IBM model M needs help

[Anonymus91]

Hello guys, first post here.
I recently found an abandoned IBM model M on the ground in an old factory. It was in pretty good shape, except it was covered in tons of dust and had it's cable ripped of.

20180201_214442.jpg (2.03 MiB) Viewed 1677 times

Took it home and cleaned up and then I tried to replace the PS2 cable with a new one from another keyboard, obviously it didn't work so I had to rewire it. Left existing ripped off cables and right side the new PS2 cable.

Screenshot_20180202-111416.jpg (98.59 KiB) Viewed 1677 times

Keyboard came online and numpad lit up. Thought it would work but most of the keys don't work and pressing a letter from mid row would output 2 characters like so: S would type - SW, F would be FR, and the top row would output the top row key plus the mid row W->WS...
Now my question is, could this be because of faulty wiring? Or if the cables were not in the correct true the keyboard would not work at all?
And most important how can I make all keys work given that mechanically they all work just fine.
Also saw it I moved around the PCB where the ribbon connects some keys would come online and others wouldn't work.
If anyone could help me I would be really grateful, been searching for a long time for an IBM model M:)

20180130_201336.jpg (2.23 MiB) Viewed 1677 times

20180201_214359.jpg (2.01 MiB) Viewed 1677 times

[PlacaFromHell]

Hello. For what I understant at least your keyboard is being recognized by your computer. I had a similar problem while wiring and connecting my model M122's soarer converter, but instead of multiple keypresses I was getting null hexadecimal codes. Check the data and clock wires (I believe that you would'nt damage the keyboard if you swap them, maybe someone can confirm that) and see if the VCC and ground are properly wired. I solved that problem with just that, swap the clock and data cables that we're inverted and resolder the VCC that was making wrong contact.
You can see the PS2 port pinout too and use a tester to see what pin correspond to what wire. You can get a molex and just solder each cable directly to the new tab for be sure about the contact issues.
Also I had a more similar problem with other keyboard (rest in peace, Unitech dome with slider), the controller whas fucked but if it the case you could buy one from Unicomp I think.

[Anonymus91]

Hello. For what I understant at least your keyboard is being recognized by your computer. I had a similar problem while wiring and connecting my model M122's soarer converter, but instead of multiple keypresses I was getting null hexadecimal codes. Check the data and clock wires (I believe that you would'nt damage the keyboard if you swap them, maybe someone can confirm that) and see if the VCC and ground are properly wired. I solved that problem with just that, swap the clock and data cables that we're inverted and resolder the VCC that was making wrong contact.
You can see the PS2 port pinout too and use a tester to see what pin correspond to what wire. You can get a molex and just solder each cable directly to the new tab for be sure about the contact issues.
Also I had a more similar problem with other keyboard (rest in peace, Unitech dome with slider), the controller whas fucked but if it the case you could buy one from Unicomp I think.

Ok I will try to swap the cables see if that helps. One other thing I noticed is that the PCB has 1extra pin that doesn't touch the contacts on the membrane and don't know if that's normal.

[PlacaFromHell]

Ok I will try to swap the cables see if that helps. One other thing I noticed is that the PCB has 1extra pin that doesn't touch the contacts on the membrane and don't know if that's normal.

Normaly much model M's come with an extra space between the membrane traces and leaves a pin with nothing to do, I never know why. Anyway, if isn't that I'm saying you can take a picture of it and see if someone knows what is this. Also the membrane can be weet, causing bridges between the traces and making weird presses, a friend told me about using a cable to test te controller for confirm that kind of issues.
Remember never swap VCC and ground, that can kill your board

[Anonymus91]

20180202_202946.jpg (2.08 MiB) Viewed 1602 times

Here is a picture of the pins.
Cleaning the membrane could solve this then? And is it possible to reach the membranes without doing a screw mod?

[PlacaFromHell]

20180202_202946.jpg

Here is a picture of the pins.
Cleaning the membrane could solve this then? And is it possible to reach the membranes without doing a screw mod?

The membrane looks like one trace out of phase, I can't say if this is normal or not, but maybe here is the problem. I don't think you'll need to open the sandwitch, only one trace looks a bit weathered and my model M works with much more punishment than that. If you want to clean the membrane you have to destroy the rivets to open the keyboard. Some people just glue the plastic to the backplate instead of do a screw or bolt mod, but I don't recomend this.
If you want you can try luck and give him a good internal cleaning (and with a boltmod necessarily), if this don't work you can just ask to Unicomp for a new controller, just contact them and ask if they have a controller for your keyboard part number (the Classic 101 should be pretty much the same, but ask them anyway if you are considering this option).

[Anonymus91]

Did a screw mod on it, had to clean the membranes with alcohol since rust got in then redrew the circuits with graphene pencil. Put it together and everything is working fine except the insert button. Also managed to find another one from '85 that needs just a cleaning. Can't decide witch one to keep for home.

20180206_230749.jpg (2.3 MiB) Viewed 1542 times