keyboard lights is weird one. Keyboards blink once on their own, then again during POST when keyboard controller is initialized. You arent getting that far since you get no post codes, so the keyboard itself is hanging while initializing itself? measure keyboard port 5V supply.

That is a weird one with the KB lights staying on. Maybe the Tant caps are failing but have not blown yet ? or worse the '206 has failed.

Horun wrote on 2023-03-19, 01:40:

That is a weird one with the KB lights staying on. Maybe the Tant caps are failing but have not blown yet ? or worse the '206 has failed.

Incidentally, I have an ECS 386 motherboard that had battery damage and that I bought for spares. I thought I'd try to repair it. It has exactly the same chipset and, after replacing the BIOS ROM, it is exhibiting the same symptoms. No POST, just keyboard lights staying on. POST analyser card shows all rails are good and there is a clock signal. Reset also isn't stuck.

I'm also suspecting one of the chipsets. What makes you believe it's the '206 chip and not the 495XLC chip?

same keyboard? 😀

rasz_pl wrote on 2023-03-19, 08:22:

same keyboard? 😀

Yes, but I did test the keyboard on another system and no issues. Also, when I unplug the keyboard, system still doesn't want to POST.

jesolo wrote on 2023-03-18, 21:56:

I've attached a picture of the motherboard as mentioned in the subject header. a-trend-495slc-revc.jpg I bought this motherboard […]
Show full quote

I've attached a picture of the motherboard as mentioned in the subject header.
a-trend-495slc-revc.jpg
I bought this motherboard a couple of years ago as a "backup" motherboard and, when I received it, I immediately removed the barrel battery and there was no battery leakage.
Upon testing the motherboard, everything worked fine and I placed it back in storage.

I recently retrieved this motherboard from storage and, when I tried it start it up again, it was just completely dead.
I've tried the following to find the fault:

1. I've used a POST analyzer card and all the rails are fine (no shorts). I also checked with a multimeter. However, there is no activity on the card and all it displays are the 4 solid lines on the LCD display (no error codes). The reset light on the POST analyzer card does not stay on but, does come on momentarily when I reset the motherboard (which is normal behaviour).
2. This led me to believe that it could the BIOS ROM. So, I burned a new EEPROM and tried it in the motherboard. No change.
3. What I did notice is that, when I plug in a keyboard, the keyboard lights just stays on (it doesn't switch off).
4. I then swopped out the AMIKey keyboard BIOS with a keyboard BIOS from another 486 board (although, not exactly the same one). Still no change. Must I use exactly the same keyboard BIOS?

Based on the above symptoms, where else could I be looking to try and find the fault?

Why do not post actual photo with CPU, memory installed?

jesolo wrote on 2023-03-18, 21:56:

I recently retrieved this motherboard from storage and, when I tried it start it up again, it was just completely dead.

I see some bent jumper or conenctor pins on the right hand side, make sure these are not shorting (and preferably make them straight). Re-seat CPU and all jumpers, and also the socketed chips in the bottom right corner (I assume these are clock generator and a buffer). Check for clock, if you don't have a scope then a multimeter can be used to some extent - you should see roughly 1/2 of the 5V DC on the clock like since pretty much all clocks will be about 50% duty cycle. You can also test ISA data and address lines like that but that will be inconclusive in most cases, since you won't be able to tell if the line is constantly L/H or if there are pulses but very short.

Deunan wrote on 2023-03-19, 12:09:

jesolo wrote on 2023-03-18, 21:56:

I recently retrieved this motherboard from storage and, when I tried it start it up again, it was just completely dead.

I see some bent jumper or conenctor pins on the right hand side, make sure these are not shorting (and preferably make them straight). Re-seat CPU and all jumpers, and also the socketed chips in the bottom right corner (I assume these are clock generator and a buffer). Check for clock, if you don't have a scope then a multimeter can be used to some extent - you should see roughly 1/2 of the 5V DC on the clock like since pretty much all clocks will be about 50% duty cycle. You can also test ISA data and address lines like that but that will be inconclusive in most cases, since you won't be able to tell if the line is constantly L/H or if there are pulses but very short.

Yes, I think my next step would be to measure the clock and address lines with an oscilloscope. Unfortunately, I don't have one so, will have to take it to a friend that has the right equipment.