ggalvan wrote on 2024-05-21, 10:44:

Hello, […]
Show full quote

Hello,

I'm trying to repair this 286 motherboard. (https://theretroweb.com/motherboards/s/octek-fox-m-286) with AMI BIOS.

I'd replaced 3 tantalum capacitors, one expode and other one made a short circuit and power supply did'nt started.

With this fix i could start the board wiouth memory. 3 beeps error.

I'd installed 4 simms 30c 1Mb 9 chips and 3 beeps disapared but obtein the code "02-01" in the test card.
I'd installed VGA and Multi I/O card 16bits ISA cards too.

Reparing two circuit damaged by battery acid i obtain the same code in the test card.

I think in replace all tantalum capacitors, any other idea?

Thanks.

Post pics of the board. Add some of the area damaged by the battery.
01-02 means it is at least partially alive.

Stay positive but be aware it could take some time to repair it.

The three keyboard leds blink when power on but doesn't respond later if i push num lock or shift lock...

The 2 circuits repaired go to MC146818P
mc146818p - PIN 3 - OSC1
mc146818p - PIN 13 - /CE

The other circuit seems to work but I can't find where it goes.

Try deoxit or similar contact cleaner on the bios and keyboard controller (the one with keyboard bios on it)

Remove the 287 chip, not needed and rules one more thing out.

Go for a second round of continuity tests, 01 02 points to an RTC error. Probably some trace is broken.
https://pdf1.alldatasheet.com/datasheet-pdf/v … A/MC146818.html

Keyboard controller:
https://pdf1.alldatasheet.com/datasheet-pdf/v … NTEL/P8042.html

If you have an oscilloscope check all crystals.

Trick, if you need close detail pic use a magnifying glass in front of your phone camera. In the end they are all lenses.

Could you take a picture of the board from the top without cards installed?
Also the bottom.

Did you thoroughly check the jumpers to be correct according to the manual?

I'm removing BIOS and cleaning contacts ...

Hmm where is the odd BIOS chip ? I have same board and it has two (even and odd) same as those on TRW. Also my board J2 is set for 1-2......

You can see the two bios chips in the firsts images, in the sencond post i was cleaning pins and sockets ....

In my board JP2 is config in 1-2, i test with 2-3 but same result.

ggalvan wrote on 2024-05-21, 10:44:

I'd installed 4 simms 30c 1Mb 9 chips and 3 beeps disapared but obtein the code "02-01" in the test card.

BIOS checksum error. If treating the sockets with deoxit didn't help you might have to try some other image to make sure it's not a problem with one of your chips.
Do you have EPROM programer? Can you copy your chips? I can verify the checksum, it won't tell you 100% that the chips are good (sometimes the chips get slow with age, work in programmer but not in the system) but if there is a glitch then at least you can rule out mobo damage.

I've read the eproms with my arduino eeprom programmer, they are very similar to the "retroweb" upload images ....

ggalvan wrote on 2024-05-22, 17:18:

I've read the eproms with my arduino eeprom programmer, they are very similar to the "retroweb" upload images ....

The good news is the image checksum appears to be correct. The bad is I've disassembled the code and 02 is indeed BIOS checksum test. If it never gets to 03 then the test fails, or the CPU somehow glitches. This is all done before any part of RAM is used, and should result in beeping.

If there is no beeping now but removing RAM sticks make a difference then perhaps the problem is not with BIOS as such but the system thinking it's a warm reset instead of cold one and not setting up the HW correctly. All kinds of things can happen then. The chip responsible for providing the status bit of cold/warm reset is the keyboard controller.

One more option is somehow one of the address or data bits going to the BIOS ROMs is bad. For data it wouldn't be completly but perhaps weak, like bad bus driver chip. Address bit could be completly cut - but not one of the lowest bits. Too much of the BIOS still works for that, I would guess one of the higher bits like A15 might be stuck high. Nope, can't be that, the BIOS checksum only covers the upper 32k of the segment so that would still work...

Thanks for the info. When i reset several times with reset button same times obtain a "0301" code ...

The keyboard controller i think is the 40 dip chip labeled with "Keyboard bios setup", can i replace it with other keyboard controller?

ggalvan wrote on 2024-05-23, 06:07:

Thanks for the info. When i reset several times with reset button same times obtain a "0301" code ...

The keyboard controller i think is the 40 dip chip labeled with "Keyboard bios setup", can i replace it with other keyboard controller?

Yes.

I replace with a working keyboard controller of other 386 motherboard and the same result.
Without keyboard controllar i get "03-10" in test card.

ggalvan wrote on 2024-05-23, 10:00:

I replace with a working keyboard controller of other 386 motherboard and the same result.
Without keyboard controllar i get "03-10" in test card.

Have tried a different RTC?

I don't have other RTC, i'll buy a new one

ggalvan wrote on 2024-05-23, 10:09:

I don't have other RTC, i'll buy a new one

I posted the RTC's datasheet, you could check if everything is connected correctly.

ggalvan wrote on 2024-05-23, 10:00:

I replace with a working keyboard controller of other 386 motherboard and the same result.
Without keyboard controllar i get "03-10" in test card.

03 code is for testing KBC. The controller is soft-reset and responses are checked, after the test passes the next code should be 14.

Since replacing KBC didn't change anything I would assume it's a some sort of mobo issue. Weak/dead bus transciver chip or some broken connection maybe. On this mobo KBC and RTC are pretty far from the battery (and curiously also far from the 32k oscillator circuit) but also in the corner of the PCB. Corners tend to flex more and can get damaged, so look for any signs of the mobo being dropped on that corner, or any scratches on the traces (usually on the bottom, the top is "protected" by the chips).

Deunan wrote on 2024-05-23, 11:20:

ggalvan wrote on 2024-05-23, 10:00:

I replace with a working keyboard controller of other 386 motherboard and the same result.
Without keyboard controllar i get "03-10" in test card.

03 code is for testing KBC. The controller is soft-reset and responses are checked, after the test passes the next code should be 14.

Since replacing KBC didn't change anything I would assume it's a some sort of mobo issue. Weak/dead bus transciver chip or some broken connection maybe. On this mobo KBC and RTC are pretty far from the battery (and curiously also far from the 32k oscillator circuit) but also in the corner of the PCB. Corners tend to flex more and can get damaged, so look for any signs of the mobo being dropped on that corner, or any scratches on the traces (usually on the bottom, the top is "protected" by the chips).

Do you have a complete sequence with codes?
I had a similar case I'm still trying to solve and had codes up to 08-09, randomly stopping before those at different points.

Anyway, if it issues codes could it be a 74xxx dead in between?