Ok, What exactly do you mean by "does not start"?

Your photo shows the board obviously powered on (POST LEDs are lit), but it has no visible video
card or RAM installed... I assume you know these are normally requires to POST?

I do see a header with a couple of wires at lower right ... is this for a speaker?
Are you trying to get "error beeps" from a BIOS where RAM fails? I've never found
this terrible reliable...

Assuming you have tried it with RAM and Video.. and it still doesn't POST.. Since you have a scope,
can you confirm that the BIOS ROM is getting selected (ie: at least a few instructions are fetched
from it) right after RESET? If no, can you confirm that the RESET and clock lines to the CPU are
working as expected?

DaveDDS wrote on 2025-10-09, 03:09:

Ok, What exactly do you mean by "does not start"? Your photo shows the board obviously powered on (POST LEDs are lit), but it ha […]
Show full quote

Ok, What exactly do you mean by "does not start"?
Your photo shows the board obviously powered on (POST LEDs are lit), but it has no visible video
card or RAM installed... I assume you know these are normally requires to POST?
I do see a header with a couple of wires at lower right ... is this for a speaker?
Are you trying to get "error beeps" from a BIOS where RAM fails? I've never found
this terrible reliable...
Assuming you have tried it with RAM and Video.. and it still doesn't POST.. Since you have a scope,
can you confirm that the BIOS ROM is getting selected (ie: at least a few instructions are fetched
from it) right after RESET? If no, can you confirm that the RESET and clock lines to the CPU are
working as expected?

Hi.
Here is the video of 'power on':
https://www.youtube.com/watch?v=kvh5GUMjgco

I gave it a through 'look' with the scope again and (my) conclusion is that address lines work OK but data lines are screwed.

BIOS:
CE - yellow,
OE - blue

Waveforms on address lines look really good (unfortunately, I did not save the picture).

And D0,D1,D3 on ISA... 🙁
Signals on BIOS 'D' lines ain't look better (again, I did not save the picture 🙁 ).

Looks like POST CARD gets proper address signals and crap on data lines is constantly interpreted as '11111111' -> 'FF' ?

Checked BIOS 'D' lines for resistance ~600ohm seems normal.
Then I've checked voltage on those lines without EPROM inserted and got 3V (???) on them.
Those lines connect directly to KBC so i desoldered AMIKEY-2.

Now signals on BIOS datalines look IMO pretty good, on ISA still ain't great....
BIOS D0 - blue
ISA D0 - yellow

There is 74ALS245 transceiver between them that may be damaged so I've desoldered it and then looked at the clock :]
TBC.

PS: Some 'lines' on dipswitches JPCU1 & JCPU2 didn't make contact - would you believe that.... ?

Locutus wrote on 2025-10-10, 12:48:

Here is the video of 'power on':

Same as your photo - no RAM or Video card... have you tried with?

You do have a little speaker attached, and BIOS should at least beep if no RAM or video (it can't go much farther)
But, I've run into systems/setups where this wasn't reliable.

I've also ran into a system once where RESET depended in the hard drive reset and didn't always POST without a hard drive.
Easily "fixed" with a manual "reset button".

This is why when testing a new MB, I always have RAM and a video card installed, and always have a RESET button connected.

DaveDDS wrote on 2025-10-10, 16:37:

Same as your photo - no RAM or Video card... have you tried with? You do have a little speaker attached, and BIOS should at leas […]
Show full quote

Locutus wrote on 2025-10-10, 12:48:

Here is the video of 'power on':

Same as your photo - no RAM or Video card... have you tried with?
You do have a little speaker attached, and BIOS should at least beep if no RAM or video (it can't go much farther)
But, I've run into systems/setups where this wasn't reliable.
I've also ran into a system once where RESET depended in the hard drive reset and didn't always POST without a hard drive.
Easily "fixed" with a manual "reset button".
This is why when testing a new MB, I always have RAM and a video card installed, and always have a RESET button connected.

Hi.
The first boot attempt was with the memory, graphics card, and speaker installed.

If the motherboard is operational, the lack of RAM/VGA doesn't (or shouldn't...) cause such behavior - motherboard reports then appropriate error code to port 80h.
In my case, address bus seems to receive proper signals (at least for the port '80h' used by diagnostic card) and on the data bus bits 0-7 have constantly logical value of one.
That is why the board immediately reports the error code 'FF' (that's my theory that might be incorrect).

Anyway, I've probably found a major hardware problem.
On terminals of two capacitors there is very low resistance (I've ruled out cap failure by desoldering them).
Perhaps we're dealing with an interlayer short in the PCB 🙁

What I'd probably do next is to confirm that the system is trying to read the BIOS - you should see a few activations of the chip select.
(Even if it's reading FF it should do so at least once, FF is an invalid 8086 opcode (don't know about more recent devices) and that should
trigger a sofrware int) - so I would expect several accesses to the BIOS chip.

I see that it is a flash - and I have had flashs go bad - can you read the flash itself and see if it is FFs.

Another option - pull the flash and use some pull-down resistors to "convince" the data bus to assume 00 when not driven.
- with power-on, does it still read FF - if yes, something is indeed driving the data bus high.

Hi.

DaveDDS wrote on 2025-10-15, 14:02:

What I'd probably do next is to confirm that the system is trying to read the BIOS - you should see a few activations of the chip select.
(Even if it's reading FF it should do so at least once, FF is an invalid 8086 opcode (don't know about more recent devices) and that should
trigger a sofrware int) - so I would expect several accesses to the BIOS chip.

The oscillogram of the waveforms on pins OE and CE, which I've posted here:
download/file.php?id=228522&mode=view
seems to confirm series of reads ( CE & OE are active low inputs).
I have a PCB of logic analyzer that I've never had time to solder.
Maybe it's the right moment to finish it... it would make analysis a whole lot easier...

DaveDDS wrote on 2025-10-15, 14:02:

I see that it is a flash - and I have had flashs go bad - can you read the flash itself and see if it is FFs.

The flash memory was programmed by me and for sure contains correct BIOS for this motherboard.
I've checked several available versions downloaded from retroweb with the same result.

DaveDDS wrote on 2025-10-15, 14:02:

Another option - pull the flash and use some pull-down resistors to "convince" the data bus to assume 00 when not driven.
- with power-on, does it still read FF - if yes, something is indeed driving the data bus high.

That is a very good idea!
But first, I'll try to figure out this mysterious short circuit on pins of capacitors.

Problem pinpointed and solved !
A virtually impossible defect - two closely spaced vias with a solder meniscus were connected, appearing to be a properly soldered trace - likely the result of an impact.
Viewed perpendicularly at the lying board, it was indistinguishable from the other traces.
I've removed this shortcut and voilà- I get a proper 'C1' postcode code.

Awesome! - congrats!

Nice work, hard to believe you even found that