. 

Rinsing in water and repeating does help get off more of the gunk but 
it's never all gone. In any event, I've at least slowed (maybe stopped) 
damage to the PC board, and after drying off I find it still works - in 
some cases works better because we don't have all the gunk anymore. 
When I see the gunk I see it mostly as an insulator that will stop 
conductivity, but of course it also bridges together traces and 
probably gives very real leakage in some cases too. Incidentally in 
many cases the board doesn't work right until it is really really dry, 
so I think that in some cases just the leakage due to the rinse water 
is enough to inhibit normal operation. 

But now, say I want to remove a somewhat corroded IC socket off the 
board because I suspect it's causing problems. I put my soldering iron 
on the pad and try to heat up and melt the solder. I heat, and heat, 
and heat, but nothing's melting! Check that the soldering iron is 
working and that the tip isn't covered with crud... well, maybe there 
was some crud but after wiping it off the soldering iron is working 
fine. 

So I go in with a wire brush and an X-acto knife and try getting past 
whatever chemical is still covering the pads and traces, and find it's 
really tough stuff. The wire brush works but it also erodes the traces. 
The X-acto knife works somewhat on pads with solder on 'em. After 
repeated attacks, though, the solder still doesn't really want to melt. 
I try putting on some new solder thinking that the rosin and new new 
metal will help, but all it does is ball up and roll off. 

So what I'm left with is a PC board that is somewhat/mostly working but 
in cosmetically poor shape, and feeling that I should do more. What's 
semi-ironic about all this is that the board usually doesn't have any 
more than $10 worth of parts on it but an instrument that cost $10K 
twenty years ago (and is maybe worth $1K today on the used market) is 
dependent on this still crud-covered PC board that I can't even fix the 
sockets on. 

Buying a replacement PC board is usually out of the question because 
they haven't made this doohickey in 20 years, and all the boards out 
there all have NiCad leakage damage on them. 

One solution would be to lay out a clone PC board that never had the 
NiCad damage done to it and stuff it with new parts. Realistically in 
some cases this would take less time to do than I've already spent 
cleaning/rinsing/drying/attempting to solder on the existing board. 

But is there some magic chemical or mechanical means that will render 
the existing cruddy PC board solderable so I can replace these flaky 
sockets? 

Tim.