PDA

View Full Version : Bad Keyboard Interface



Stone
April 20th, 2017, 10:47 AM
I have a few AT class motherboards which give a 'keyboard interface error press f1 to resume' message when booted. None of these uses a PS/2 style keyboard -- they all use the AT style keyboard connector.

Is there any remedy or means to bypass this problem since there's no USB keyboard possibility with any of these boards? Without a keyboard there's no way to boot any of these motherboards.

I've checked the connectors themselves for open circuits and cold solder joints but that isn't the problem with them.

Chuck(G)
April 20th, 2017, 01:37 PM
Most use dedicated MCUs for the keyboard interface. (8742 type). If those go bad, the only remedy is to substitute a new-or-reprogrammed unit. Some BIOSes would allow you to boot without a keyboard.

FWIW, there's no practical difference between PS/2 and AT interfaces, other than the connector--and adapters are/were available. But I'm sure you knew that.

snuci
April 20th, 2017, 01:44 PM
I;m sure you are aware of this already but did you confirm they did not require an XT keyboard? Just putting it out there as they use the same connector as AT but are a different protocol.

MikeS
April 20th, 2017, 02:19 PM
Not a keylock issue by any chance?

Stone
April 20th, 2017, 03:11 PM
Most use dedicated MCUs for the keyboard interface. (8742 type). If those go bad, the only remedy is to substitute a new-or-reprogrammed unit. Some BIOSes would allow you to boot without a keyboard.

FWIW, there's no practical difference between PS/2 and AT interfaces, other than the connector--and adapters are/were available. But I'm sure you knew that.Since I can't access any of these BIOSes I don't have that possibility even if they do offer it. :-)

I only stated AT keyboard to give a more complete description of the situation.


I;m sure you are aware of this already but did you confirm they did not require an XT keyboard? Just putting it out there as they use the same connector as AT but are a different protocol.I have used all of these boards many times. As it turns out the keyboard problem appeared intermittently, at first, and gradually worsened on several of the motherboards.


Not a keylock issue by any chance?Nope, same results with all keyboard I tried.

Bungo Pony
April 20th, 2017, 04:23 PM
I have a few AT class motherboards which give a 'keyboard interface error press f1 to resume' message when booted. None of these uses a PS/2 style keyboard -- they all use the AT style keyboard connector.

I had this problem once with a 386 of mine. The problem was a blown board mount fuse. They're green in color and look like this (http://media.rs-online.com/t_large/F5413243-01.jpg). It was located close to the keyboard connector.

After replacing it, my keyboard worked fine.

MikeS
April 20th, 2017, 05:46 PM
Nope, same results with all keyboard I tried.
No, I meant the lock connector/jumper on the motherboard, but if it was intermittent and getting worse I guess that ain't it.

Corrosion?

Stone
April 20th, 2017, 06:00 PM
I had this problem once with a 386 of mine. The problem was a blown board mount fuse. They're green in color and look like this (http://media.rs-online.com/t_large/F5413243-01.jpg). It was located close to the keyboard connector.

After replacing it, my keyboard worked fine.I'll have a look at the fuses on the boards but since they seemed to get progressively worse until there was no keyboard recognition at all I am doubting it's the fuses since they're either good or they're not.


No, I meant the lock connector/jumper on the motherboard, but if it was intermittent and getting worse I guess that ain't it.

Corrosion?These boards do have some corrosion but the batteries were all removed several years ago (in favor of external batteries) so any ill effects from leakage should have been noticed years ago.

Chuck(G)
April 20th, 2017, 06:14 PM
These boards do have some corrosion but the batteries were all removed several years ago (in favor of external batteries) so any ill effects from leakage should have been noticed years ago.

I suspect that corrosion has slowly progressed, even though you've probably done a good job of cleanup. Things get ugly when the nasty stuff creeps into the inside layers. Still you might see what a continuity tester shows.

Stone
April 21st, 2017, 04:34 AM
I had this problem once with a 386 of mine. The problem was a blown board mount fuse. They're green in color and look like this (http://media.rs-online.com/t_large/F5413243-01.jpg). It was located close to the keyboard connector.I have five boards with this problem... three 386 and two 486. None of the fuses was green. Mostly black, one beige and one clear. Anyway, all were still good.

Stone
April 21st, 2017, 05:41 AM
I suspect that corrosion has slowly progressed, even though you've probably done a good job of cleanup. Things get ugly when the nasty stuff creeps into the inside layers. Still you might see what a continuity tester shows.When I'm checking continuity am I just looking for opens or does this type corrosion also cause shorts?

KC9UDX
April 21st, 2017, 06:16 AM
It typically cause high impedance where you should have low impedance.