The bios ID string in your file is 40-0101-001133-00101111-111192-SIS461. the 1133 relates to Micro Leader Enterprises. According to Wims BIOS it should be 1247 for Abit.
It is possible it is the right bios as Abit did have some boards made by other manufactures, specially the older ones. There is no identifiers on the board itself that show the Abit logo.
Unfortunately I cannot find a BIOS for the AV4, just the AH4 and AN4R2. Maybe someone else will have better luck,

Horun wrote on 2020-03-26, 17:40:

The bios ID string in your file is 40-0101-001133-00101111-111192-SIS461. the 1133 relates to Micro Leader Enterprises. According to Wims BIOS it should be 1247 for Abit.
It is possible it is the right bios as Abit did have some boards made by other manufactures, specially the older ones. There is no identifiers on the board itself that show the Abit logo.
Unfortunately I cannot find a BIOS for the AV4, just the AH4 and AN4R2. Maybe someone else will have better luck,

Yeah, I found the same info. All I have to prove this is a AV4 is the information from TH99 (and they might be wrong).

I found a POST screen picture here: http://pcrebuilding.altervista.org/49/57/2nd+ … +Lightning.html

The BIOS string there is 40-00AG-001247-00101111-060692-SIS3486-F .

Where did you find the AH4 and AN4R2 BIOS images? I might try those, the boards are quite similar.

Thanks for your time 😀

Did you notice that the board in the PC Rebuilding link uses the SiS460 rather than the 461 chipset, and there are differences in some of the IC / socket layouts as well. Your board looks more like this one

https://translate.google.com/translate?hl=en& … 687&prev=search

I do not think the AN4R2 bios will work: "Award Software Inc. 08/30/95-SIS-85C471E/G-2C4I9A12-02"
Different SIS chipset. https://soggi.org/motherboards/abit.htm#S486
And if you use the wayback machine: https://web.archive.org/web/19971221083125/ht … ml/download.htm
shows AH402.EXE for the AH4 and AN4R2 which would be ah4_02.bin as here: ftp://retronn.de/mirrors/ftp.abit.com/pub/dow … d/fae/ah402.exe

Was taking a long break from doing some PCB work in KiCAD and saw this thread. I did find a ROM like that, but it had to be dug up from an old MAME 0.209-0.210 ROM update collection.

Found using hex editor:
40-00AG-001247-00101111-060692-SIS3486
AV4 ISA/VL-BUS SYSTEM BIOS

AmiSapphire wrote on 2020-03-27, 01:28:

Was taking a long break from doing some PCB work in KiCAD and saw this thread. I did find a ROM like that, but it had to be dug […]
Show full quote

Was taking a long break from doing some PCB work in KiCAD and saw this thread. I did find a ROM like that, but it had to be dug up from an old MAME 0.209-0.210 ROM update collection.

Found using hex editor:
40-00AG-001247-00101111-060692-SIS3486
AV4 ISA/VL-BUS SYSTEM BIOS

Good job !

Thank you all! I can't wait to try. I will have to wait until tonight.

I hope this will shed some light on what's wrong with the board.

Would the lack of a BIOS battery prevent it from powering up?
By the way, I see some green residue at where the battery has been. Are all traces intact?

H3nrik V! wrote on 2020-03-27, 06:39:

Would the lack of a BIOS battery prevent it from powering up?
By the way, I see some green residue at where the battery has been. Are all traces intact?

All traces look fine (visually), and I have checked continuity and I cannot find anything wrong. I tried powering on with an external battery pack.

The PSU spins up and keeps spinning, but there is no further activity. No POST code on the POST card, no beeps.

Another thing I can check is the tantalums for shorts, although I would expect that the PSU would shut down with a short. Not all tantalums near the PSU connectors are the same, maybe somebody already replaced some? It wasn't me.

Since I worked from home today I could do some tinkering.

- Changing the BIOS does not matter
- I thoroughly checked all connections where any acid damage was visible. I de-soldered the keyboard connector to check underneath. Everything had continuity. Two vias are nasty though, they do connect from top to bottom. I let them soak in vinegar for a while and cleaned them off again. No change
- Removed all cache chips, no change
- I notice on my POST card the reset line stays high. Not unexpected I think?
- When I press the reset button and hold it, the PSU fan slows down a bit
- Voltages on CPU socket and ISA slots seem fine. 5V is 4.8V, but I think that should be acceptable.

Any thoughts?

Have you checked for shorts? Pull cpu and ram, make sure it a bare board other than cache. Use a good DVM and check resistance from all AT PSU connections to ground using proper polarity (ground being black lead 🤣). If any are less than 10Kohm make a note which ones. Is common for the +5v to be 500-2kohm. Next is the cache, have had boards fail to boot or give any sign of life with bad cache or bad memory. This is trickier so start with the possible shorts...

Horun wrote on 2020-03-28, 01:37:

Have you checked for shorts? Pull cpu and ram, make sure it a bare board other than cache. Use a good DVM and check resistance from all AT PSU connections to ground using proper polarity (ground being black lead 🤣). If any are less than 10Kohm make a note which ones. Is common for the +5v to be 500-2kohm. Next is the cache, have had boards fail to boot or give any sign of life with bad cache or bad memory. This is trickier so start with the possible shorts...

Okay, had a spare moment. DVM set to auto range. Bord was empty (no CPU or RAM), since cache was already out I kept it out. Power good and +12V are suspicious I think? Another board did give results on those also.

P8
1: Power good: Out of limits
2: -5V: 1Kohm
3: +12V: Out of limits
4: -12V: 12Mohm
5: 0
6: 0

P9
1: 0
2: 0
3: -5V: 13Mohm
4: +5V: 1Kohm
5: +5V: 1Kohm
5: +5V: 1Kohm

Those all look ok but maybe the PG and +12v as you said. The Power Good usually sits 3M-10Mohm but is board dependent. You can check it with the PSU connected and turned on (with CPU in) and see if it comes to near +5v after a second or two. The +12v should be close to the -12v in Mega ohms too. Luckily you have no shorts but maybe a possible open +12v circuit.

Horun wrote on 2020-03-29, 15:07:

Those all look ok but maybe the PG and +12v as you said. The Power Good usually sits 3M-10Mohm but is board dependent. You can check it with the PSU connected and turned on (with CPU in) and see if it comes to near +5v after a second or two. The +12v should be close to the -12v in Mega ohms too. Luckily you have no shorts but maybe a possible open +12v circuit.

PG comes to 4.8V (same as 5V). CPU (Intel 486 DX-33) gets warm (hot-ish even).

Not sure what to do next. Will try another PSU for 'fun', to rule out a non-working combination.

Cyrix200+ wrote on 2020-03-27, 14:59:

- I notice on my POST card the reset line stays high. Not unexpected I think?

I think this is the problem to be solved first. Take a working board, remove CPU, BIOS and Keyboard Controller. Reset is off on such board. Think you have some short or broken trace on your board...

Predator99 wrote on 2020-03-30, 10:05:

Cyrix200+ wrote on 2020-03-27, 14:59:

- I notice on my POST card the reset line stays high. Not unexpected I think?

I think this is the problem to be solved first. Take a working board, remove CPU, BIOS and Keyboard Controller. Reset is off on such board. Think you have some short or broken trace on your board...

That is something I'm also keeping in mind, I have some corroded traces, they check out, but I also have some corroded vias. They have continuity from one side of the board to the other, but I think this board also has a middle layer? Or is is just top and bottom?How do I find out?

AmiSapphire wrote on 2020-03-27, 01:28:

Was taking a long break from doing some PCB work in KiCAD and saw this thread. I did find a ROM like that, but it had to be dug up from an old MAME 0.209-0.210 ROM update collection.

... which probably means it's around here somewhere. There's been someone (not me) adding some of the BIOS dumps found on VOGONS/VOGONSDrivers to MAME.

Was not me and is not in the Vogons drivers library but did not search the 486 bios image archive very thoroughly . Checked and it was added to Mame rom update 4.30.2019. Pred added a bunch of SIS 486 bios images so maybe it was one of his...

How far did you follow the corroded traces to test them? (continuity test, I assume) Multimeter or logic probe?

I've got some non-working boards that have corroded but intact traces, but I highly suspect there's broken contact between one or more inter-layer jumper trace bridges. (or whatever those are called: the tiny holes in the board jumping a trace from one side to the other of a PCB)

I haven't attempted repairs of these yet, but I assume it'll involve cleaning out those jumper holes and patching them with solder and/or a copper wire strand + solder.

Any un-broken jumpers will probably get damaged in the cleaning process anyway, so I'll have to fill any one I mess with. (aside from any I can surface-clean well enough to test on both sides of the board)

Those sort of jumpers are also often connected to tiny, almost hair-thin traces, sometimes closely grouped together, so not fun to have to fix. (especially without an advanced solder station, magnifying glass, clamps/clips, etc and a very fine, sharp-pointed soldering iron tip, or wedge/blade tip with very square, sharp corners)