Check the keyboard controller (the large intel chip on the left of the 387 on your picture), try to reseat it. Maybe it's just a bad connection. Jumpers near this area are probably controlling things related to something close to them, maybe the KBC is part of this. It only has 4 buttons, so 16 possible values. It shouldn't be too much of a hassle to try them all.

Also I'm not sure if it's the lighting or if it's really something : I see white stuff under the 387 and the keyboard controller. Is that just an optical illusion (like I said, perhaps it's just lighting), or was something spilled on there ?

On your backplane I see two connectors with "KB-IN" and "KB-out" marked nearby. Maybe your backplane expects something to connect there related to the keyboard ? Maybe your SBC needs to connect something to these in some way ? (Maybe through the header next to the ISA slot ?)

Yeah, there's no standard raw KB data over ISA protocol AFAIK so not sure how KB port on backplane would be meant to communicate with KB controller on board if not for a jumper cable to those headers. Although it's not real obvious where on the board you could obtain that.

Edit: with seeing nothing remotely likely elsewhere, I am wondering if that jack socket J6 is anything to do with the keyboard, maybe clock and data with the original machine providing KB power and ground elsewhere???

Yes, the keyboard connector on your backplane ("KB-IN") needs to be connected to the keyboard header on the SBC. I dont see it on your photo of the SBC. Be sure to check the pinout, ground and +5V should be easy to identify...

I think you'll have an oscilloscope to check and find the keyboard data and clock signals.
Maybe some pins on the 4-pin jumper post between the KBC and the (possibly leaking) battery?

retardware wrote on 2021-09-07, 21:38:

I think you'll have an oscilloscope to check and find the keyboard data and clock signals.
Maybe some pins on the 4-pin jumper post between the KBC and the (possibly leaking) battery?

Or that thing that I thought was test points to the left of the edge connector. Maybe has keyboard and ps/2 mouse? Or has activity/power light connections too.

retardware wrote on 2021-09-07, 21:38:

I think you'll have an oscilloscope to check and find the keyboard data and clock signals.
Maybe some pins on the 4-pin jumper post between the KBC and the (possibly leaking) battery?

that 4 pin header is almost certainly for an external battery, my money is on the 8-pin guy next to the edge connector

That's a really cool looking board, all those right angle traces and multi layers 😀
The ISA bus doesn't have keyboard data going through it and that SBC predates standardised conventions for connecting keyboards in industrial computers, I think it's from about 1988? It looks like this SBC has forever been a mystery though 😁 Take a look at this: https://groups.google.com/g/comp.sys.ibm.pc.h … /m/Tm6jOe0e58oJ

Since it was designed to be pushed into a slot and have no connectors directly attached to it, my guess is that it will be some pins on that white connector just after the end of the ISA slot, that could have pushed up against some contacts on another board that handled the keyboard connection.
You need to find 5v (use the 5v pad on the isa slot for reference), ground. If you look up the keyboard controller's datasheet, or any DIP40 KBC datasheet, it should list where the KBCLK and KBDATA pins go.
Then just test continuity from one of those pins on the KBC then probe around that white connector by the ISA slot, then the two 40 pin connectors, it could be hidden somewhere in there.

I wonder what that headphone jack does - maybe that's the PC speaker?

Hello everyone! For some reason, I didn't get emailed/notified that there were replies here! I appreciate people trying to help 😁

I've actually made progress here!! This board is by a company called Texas Microsystems. They've made other SBC boards, like a 286 one. This 286 board has an 8-pin connector that has signals for the KB data/clock!

Looking at my board... the white, 8-pin connector by the ISA connector seems to be a match! Using my multimeter, I was able to determine that the +5V and GND pins on my board line up with the J5 connector on that 286 one.
So, if that's the case, maybe the other pins are the same, too...

I was able to use the 1st two pins to reset the computer, so those pins match. And the last 2 pins are indeed the PC speaker! So, this means I need to connect the KB data and clock here.

Now, it all works! The KB works and I can get into the BIOS screen, which says "Texas Microsystems"! 😀