View Full Version : DTK PTM-1000 Phoenix BIOS error- keyboard controller

June 8th, 2014, 07:42 AM
Just got this board.
DTK PTM-1000 V.2 w a 286.


It's got some battery corrosion. Plugged it in to see if it worked. Got no video, but I hooked a speaker up and got 3-2-4 beeps.
Looked it up and get this:
3-2-4 Keyboard controller failure The keyboard controller has failed. Replace the IC if possible

Is this the Keyboard controller IC? I tried to look up the numbers but got nothing.

I reseated this chip and got the same error. Could be the error is from the corrosion and the IC is fine though.

June 8th, 2014, 07:43 AM
Could I swap this IC from another board?


June 8th, 2014, 10:26 AM
Yes, that's the keyboard controller (M5L8042). It's basically the same as an Intel 8042 (preprogrammed, of course). You might try the other controller--I doubt that it will harm anything, but it also will probably not work.

I have the motherboard manual to that board, I think (or a DTK one very close to it), but I doubt that it will be of any help. Corrosion from the motherboard battery is always a problem with these. When I'd get a new board, I'd clip the battery as one of the first things and install an external one. As an engineering manager once told me, "A PCB is no place for a battery." Wise words indeed.

June 8th, 2014, 11:55 AM
You doubt swapping the chips will work because of the chip differences, are is it because you find it more likely that the corrosion did this board in?

June 8th, 2014, 12:58 PM
I think he's saying the problem is more likely the motherboard than the IC - but it wont hurt if you want to try it.

The right hand side of that pic shows a bit of damage, might be worth checking all those traces are connected and possibly replacing the components with green legs - that oscillator sure doesn't look happy.

June 8th, 2014, 01:15 PM
I think he's saying regardless of the motherboard's condition, be it good or bad, that swapping keyboard controllers can't hurt anything but since they are not the same chip, even if everything else is good the chip swap is unlikely to be successful. IOW, only the correct chip can work in this (or any) board.

June 8th, 2014, 01:53 PM
What I'm saying is that BIOS manufacturers had this tendency to make sure that only their keyboard controller would work with their BIOS (i.e. they added "features"). But it's worth a try, although in my experience, keyboard controllers rarely go bad.

June 8th, 2014, 02:06 PM

An analogy. You're a battlefield commander. All of the troops that you send into sector F never return. No comms, nothing. You have to report to HQ only three words to describe the issue. You report "Sector F problem". Turns out that sector F is good; the troops are being ambushed on the bridge between you and sector F.

That's like what is going on when the POST reports a keyboard controller failure (and other types of failures); it could be the keyboard controller at fault, but instead, it could be anything else that stops the keyboard controller test being successfully executed.

I second SpidersWeb's suggestion. The keyboard controller test is failing, and right in that area is potential damage from battery leakage. I think the odds of it being coincidence are low.

June 8th, 2014, 05:52 PM
I did verify that I do have the manual for this board, FWIW.