80386 board with acid damage

[naujoks]

I'm trying to fix a 80386 board that has damaged from a leaked NiCd battery.
I tried to clean it as much as possible with vinegar and isopropyl alcohol.
One trace had been completely eaten away and I soldered in a wire to repair it.
I put in a new battery.
I replaced the socket for the keyboard controller IC. I cleaned all legs of the other socketed ICs (they weren't corroded).
Desoldering anything on the corroded bit is a major PIA. Solder tin won't stick to the legs, flux doesn't help to make the old solder tin melt.
By unsoldering things I felt I did more damage than good, the desoldering gun lifted a couple of soldering pads, so i had to fix some traces.
I checked the voltages on the ISA sockets, and they're all good. I checked voltages on the ICs across the board and they're good.
The board shows no sign of life, no beep, no picture. The CPU and a couple of other ICs get slightly warm, but that's it.
Any suggestions what else I could try?

[TheMG]

Corrosion damage from a leaking battery (btw, it's a base, not an acid, NiCd uses potassium hydroxide electrolyte) can be very sneaky.

Even if a trace *looks* intact, test it anyways. Test EVERYTHING within the affected area for continuity. Don't overlook via and through-hole plating, the stuff loves to eat those away and create open circuits.

Very likely something in there is still open-circuit.

When repairing corrosion damaged traces, go all the way back to a completely unaffected part of the trace to make your repair, don't even attempt to solder to the corroded areas, skip over it altogether.

[shakalnokturn]

Cut "power good" line from AT header?
Just a guess among many other options...

[Rasz]

By unsoldering things I felt I did more damage than good, the desoldering gun lifted a couple of soldering pads, so i had to fix some traces.

most likely you didnt. If it fell apart it was marginal to begin with, would work for a few hours until breaking randomly.

I checked the voltages on the ISA sockets, and they're all good. I checked voltages on the ICs across the board and they're good.
The board shows no sign of life, no beep, no picture. The CPU and a couple of other ICs get slightly warm, but that's it.
Any suggestions what else I could try?

this topic goes over a lot of troubleshooting steps
https://www.eevblog.com/forum/repair/how-to-troubleshoot-your-defective-386486-motherboard-with-an-oscilloscope/

[naujoks]

I'm still working on this and after a lot more cleaning and replacing of parts I'm still not getting any life out of this board.
I was wondering, as I'm using an ATX PSU with an AT adapter, I don't have a -5V line. Am I correct in thinking that the -5V isn't necessary to run this board?

[Inhibit]

Since I was curious I took a looksie at the Power Supply (computer) Wikipedia page (easiest place to find a mile high overview of the AT PSUs these days).

And the answer is probably not:  "A −5 V rail was provided for peripherals on the ISA bus (such as soundcards), but was not used by any motherboard other than the original IBM PC motherboard." Unless you're using an ISA bus peripheral; I don't know the effect that'd have on power-up.

It's got a basic but useful breakdown of the common computer PSUs: https://en.wikipedia.org/wiki/Power_supply_unit_(computer)#Original_IBM_PC,_XT_and_AT_standard

[Rasz]

I'm still working on this and after a lot more cleaning and replacing of parts I'm still not getting any life out of this board.
I was wondering, as I'm using an ATX PSU with an AT adapter, I don't have a -5V line. Am I correct in thinking that the -5V isn't necessary to run this board?

did you read the topic I listed?
-5 was used by some ancient ram chips, -12V by some sound cards

[naujoks]

I managed to destroy one of the diodes seen on the first picture in position D1-D4.
Is there a way to tell what kind they are and what I can use to replace it? There's nothing written on them.

[Rasz]

most likely something as common as 1N4148