• Please review our updated Terms and Rules here

Keyboard error, press <F1> to enter setup OMG o_O

A

aSaSSinoooo

Member
Joined
Mar 24, 2009
Messages
19
Hello everyone,

first of all, sorry for my bad english, I'm Italian. I found this great forum jumping around with google :)

I'm in trouble with my old 386 pc, I cannot identify motherboard brand/model, the amibios string is "30-01]1-ZZ1326-00101111-043091-HT18SX-0" and it has a 386sx onboard and 387 coprocessor on slot.

Yesterday I tried to boot it after some year, 3 times it booted ok, after 3 times it freezed on boot with "Cmos error, default loaded" and "Keyboard error, press <F1>". Battery had some electrolyte out, and I've seen on some forum that cmos battery low may cause bios fault, so i unsoldered old ni/cd battery and soldered a new one, now I've only the keyboard error.

Sometimes the keyboard seems running ok for 5, 10, 20 seconds, in one of this rare moments I've entered bios setup and set bios to not halt on keyboard errors, but i've not resolved my problem. Btw so I think that (fortunately) keyboard bios is ok.

Tomorrow I will try to post some photo of the motherboard!
Can someone suggest to me something to try to fix it? (I don't want to buy another one :) ) I've tried to put a deoxidant spray on the keyboard connector and checked the three fuses near this connector (I suppose that components marked as "FB" are fuses).

Thank you in advance.
 
I've forgot to tell you that the three leds on the keyboard flashes at startup, even with the keyboard error.

Tomorrow I'll try to re-solder contacts of the keyboard connector... hope it will fix the mobo :confused:
 
aSaSSinoooo said:
Sometimes the keyboard seems running ok for 5, 10, 20 seconds, in one of this rare moments I've entered bios setup and set bios to not halt on keyboard errors, but i've not resolved my problem. Btw so I think that (fortunately) keyboard bios is ok.
Click to expand...
It really sounds like the acid has damaged a copper track that is part of the keyboard circuitry, which makes sense because the battery is normally located near the keyboard connector/circuitry.

It could also be a bad solder joint where the keyboard socket is soldered to the motherboard. Try powering on the machine with your hand pushing the keyboard plug slightly to one side, then try the other direction, etc.

It could also be a bad keyboard. You don't have a spare to try?

aSaSSinoooo said:
and checked the three fuses near this connector (I suppose that components marked as "FB" are fuses).
Click to expand...
No. FB is an acronym for ferrite bead. Think of them as a doughnut of ferrite that sits around the wire. They reduce RF signals getting in/out of the computer.

aSaSSinoooo said:
I've forgot to tell you that the three leds on the keyboard flashes at start up, even with the keyboard error.
Click to expand...
Keyboards normally do that when they get power.
 
First of all, thank you for your prompt reply :D

modem7 said:
It really sounds like the acid has damaged a copper track that is part of the keyboard circuitry, which makes sense because the battery is normally located near the keyboard connector/circuitry.

It could also be a bad solder joint where the keyboard socket is soldered to the motherboard. Try powering on the machine with your hand pushing the keyboard plug slightly to one side, then try the other direction, etc.
Click to expand...

I tried this, but no results. I've seen something strange on a copper track, but copper tracks are covered by a green plastic sheet, and they seems to be tracks from ali to isa slots, not related to keyboard connector. Tomorrow I will post some photo...

modem7 said:
It could also be a bad keyboard. You don't have a spare to try?
Click to expand...

Yes, I've 3 AT keyboards, any of these running on my 486 so i can exclude bad keyboard.

Thank you!
 
Is the keyboard controller (a 40 pin DIP package usually near the keyboard connector) in a socket? If so, try removing it (gently) from the socket, give the pins a visual inspection and a bit of cleaning if corroded and then replace it. Sometimes, mechanical stresses will cause a socket to produce an intermittent connection.
 
When I powered up my 486 after being dormant for many years I found the battery electrolyte had eaten through some of the tracks and the mouse wouldn't work. (Serial port U/S)

I put another board in and that worked fine for a day or so (after also cleaning off the electrolyte as this one had also rotted, but not as bad). There were a couple of tracks under the keyboard socket which I scraped clean and tinned, I also cleaned the crud off the DIN socket before I refitted it, but then the keyboard stopped working after a day or so. Although the tracks were supposedly OK when I fitted the board, powering it up must have finished it off as on examination one track had developed an obvious gap under the green lacquer and measured O/C. I scraped the lacquer off and soldered a wire across it and it cured the fault. So it seems the nasty stuff from your NiCd can cause more damage after you thought you'd dealt with it ;)


BG
 
Chuck(G) said:
Speaking from experience, battery electrolyte is nothing compared to cat urine when it comes to dissolving PCB traces...:puke:
Click to expand...
What! You mean that all this time I've been buying expensive chemicals to etch my boards, when I could have been using my cat's litter tray instead !
 
Thank you for your interest in my trouble :D

Here are some photos of the board (sorry, i've not put imagese here because they are too big to fit in the webpage):

First to change battery, board w/oxide and electrolyte tracks: http://i40.tinypic.com/2gxkzur.jpg

After battery change, used a spray to clean contacts, I removed also AMI KEYBOARD BIOS and cleaned pins: http://i44.tinypic.com/24w57bs.jpg

Zoom of another place covered by electrolyte and oxide: http://i41.tinypic.com/2zi1rvc.jpg

Zoom of 3 copper tracks that seems darker than normal (2 biggest, 1 smallest near the resistor R3): http://i40.tinypic.com/124ja69.jpg

It seems that the green protection over the copper tracks is intact, but tracks under are darker, do you think that they may be corroded?

See you!
 
I'd start by testing for continuity/resistance on the suspect-looking tracks.
aSaSSinoooo said:
Zoom of 3 copper tracks that seems darker than normal (2 biggest, 1 smallest near the resistor R3): http://i40.tinypic.com/124ja69.jpg
Click to expand...
That narrow track leading from the keyboard connector towards C5 is definitely worth looking at, as is the diagonal one passing the positive end of the battery (where does that go?). Looks like there's also a plated hole at the top centre of that photo that may need checking. R3 is also chipped but I don't think that should matter too much.
 
Solved!!!! :d

Solved!!!! :d

cosam said:
I'd start by testing for continuity/resistance on the suspect-looking tracks.

That narrow track leading from the keyboard connector towards C5 is definitely worth looking at, as is the diagonal one passing the positive end of the battery (where does that go?). Looks like there's also a plated hole at the top centre of that photo that may need checking. R3 is also chipped but I don't think that should matter too much.
Click to expand...

SOLVED!!! :D :D

I haven't seen the "hidden" track from the keyboard connector and C2:

r7udl5.jpg


so i turned upside-down the motherboard and tested continuity with a multimeter, the track was broken so I made a "bridge" with a simple wire and a soldering iron. (later i will post photos)

Now this pc runs better than ever :D

I want to thank you everyone for helping me, hope to contribute to this forum some day!

See you!

Luca
 
Last edited:
Speaking from experience, the keyboard is always the first thing to go when the battery leaks. If the damage is not too bad, it may be repairable, but most of the time, it ain't worth the effort. So many hidden tracks, etc...

--T
 
I don't know HOW, but I've LOST my Usb cable to connect camera to pc. Obviously it is the same cable used to connect the card reader to my pc, so I cannot upload photos... I hope to find it in a few days.

In the meantime, I'm trying to install something on this 386, actually I added a 166M Hard Disk and an Atapi CDROM, but with some trouble... tomorrow I'll try to install some kind of Windows, just for testin purpose...

Anyone can suggest me some good -light- actual linux distro?

Thanks!
 
Hello,

I had/have the same problem as described in this thread. After 15 years I decided to power up my 386 again but I never removed the battery,
how dumb ... Nevertheless I tried to boot the system and it came up with the bios utility, after some configuration changes I restarted the
machine and it loaded DOS without problems.

I decided to leave it for a while and about 30 minutes later I noticed that characters were appearing on the screen, not normal character but something
like this ^C^@ etc.

Maybe it was the keyboard that has gone bad, I connected the original one from that time, knowing that it is not in a best condition, so my conclusion
was that the keyboard was responsible for those chars.

After a while i decided to connect a harddrive to it, but after installation and booting it up again I got the above mentioned keyboard error and was
not able to boot DOS again ...

I searched the internet and found this thread and decided to change the battery (I would have anyway ...).

The system is booting now again without problems, my Samsung 500Mb drive is spinning happily, but the random characters problem has not been
solved. I am using an other keyboard now and that didn't help either.

I cleaned the motherboard but like the pictures in this thread, my board also has some traces of damage from the leaking battery. Also a small part
of the green coating has come off.

It looks like if the system is frozen but after some random key pressing it comes back again (sometimes it takes quite a long time) with the c:\>
prompt and the pc goes on as nothing has happened.

Could those random characters been caused by this damage ?

The battery seems ok, the bios keep its settings (power cord disconnected for like 12 hours).

I did some test with hardware utilities but none of them reported any faults.

Anyway, if more information or pictures are needed, please let me know because I would like to fix this baby.

Gtx,
Marco
 
Bridge the thin traces (yes there are two from the keyboard connector) that corroded off and it should work. The other two come from the power socket, so you'd see much worse things than a keyboard error if they weren't connected - the one from the keyboard socket, however, is quite likely.

On another note, where did you get this battery? I have original batteries on all of my boards, and on ones where it has corroded badly, I just plucked them off and left them be. On my really tiny 386 mobo, it corroded, but didn't affect the function of the board - it even remembers CMOS info - but I left the battery there, after cleaning the stuff it leaked while in storage... probably not a great plan.
 
Check the keyboard controller - it's usually a long thick chip bigger than the BIOS one (the BIOS one will be marked, so you can tell what one THAT is). You can get new ones if yours is too badly damaged, but I'm willing to bet there's just some corrosion on the pins someplace. Pull it out of the socket and inspect/clean it. Hopefully that's your problem.

Marco said:
Hello,

I had/have the same problem as described in this thread. After 15 years I decided to power up my 386 again but I never removed the battery,
how dumb ... Nevertheless I tried to boot the system and it came up with the bios utility, after some configuration changes I restarted the
machine and it loaded DOS without problems.

I decided to leave it for a while and about 30 minutes later I noticed that characters were appearing on the screen, not normal character but something
like this ^C^@ etc.

Maybe it was the keyboard that has gone bad, I connected the original one from that time, knowing that it is not in a best condition, so my conclusion
was that the keyboard was responsible for those chars.

After a while i decided to connect a harddrive to it, but after installation and booting it up again I got the above mentioned keyboard error and was
not able to boot DOS again ...

I searched the internet and found this thread and decided to change the battery (I would have anyway ...).

The system is booting now again without problems, my Samsung 500Mb drive is spinning happily, but the random characters problem has not been
solved. I am using an other keyboard now and that didn't help either.

I cleaned the motherboard but like the pictures in this thread, my board also has some traces of damage from the leaking battery. Also a small part
of the green coating has come off.

It looks like if the system is frozen but after some random key pressing it comes back again (sometimes it takes quite a long time) with the c:\>
prompt and the pc goes on as nothing has happened.

Could those random characters been caused by this damage ?

The battery seems ok, the bios keep its settings (power cord disconnected for like 12 hours).

I did some test with hardware utilities but none of them reported any faults.

Anyway, if more information or pictures are needed, please let me know because I would like to fix this baby.

Gtx,
Marco
Click to expand...
 
Thanks for the help sofar. I pulled the keyboard chip out of the socket and
it looked very clean to me, still under the socket there was a lot of dust
and dirt ;) - Nevertheless, i cleaned that and put the chip back in. I rebooted
the computer and its on now for almost an hour without problems ....

Hopefully that solved the problem, but I we will see.

Another question, what do you exactly mean by bridging, just soldering
a wire ? It is necessary to cut the original 'wire' in the board or can it
be left alone ?

Thx,
Marco
 
I was talking to the other guy, apparently in the time it took me to write my post he solved it, or I skimmed over him solving it before.
 
You must log in or register to reply here.
Back
Top