I've got this 122-key monster a few days ago. It had an RJ-45 connector, and I only had USB, so all I could do was cut the cable, and solder a Teensy with Soarer's firmware on it, which I did.
I also cleaned it thoroughly. In the hindsight, maybe I shouldn't have. The fact is: the Teensy itself works, but no keys whatsoever are registered. I checked pinout and wire coloring twice before I soldered it, so getting Vcc and ground is ruled out. I also checked with a multimeter that joints themselves are OK — they deliver 5V to the controller. So, it must be either membrane or controller.
While cleaning, I did unscrew the controller from its place, and reseated the membrane connections. The membrane itself is quite old, and the connection points are blackened with oxidation, except for the parts that actually make contact with the connector — they look like polished silvery spots.
So, I don't know if it's the controller that's dead, or is it the membrane. To replace the membrane, I'd need to actually do the full bolt mod, because the rivets will have to go, too. I'm a bit reluctant to do that.
Also, if I replace everything — say, order a new controller from Phosphorglow, a new membrane from Unicomp, etc — it will be a hefty premium. So I want to positively rule out which component is at failure.
The question thus is: how do I do that? I thought about connecting the controller, sans membrane, to the Teensy, running hid_listen, and trying to short rows and columns at the connector itself to emulate keypresses. If it's the membrane, the controller would register keypresses. If the controller is dead, well, it's dead.
Or go through the pains of checking continuity at the membrane itself. Still a pain, as I would have to take it out (and order all the nuts and bolts).
Suggestions welcome.