Please help identify component on keyboard logic board

[newtekuser]

Hello all,

I'm trying to revive an old xt/pc keyboard that throws a keyboard error and started to poke inside it and found these two components encased in a clear case which look like fuses to me.

Can someone please confirm? If that's the case they are dead as I put my meter in continuity mode and didn't get any beeps across the leads.  Furthermore, what fuse value should I replace them with if that's what they really are?

[Ian.M]

https://www.eevblog.com/forum/metrology/mlccs-encapsulated-in-glass/

[fzabkar]

Can you measure +5V at the keyboard connector? If not, look for an open polyfuse.

[newtekuser]

Thank you both!

I did find what looks to be a polyfuse. If this is what it is then it's dead (measured resistance across the leads/no continuity). What replacement fuse would I need for this application? I take something rated for 5v, maybe 500mA max?

As for measuring voltage on the keyboard, do you mean while pc is powered on, or should I inject 5v into the end coming from the pc side?

[fzabkar]

That's a ceramic capacitor.

In my experience you should find a picofuse. (I was wrong to suggest a polyfuse, as these are resettable components.)

Something like this, probably 1A ...

https://au.rs-online.com/web/p/non-resettable-fuses/5413287p

In fact I was wondering whether your first photo has such a device.

As for the keyboard connector ...

https://old.pinouts.ru/InputCables/KeyboardXT_pinout.shtml

[newtekuser]

As far as I can tell it's got: resistors, ceramic caps, zenners, schottkys and some tantalums.

Also keyboard connector is proprietary not the standard DIN, pincount is same

[fzabkar]

Find the +5V pin in the power connector, then perform a continuity test with each pin in the keyboard connector. Then power up the motherboard and see if you can measure +5V at this pin.

Edit: If there is no continuity with the +5V supply at the keyboard connector, then this would suggest that there is an open fuse.

[newtekuser]

Find the +5V pin in the power connector, then perform a continuity test with each pin in the keyboard connector. Then power up the motherboard and see if you can measure +5V at this pin.

Good idea, thanks!

[newtekuser]

I cannot find any pinout diagram for this connector but I measured +5v, +5v and +12v on some of its pins.

[fzabkar]

I've never seen an XT/AT/PS2 keyboard that required a 12V supply. Very strange ...

[newtekuser]

I've never seen an XT/AT/PS2 keyboard that required a 12V supply. Very strange ...

This is similar with what I have: somewhere in the article it says it runs on 12v

https://www.reddit.com/r/MechanicalKeyboards/comments/mrzj1j/i_successfully_brought_this_compaq_portable_1/

[fzabkar]

I've never seen an XT/AT/PS2 keyboard that required a 12V supply. Very strange ...

This is similar with what I have: somewhere in the article it says it runs on 12v

https://www.reddit.com/r/MechanicalKeyboards/comments/mrzj1j/i_successfully_brought_this_compaq_portable_1/

I still can't see which component needs 12V. :-?


Edit:

Is there a 3-terminal 7805 regulator under the extended metal area? If so, maybe the incoming 12V supply is being down-regulated by the 7805. I would measure those voltages.

https://i.redd.it/ki4ja6hd6it61.jpg

[newtekuser]

I've never seen an XT/AT/PS2 keyboard that required a 12V supply. Very strange ...

This is similar with what I have: somewhere in the article it says it runs on 12v

https://www.reddit.com/r/MechanicalKeyboards/comments/mrzj1j/i_successfully_brought_this_compaq_portable_1/

I still can't see which component needs 12V. :-?


Edit:

Is there a 3-terminal 7805 regulator under the extended metal area? If so, maybe the incoming 12V supply is being down-regulated by the 7805. I would measure those voltages.

https://i.redd.it/ki4ja6hd6it61.jpg

I measured voltage on all ICs and each is getting 5v. Didn't measure the voltage reg since I assumed to be working with everything being 5v downstream.

[fzabkar]

I suspect that the keyboard controller is an 8048 or 8748 or another mask ROM-ed member of that same series. I would confirm its supply pins and try to read it in a programmer. Failing that, find a replacement chip from any other XT keyboard and see if it at least communicates. The row and column matrices will probably be different, though.

Edit: I can see "8048" in the part number.

[newtekuser]

I suspect that the keyboard controller is an 8048 or 8748 or another mask ROM-ed member of that same series. I would confirm its supply pins and try to read it in a programmer. Failing that, find a replacement chip from any other XT keyboard and see if it at least communicates. The row and column matrices will probably be different, though.

Edit: I can see "8048" in the part number.

Mine is actually using an Intel P8035L

Edit: I found one on ebay, would a direct swap work or does this need to be programmed with the same code as the original? (Which I assume to be the case) stupid question, these are not eeproms and cannot be programmed more than once.

https://www.ebay.com/itm/283861894704?hash=item42177cc630:g:X9EAAOSwxBNeqIZg

Edit2: I found the datasheet for this controller and measured the reset pin and found it high (reading 5v)

https://datasheetspdf.com/mobile/541972/Intel/P8035AHL/1

[fzabkar]

The 8035 has no ROM. Therefore there must be an external EPROM chip. :-?

https://en.wikipedia.org/wiki/Intel_MCS-48

Edit:

I didn't see your pics before I posted.

[fzabkar]

Edit2: I found the datasheet for this controller and measured the reset pin and found it high (reading 5v)

https://datasheetspdf.com/mobile/541972/Intel/P8035AHL/1

The reset pin is active low. Therefore it should be sitting at 5V when the uC is running.

Do you have an oscilloscope? If so, check the oscillator, and look for activity on the address and data lines of the EPROM.

Alternatively, if you have a programmer, dump the EPROM several times and compare the dumps. There may even be a checksum which we could verify.

[newtekuser]

Edit2: I found the datasheet for this controller and measured the reset pin and found it high (reading 5v)

https://datasheetspdf.com/mobile/541972/Intel/P8035AHL/1

The reset pin is active low. Therefore it should be sitting at 5V when the uC is running.

Do you have an oscilloscope? If so, check the oscillator, and look for activity on the address and data lines of the EPROM.

Alternatively, if you have a programmer, dump the EPROM several times and compare the dumps. There may even be a checksum which we could verify.

Thank you for the correction!
I do have both, unfortunately my scope (tds5054) only plots a square wave (otherwise just noise) when hooking up the probe to its signal calibration pin and my programmer (mikroprog) only does 5v programming

[fzabkar]

Just read the EPROM several times. That only requires 5V. Then upload the dump(s) and we can look for a checksum.

[newtekuser]

Looks like the scope wasn't dialed in properly (I'm a newb). I do see activity when probing pins 2, 3