Keyboard repair: MY Cherry has failed

[ricercar]

So I bought a Cherry keyboard for a buck, on principle. How can you go wrong? The month before I had bought a NIB Cherry Point of Sale (POS) Keyboard with magstripe reader for $4 USD, and read all my credit cards. What Fun. The POS keyboard was a behemoth, and this one was compact. How can I lose?



Ha. Some of the keys are dead:



I spent some time on geekhack trying to discover the keyswitch name, because this keyboard has neither MX nor ML keys.

Not MX, Not ML


In fact, the Cherry key action appeared to have nothing to do with the electrical connection, despite the copper portion of the mechanism. The mechanism was mounted over two Mylar sheets, and the Cherry action simply mashed the Mylar sheets together. To wit, pressing a finger on the Mylar in the proper location produces a keystroke just as if the Cherry action was installed.





After a long while I discovered a hidden link to the MY switch. Clever to hide this at the top of the page. You'd almost have to read linearly to find it.

What we have is a bunch of mechanical switches pressing two Mylar sheets together over a steel plate.


So anyway, my question:

Has anyone repaired MY type keyboards successfully?

In theory all I have to do is remove all the Cherry mechanisms and wipe clean the Mylar faces, and re-install. I've tested the PS/2 controller board and made a matrix for it. It is electrically able to produce all keystrokes as expected.

Is there anything to lose by removing all MY switches from my keyboard, wiping the Mylar clean, and re-assembling?

[huha]

Seems to be a problem with the matrix. Have you tried reseating the sheets?
I'd ask keyb_gr about advanced repairs, though, he does have quite a collection of MY boards.

-huha

[keyb_gr]

I have yet to dig deeper into my parts G81 - also having a range of keys not working, which is where my functional testing stopped at the time - but my main suspect is the connection between membrane and controller board. (At least that's the only potentially critical point I can think of.) Back in the olden days it was done as in the Model M, with dedicated connectors on the board, but at some point they changed it - not for the better it seems.

Perhaps I'll tear it open when I get home today, would seem a good opportunity.

EDIT: Nah, sry, not today. Thursday is a holiday though.

[IBI]

For reference the mechanism is called FTSC on the Cherry UK documentation so if you've been searching for other repairs that's another term to try.

[keyb_gr]

OK, so here's a pic of the controller PCB and its surroundings in my parts G81-3000 (that new-style case really is a bit of a pain to open btw):

Now I guess I'll have to remove it without screwing anything up.

OK, that's off. (Think it's Torx T8 screws.) Unscrew, then lift up at top left corner.
Hmm, that looks suspicious:

For some reason, there is something on the contacts, at least those with the traces coming from below. Some mechanical issue, e.g. traces rubbing when the case flexes or something?
OK, I'll grab my can of contact cleaner and some cotton swabs now...

On with the cleaning: Upon removing the goo on the traces, I discovered the potential root cause: Underneath there is a black plastic rim pressing the membrane against the PCB contacts, and for some reason this does not work equally for all of them. Maybe there can be some relative movement here, which abrades the silver traces on the affected contact sets. And no, I'm not an Athlon unlocking guru with conductive silver paint at hand.

Needless to say, there still are a large number of keys which register very intermittently or not at all.

IIRC, the G81-12100 has the old-style connectors for membrane interfacing. Guess they didn't want to spoil reliability on what was a very expensive board at the time...

Oh, and only having looked more closely at the first post now, the OP's G81-1800 seems to have something more like the old-style connectors. I'd try unlocking those, possibly cleaning the contacts and putting things together again.

The OP's G81-1800's failure pattern would coincide with a problem for lines {n,j,m} in huha's traced G80-3000 matrix.

[bsvP585hUO2Y6]

I noticed corroded/abrased contacts as well back when I disassembled
my G81-3000 for cleaning. Looks like it is that way by design. I hope
cleaning the contacts is all it needs to get your G81 working again...

[pmyshkin]

I guess the 50 million keystrokes rating doesn't mean much when other parts of your board fails. Hopefully my g80-3000 doesn't go the same way (although I've already had some intermittent problems with keys sticking).

[keyb_gr]

I guess the 50 million keystrokes rating doesn't mean much when other parts of your board fails.

Some poor old nail could use some aspirin now.

Hopefully my g80-3000 doesn't go the same way (although I've already had some intermittent problems with keys sticking).

I don't think you need to worry much about G80s. Those with the old-style case are uncritical anyway (everything on one PCB), and while those with the new-style case do feature a ribbon cable, the construction looks entirely uncritical.

Sticking keys shouldn't happen though. If particular ones do it repeatedly, you can try getting a can of compressed air and blowing out any dirt that may be in there.

@ricercar: Mind counting the lines going to each connector? For the smaller one it's 8, for the bigger one it could be 18 or 19, hard to see that properly on this low-res image.

[ricercar]

my matrix seems to be 19×8. Unfortunately the matrix doesn't spotlight any "Oh! THIS here is the wire that's broken!" type obvious failure. For any group of 8 keys on one wire, where there's a failed key, there's also a key that works. This is why I leapt to the conclusion I need to do a general cleaning of the Mylar.

Yet additional examination of key matrix articles on the web suggests a thing. If my keyboard matrix is really 38×4 instead of 19×8, then there is a perfect correlation between dysfunctional keys and electrical traces. Real life physical-style testing now required.

[talis]

Wow, what a poor way to connect the cables to the controller board.  I guess its all to save a few pennies worth of connectors.

If my keyboard matrix is really 38×4 instead of 19×8, then there is a perfect correlation between dysfunctional keys and electrical traces.

Count the number of connections to the controller board.  a 19x8 matrix would have 27 connections, while a 38x4 should have 42.

[keyb_gr]

In the light of recent discussion, of course it could also be a spill problem. One should be able to see that on the membrane though.

It does seem that in case of contact problems at the controller board interface, differences between individual keys are highlighted. On my parts board, some keys do make contact when pushed a little harder.

We have a total of 27 connections, so it has to be 19x8. Lowpoly's sawed-up G80 apparently was 20x8. I guess with 3 blocks of keys being merged into one in the 1800s, they could do away with (or didn't have any more space for) one column.

[Bollwerk]

I'm recovering my old G81-3000 OEM. MY VERY FIRST KEYBOARD EVER!!!
(Actually there were two of them, but one got attacked by cola and the other one by me^^)

I reinstalled a Switch successfully and reconnected the PCB. It works. So, I just have to be pacient.

[JBert]

What kind of keycaps does it have? Dyed or two-shot?

[Bollwerk]

Frankly, I don't know. What characteristics do they have?

[JBert]

You can find out which kind it uses by looking at the model id.

I have two G81s, one with molded lettering and one with dyed caps.
The plastic of the dyed caps seems to have some strange feel, somehow they have more friction. Combined with the somewhat wobbly key switch, they give me an awkward feeling, a little bit like sliding your finger down a balloon.

I was wondering if someone else noticed it as well.

[keyb_gr]

The type of lettering is also encoded in the model number. S... would mean dye sublimation lettering, while a H... suffix indicates two-shot molding. Both have high contrast (compared to lasered) and are generally wear-proof, but dye sublimated lettering tends to look a touch fuzzy. Two-shot molded keys are easy to identify by looking underneath, they have characteristic black stripes.

BTW: Mind elaborating on what problem(s) the board has or had? Sounds like you had to fix some things.

[Bollwerk]

They feel strange. No kidding. The baloon-discription fits.

Hm. They look like the Caps on the pictures in this thread, I think.

[keyb_gr]

OK, some help:


G81-3000 lasered


G81-3000 dye sublimated


G80-3000 two-shot

[Bollwerk]

I thing it's dye-sublimated.