It is a Gigabyte board
https://stason.org/TULARC/pc/motherboards/G/G … 6-GA-486US.html

The video card is Gigabyte as well
https://stason.org/TULARC/pc/graphics-cards/U … VGA-GA-200.html

OP

If you want, you can probably just cut the exploded capacitor off the Tseng card and try it in the slot again. Given it is next to the slot, it is likely just to filter incoming power a little bit, and probably isn't critical to it's operation.

I'm also not a fan of replacing all the capacitors. Yes, more can fail, but more often then not, this sort of capacitor explosion is simply because the card has been left unused for a very long time in maybe sub-optimal conditions. If the machine is periodically used, chances of further tantalum cap failure are lower then in a computer that is packed away for 20 years and never used in the meantime.

Definitely get in there and cut that battery off right away. Those rechargeable cells are board killers, slowly leaking their guts out and corroding all the important bits of the motherboard over time.

The cap that popped on the OP's board was on the video card, not the motherboard. (I just had to deal with a tantalum cap on the -12V line that failed short on a 386/486 board and tripped a jumper in my PSU; will replace the caps at some point, but right now the -12V rail is dead on some of the slots)

OTOH, I don't think the OP's video card would work without that cap replaced.

jesolo wrote on 2017-11-26, 21:17:

Your motherboard was manufactured by Gigabyte. Your particular model is a very early Local Bus (not VESA Local Bus) motherboard.
CPU support is limited to 5V pre-DX2 era 486 CPU's. However, you might be able to run a DX2-50 & DX2-66 on that motherboard by using the same jumper settings as with a DX-25 & DX-33 respectively.

If you get Voltage Adapter Socket you should be able to run every CPU you want, maybe it would be picky with Cyrix 5x86, but AM5x86 does almost ever work.
This SBC is from early 1993 😉 running AM5x86 133Mhz, the 16Kb L1 Cache almost makes up for the lack of L2.

kool kitty89 wrote on 2020-05-21, 08:48:

OTOH, I don't think the OP's video card would work without that cap replaced.

Why not? I will agree that the card will not be at full factory spec until the capacitor is replaced, but there is a good chance that it would work just fine without the capacitor.

Being right next to the I/O pins, but not really grouped with a clock circuit, means it is likely there to help filter motherboard noise off the power rails. This really isn't that critical to seeing if the card/board works.

At worse, it doesn't work, then you can try soldering a cap onto it and trying again.

Thanks all. RE: CPUs, I have an overdrive 100 in now with a voltage regulator. Also just bought an Evergreen 5x86 which has a built-in voltage regulator.

RE: capacitors, I'm under the idea that I can and should replace the one that burned out. I'll just start with that one, using a straight replacement of a similar tantalum cap.

My open question is which part should I use to replace it with? Would either of the two items below work? Not sure what the 10% or 20% means.

CAP TANT 10UF 10% 16V RADIAL
https://www.digikey.com/product-detail/en/kem … -3563-ND/818424

CAP TANT 10UF 20% 16V RADIAL
https://www.digikey.com/product-detail/en/kem … 9906-ND/3726207

tolerance on the capacitance value. So the 10% can be between 9-11uF and the 20% can be between 8-12uF. If there is no tolerance mentioned on the caps that seems to be standard 20% otherwise it will be marked in there somewhere afaik.

Thank you, I ordered the 20% and I'll give it a shot once they arrive. I'll take before/after pics and a video when I actually try to plug it in and turn on the computer. Was a nice little fireball and grey electrical smoke when the capacitor blew. Hoping for less drama this time 😀

Using a higher voltage rating model of similar capacitance can be a safer bet too, though if too much physically larger it may be difficult to mount in the same location.

I was actually puzzling over these or something similar as a substitute for the 16 and 25V rated SMD caps on the -12V rail (decoupling caps for the ISA slots) on the board I had one fail on. (it turned out it was one of the 16V rated ones, but I also ended up breaking one of the J-lead legs on a 25V one during desoldering and trial-and-error troubleshooting that)

In particular I came across these 35V rated radial through-hole tantalum caps on ebay:
https://www.ebay.com/itm/10pcs-NIC-10uF-35v-1 … or/143602319481

I've dealt with mounting radial through-hole caps to surface-mount pads before (replacing electrolytics with electrolytics in that case) and for decoupling purposes I don't think any performance differences between SMD and through-hole caps would be problematic. And 35V seemed a safe over-rating to replace both the 16 and 25V ones.

I was also considering going with 10uF ceramic caps instead, but am not sure if non-polar caps are totally appropriate for this. (from reading, they seem to be fine for PSU decoupling purposes but I'd still think they'd behave somewhat differently than polar caps used for similar filtering of AC ripple and noise on a DC rail)

10uF >16v rated ceramic caps are also bulkier, but that shouldn't be a deal-breaker either given the location.

Other (higher) capacitance values may also be appropriate in this specific role (decoupling for power rails on the ISA slot) and that does seem to be the norm for electrolytic caps being used in the same role (in particular the +12 and -12V rails seem to use those higher value caps) .

And I'm pretty sure going higher on both capacitance and voltage tolerance is usually OK (or a safer, conservative bet) for power supply decoupling purposes. (so long as they can still physically fit in the allotted space, that is)
Using higher capacitance values is also similar to using pairs of similar caps in parallel to improve reliability, just without the effect of averaging out manufacturing tolerances/defects. (that and the external surface area to volume ratio is higher for 2 smaller caps than 1 big one, so there's differences for passive cooling, potential electrolyte evaporation rates, among other things)

There's the 47uF electrolytic caps used on at least those 286 boards I was looking at. Though a Shuttle MVP3 SS7 board I just looked at appears to have 25V rated 10uF electrolytic clustered around the ISA slots though I'm not positive which rails those go to. (plus the ISA slots in late S7 boards are probably assuming less power demand on those slots ... but still, some sound cards with beefy onboard amplifier circuits could be pulling a significant amount of power through there)

Oddly enough, my M-Tech Mustang R534F (SiS 5571 chipset) has through-hole tantalum caps lining the ISA slots, in spite of being a late-90s Socket 7 board.

OTOH, the sort of capacitors used onboard video or sound cards can be a lot more specific in requirement than just power decoupling. (and you've got the non-polar caps on the coupling and filtering side of things in amplifier circuits for audio output and I think some video output circuits as well)

Though if they're also being used on the decoupling side of things (right on the power supply input lines from the ISA slot) I think similar assumptions can be made regarding suitable substitutions.

Wait, I may be mistaken in my references to decoupling. I think the ISA-slot associated capacitors I'm referring to (and any other caps of fairly large values, by the standards of these sorts of circuits: ie in the high nF into the uF range) are filtering capacitors, not decoupling.

Decoupling caps are usually smaller and are used on I/O and other signaling lines (sometimes oscillator signal lines) and not power supply lines. (normally located right next to the ICs receiving the signals)

I need to do more research on filtering capacitors for DC supply lines and appropriate replacements.

edit:

No, no, they must be decoupling caps as there's live 12V (or -12V) power going across the two contact pads for the capacitors.

If they were filtering caps, they should be open-circuit unless a card in one of the ISA slots is present, closing the circuit.

And if a filtering cap failed short it wouldn't have been causing the problems I had (short-circuit on the -12V pin of the AT power socket/connector) where a shorted decoupling capacitor would be shorting between live and ground.

So these are larger(er) value decoupling capacitors doing part of the same job that filtering caps do (smoothing out AC signal noise in the DC supply line).

That also means the -12V lines on all ISA slots are still live, but there's now fewer decoupling caps running in parallel to suppress ripple/noise.

Just a quick follow-up. I soldered on a new capacitor, powered it on, and POOF the capacitor next to it blew out. Same one as the earlier poster. This resulted in a small golden-hot projectile being forcefully ejected from the card and promptly landing on the carpet, burning a small capacitor-sized hole. My wife was not super pleased about that.

BUT I did solder on a second capacitor in its place. This worked! The board is now operational and back in my case. Thanks to all for the input and suggestions, would not have been able to figure this out without your help. I appreciate the community even if my wife isn't quite as pleased with this project 😀

There's been bumps to this thread since I've started it and also just got a PM about the availability of this motherboard. Sad to say, I have discarded it a year or so ago when I was moving.