First post, by Skyscraper
I posted about this board in the "Bought these (retro) hardware today" thread a week ago.
I still have not got the board working and it could be a memory incompatibility issues as has been suggested but lets presume it's not as I only have 44256 (and 4464) memory chips.
This is how the board looked when I got it (the sellers picture). The only thing looking a bit odd other than the skewed expansion slot is the choice of placement of the memory chips. Without knowing the exakt bank layout I think someone else has been trying to get this board to run before me and I think these 4 chips are probably spread out over 3 memory banks without knowing which sockets belong to which bank or which bank are which.
To get the memory issue out of the way I have filled all 8x DIP-20 sockets with known to be good 44256 chips from a 286. This is probably 3 memory banks (2x +2x +4x) if this boards bank layout is like other similar boards but could perhaps be two banks (4x + 4x) if we assume the board came with 512KB (2x 44256 + 2x 44256) memory in the correct sockets. Filling all 8 DIP-20 sockets should correct this issue what ever the bank configuration one would think. The rest of the 10 memory sockets are 6x DIP-16 (I assume for parity) and 4x DIP-19 (I assume another memory bank, perhaps two should the 8x DIP-20 sockets only be two banks).
This is the motherboard I have been using as a template, it's not the same but very similar.
https://th99.bl4ckb0x.de/m/A-B/32279.htm
This is the memory chips I'm using and the board on the test bench a few days ago.
The diagnostic card is perhaps not 100% useful as XT boards send codes on port 60? and not on port 80 but the card still shows "FF" in my working XT clones.
Thinking the memory is not the problem I turned to the BIOS chip, once removed we have at least found an indication someone else has been tinkering with the board. There was no short and this pin is NC on many flash chips but still worth fixing.
Fixed, this did sadly not revive the board. I then tested the BIOS chip in my working XT clone and the BIOS works fine. I will dump the BIOS for member Anonymous Coward tomorrow.
I then started looking for shorts with my multimeter, with the board not connected to anything I did not find any shorts but I did find something worth investigating. When the motherboard is connected to my AT PSU with the power off measuring on an expansion card (does not matter which) shows only 48 Ohm resisitance from +5V(B3) to GND (B1 or bracket) and the resistance is stable. When measuring over caps on the motherboard with the AT PSU connected with the power off there is also only 48 ohm resistance, the same on every cap. When removing the AT PSU the resistance changes to no connection at once.
Not knowing exactly what I should think I moved over to the woking XT clone and measured the resistance from pin B1 (or bracket) to B3 on an expansion card with the PSU connected but turned off (not the same PSU) and the resistance is only 19 ohm but goes to no connection within 2 seconds. Measuring on caps on the working XT clone board with the PSU connected gives the same 19 ohm for a second then no connecion.
The behaivour is not the same is pretty much all this tells me and as the PSUs are not the same perhaps this is all normal? The PSU I'm using for testing the non working board works fine with other motherboards. I guess the next step is to try another PSU, I have already done that but not after fixing the BIOS chip leg and I have not measured the resistance from +5V to VCC on connected expanstion cards or the resistance over caps with another AT PSU connected to the board (turned off) to see if there is any change or if there is still a stable 48 ohm resistance.
This is where I am at now. Any inputs are appreciated.
New PC: i9 12900K @5GHz all cores @1.2v. MSI PRO Z690-A. 32GB DDR4 3600 CL14. 3070Ti.
Old PC: Dual Xeon X5690@4.6GHz, EVGA SR-2, 48GB DDR3R@2000MHz, Intel X25-M. GTX 980ti.
Older PC: K6-3+ 400@600MHz, PC-Chips M577, 256MB SDRAM, AWE64, Voodoo Banshee.