How to repair Mylar cable on Muppet Learning Keys keyboard

[electronizer]

I got one of these a few years ago. From the start, it worked intermittently—sometimes the letters worked correctly, sometimes the rows were shifted so that e.g. when I pressed Q, it registered as K (the letter above it). Eventually it stopped working altogether, to the point where the driver disk wouldn’t even detect it.

 

I took it apart to find “reject” stamped over the “QA Approved” stamp (uh-oh), and two flat Mylar cables coming out of the panel. I used a pencil eraser on the traces where they connect and they cleaned up nicely. However, only one Mylar cable connects to the joystick cable. The other one is connected by two “staples” that pass through the traces on both cables. When I measure the resistance between these staples and the contacts on the cable that connects to the joystick cable, one is high (several megohms), and the other is open—no connection. So, I’m guessing this is where the problem is.

 

Anyone have any ideas about how to repair this stapled connection between the cables?

[StickJock]

What pins on the joystick port does this use?  Is it using a resistor network to give each key its own resistance value which goes to a paddle input?

 

[_The Doctor__]

conductive paint, will restore contact along the stapled traces, cleaning / trimming / refolding might help with connector contacts. Basically like any other mylar keyboard fix.

[electronizer]

StickJock, good guess! Looks like it’s using pins 7, 5, and 9, which would be +5V and both paddle inputs. Resistance from pin 7 to 9 is 11Mohm, and it doesn’t change when I press the keypad.

 

Pin 7 to 5 appears to be open, again with no change when pressing the keypad.

 

The Doctor, any thoughts on how I would reconnect the stapled traces? I’m thinking I would almost have to rip out the staples, add conductive paint, and then put new staples in before it dries.

[_The Doctor__]

no you should be able to get the paint in and around the staples and along the traces... once it starts to dry and work assemble the pad.... it will finish drying  in the position it naturally sits inside at that point. If you wait for it to completely dry and then assemble it may stress and crack and flake off.

 

did you examine the rest of the circuits and test for breaks along the traces any possible supporting components?

[StickJock]

Do the individual keys have any kind of tactile feedback, or does the entire surface feel the same (key & non-key area)?

I'm wondering if it is implemented as a simple resistive panel, where the two pots read X&Y, like a Koala pad.  Do you have to do any kind of calibration when you use it?

 

[electronizer]

Thank you both for the quick replies.

 

There is no other circuitry visible and the panel looks like it would be damaged beyond repair if I tried to peel off the keyboard layer.

 

The keys are raised pads and definitely have tactile feedback, different from the non-key area. No calibration is required, however, the driver disk can detect whether the keyboard is connected to the joystick port. So, it must be looking for a particular resistance at the paddle inputs.

[_The Doctor__]

Interesting in that it also says Apple... wonder if there are differences....

if the conductive paint get some response the resistance might be out of range.. you could experiment by adding resistance to the pot lines... adjusting the the resistance....

Edited June 27, 2020 by _The Doctor__

[StickJock]

Resistance is futile. You will be assimilated. 

[electronizer]

Yes, that is interesting considering the disk I got with it has C-64 software on one side and the Atari driver on the other side. There was no special software for the Atari, instead, the driver disk just allows you to use the Learning Keys keyboard in place of the Atari’s keyboard. When it was working, it worked pretty well with Hodge Podge, where pressing a different letter gets you a picture of something that starts with that letter (e.g. “K” gives you a koala).

[electronizer]

Finally got a chance to look at this again and I figured it out. The Mylar cable was bent at the end, probably to give a better fit in the connector. The traces ended up breaking right at the bend, and the edge connector plug was only making contact on one side of the traces, the side that came after the break.

 

I solved the issue by swapping the outer two connections and flipping the connector over so that it was making contact on the other side of the cable, before the break. I also used a pencil eraser to gently clean the traces. It works very well now—happiness!