IBM Model F XT acting strange

[snacksthecat]

I restored this XT a while back. Had a lot of fun doing it too. Unfortunately, at no point in the process did I think to test it so I'm not sure if it even worked to begin with.



I went to use it and it's not behaving very well. I'm trying with a Soarer's converter I made which works fine with any AT keyboard that I have.

When I try with this XT, usually the keys don't register unless I mash them fairly hard. Then when they do register, they just keep repeating.

What typically goes wrong with these things? I didn't replace the foam, is that something that I should have done? Thanks!

[Ir0n]

Bad foam in model fs makes them pretty much useless.

Any time you restore these you should always try to replace the foam.. Not always needed, but more than likely that's the issue here.

[orihalcon]

When you restored it, did you clean the capsense PCB or flippers? Any dust in there would give you those problems. If the restore was more just cleaning externally, then that's the likely culprit. Foam replacement is nice, but not always a must in terms of functionality. If you remove and clean the barrels, you usually will destroy the foam in the process unless you got really lucky and have one with non-degrading foam which a few XTs and 3178's had, or possibly other models that got refurbished by IBM maybe. If you have non-degraded foam, you should re-use it. Should never go bad.

[JBert]

First of all, did you take the XT apart and are you sure it's properly reassembled? I have read reports in the past of people testing it outside of the case or with the screw between controller and backplate missing which caused problems like these.

Are you sure that all keycaps were properly snapped into place over the buckling springs? That is, if you press them very slowly can you hear the flipper and spring buckle? If you took off the backplate (you shouldn't unless you know you have more than a few hours and should follow a guide), did you check that the spacebar flipper isn't stuck?

(A tip w.r.t. the keycaps: the best way to attach keycaps is to set the keyboard vertical (space bar up, controller down) on your lap, keeping it steady with one hand and using the other to reattach keycaps one by one. That way the springs should be "straight" in their little tubes and the keycaps will properly sit on top of them.)

[snacksthecat]

Bad foam in model fs makes them pretty much useless.

Any time you restore these you should always try to replace the foam.. Not always needed, but more than likely that's the issue here.

The foam seemed good so I left it be.

First of all, did you take the XT apart and are you sure it's properly reassembled? I have read reports in the past of people testing it outside of the case or with the screw between controller and backplate missing which caused problems like these.

Are you sure that all keycaps were properly snapped into place over the buckling springs? That is, if you press them very slowly can you hear the flipper and spring buckle? If you took off the backplate (you shouldn't unless you know you have more than a few hours and should follow a guide), did you check that the spacebar flipper isn't stuck?

(A tip w.r.t. the keycaps: the best way to attach keycaps is to set the keyboard vertical (space bar up, controller down) on your lap, keeping it steady with one hand and using the other to reattach keycaps one by one. That way the springs should be "straight" in their little tubes and the keycaps will properly sit on top of them.)

Thanks for the tips. Yes I took it completely apart. Not sure if it worked to begin but I believe I put everything back together right. I'll start back from square one.

When you restored it, did you clean the capsense PCB or flippers? Any dust in there would give you those problems. If the restore was more just cleaning externally, then that's the likely culprit. Foam replacement is nice, but not always a must in terms of functionality. If you remove and clean the barrels, you usually will destroy the foam in the process unless you got really lucky and have one with non-degrading foam which a few XTs and 3178's had, or possibly other models that got refurbished by IBM maybe. If you have non-degraded foam, you should re-use it. Should never go bad.

I cleaned the flippers real nice and just now took it apart again and cleaned the pcb.

I've got some time today so I'm going to try and work out the issue. Here's how it's behaving:

[snacksthecat]

Here it is down to the PCB assembly:

Edit: Same result if I remove the backplate and just have the bare pcb plugged in. Bad component?

[snacksthecat]

Something seems to be pulling the voltage down. I wired it up directly to the teensy and it works much better but still some ghosting. The cable is pulling it down to ~4.5V and something else on the pcb is pulling it down further to ~4V. I think that might be my problem.

[snacksthecat]

Still a bit stuck on this (don't know much about electronics). Is there a way I can determine what is pulling the voltage down? I poked around with the multimeter but only came to the following conclusion that I posted previously

Voltage is dropping to 4.5v from the cable
Being further pulled down to 4v when I measure everything connected up

[JBert]

I wonder how you're measuring this. Though if you think the cable already is a voltage drop, couldn't you unplug it and check it with the resistance setting of your multi-meter?

[andrewjoy]

is the plate and controller grounded properly ?

have you cleaned the PCB properly ?

[snacksthecat]

Total noob mistake. I was using a Soarer's converter that didn't have the din sleeve wired to Gnd.