VOGONS


First post, by joebelter@hotmail.co

User metadata
Rank Newbie
Rank
Newbie

Finally got my no name Varta damaged 386/486DLC motherboard to the CMOS screen, but can't get past the "press F1 to continue" screen.
It says the ram is ok, CMOS changed, but keyboard doesn't work.
When I reboot I do get the three green lights on the keyboard when it initializes...
I cleaned up all my traces, looked for micro damage, now trying to check for continuity errors.
I saw a number of posts that directed to check the pinouts to the keyboard controller, but my continuity checks aren't like anyone else's!
I checked the net for pinout reference for AT computers...
This is what I came up with.
Pin 3 - Pin 1 on Keyboard controller
Pin 4- Pin 26,40 on the Keyboard controller
Pin 5- Pin 33 on the Keyboard controller. I couldn't find a "1" on the controller so assumed "U1" on the upper left of the chip where the u shaped cutout is. Check my picture above for the chip.
I tried looking this up, but makes no sense. (Oh,l and I did remove the solder bridge on the traces I had there near the keyboard controller as well as cleaned out the VIAs with a tiny .31 bit).

Really APPRECIATE any tips on where to measure voltages or continuity...
Thanks,
Joe

Reply 1 of 5, by Horun

User metadata
Rank l33t++
Rank
l33t++

Need to know the exact KB controller model. Some have diff pinouts and you are missing the DATA signal.
On a 8042 types (that are direct pin replaceable to Intel8042/8047, HT6542B, VT82c42, W83C42). Pin 1 is KB Clock and Pin 39 is DATA iirc.
Where the Ground and +5 come from should be the power planes, also Pin 20 and pin 40 of KBC repectively...
Note: the Clock and DATA may go thru a buffer or inverter ic before going to the KBC on some boards...

Hate posting a reply and then have to edit it because it made no sense 😁 First computer was an IBM 3270 workstation with CGA monitor. Stuff: https://archive.org/details/@horun

Reply 2 of 5, by joebelter@hotmail.co

User metadata
Rank Newbie
Rank
Newbie

Thanks for the reply... Just as a followup...
Ok, I'm stupid. I admit it!
I don't know what the HECK I was doing/thinking...
But I got out a working 386sx and did a comparison pin for pin between the keyboard connector and the keyboard programmer.
EVERYTHING was exactly the SAME except the data pin to pin 38 on the Keyboard programmer. No continuity there.
I discovered that TINY trace was missing between the back of the keyboard socket and where the trace is on the board. I put a wire bodge wire there and BANG - continuity!!
I then checked the two boards against each other for continuity to the PSU connector on the board. They are identical!
My heart pounded with excitement as I booted the board...
but alas... NO CHANGE!
Noooooooooooo!
What now?

Joe

Reply 3 of 5, by joebelter@hotmail.co

User metadata
Rank Newbie
Rank
Newbie
Horun wrote on 2025-02-19, 03:31:
Need to know the exact KB controller model. Some have diff pinouts and you are missing the DATA signal. On a 8042 types (that a […]
Show full quote

Need to know the exact KB controller model. Some have diff pinouts and you are missing the DATA signal.
On a 8042 types (that are direct pin replaceable to Intel8042/8047, HT6542B, VT82c42, W83C42). Pin 1 is KB Clock and Pin 39 is DATA iirc.
Where the Ground and +5 come from should be the power planes, also Pin 20 and pin 40 of KBC repectively...
Note: the Clock and DATA may go thru a buffer or inverter ic before going to the KBC on some boards...

You probably don't feel like telling me how to go about checking this out... I'm really new to this game and learning SLOW as I go. I'm old, retired, and willing to learn though!!

Joe

Reply 4 of 5, by MikeSG

User metadata
Rank Member
Rank
Member

To get the exact KB controller model you need to take the sticker off the top, and it should show name and part number. Then you can track down a datasheet and look at the pinout.

After you have the pinout then you can track where the lines should go.

Clock & Data often go through a buffer chip between KB controller and connector. Eg. 74F04N

Reply 5 of 5, by Nunoalex

User metadata
Rank Member
Rank
Member

The problem can be north of the keyboard controller

can you inspect where every trace from the kboard controller goes and if you have continuity ?
Also if there are some diodes on the way would be a good idea to check if they are ok... i.e. if they conduct in one way but not on the other

Good luck !

Nuno