kbt pure right shift stopped being recognized

[nevin]

looking for a little help. i have a kbt pure (original, not pro) and today the right shift stopped being recognized.

now, i've pulled the pcb out of the case and the switch is completing the circuit (continuity test with meter) but is not being recognized on the computer.

at first i thought it was not making the connection when pressed, thinking it was something internal in the switch, but it is by the above test, so i'm not sure what's up.

this is the first i've ever pulled the board apart and it does not look like any traces are missing/scratched through, etc..

i checked it on another computer & it does the same thing so it doesn't look like it's something software related on the computer end.

any help/suggestions would be greatly appreciated.

thank you all in advance, this is a great community.

[ohaimark]

If none of the other keys in the row/column are having issues, I doubt it's a trace issue.

Are the solder joints solid?

It's pointing to a controller issue right now. Can you check to see if it's outputting a code that isn't assigned to anything? If not, you'll need to take more drastic measures.

[nevin]

If none of the other keys in the row/column are having issues, I doubt it's a trace issue.

- no, it's only that key

Are the solder joints solid?

- yes, all look good

It's pointing to a controller issue right now. Can you check to see if it's outputting a code that isn't assigned to anything? If not, you'll need to take more drastic measures.

- using Many Trick's Key Codes app (i'm on a mac) to view the actual output of the keyboard i don't see anything when i press the right shift key. using it as the right shift or turning on the "number cluster", it has no output either way. other keys register as they should.
- can the controller go bad or the info on the controller get corrupt? never heard such issues with something you can't program or that doesn't have a reset switch or jumper....

i guess i'll try to open the switch, but LED will have to come out first.

any suggestions before i do that?

thanks again for your quick reply

[ohaimark]

Those solder joints look quite healthy.

Try measuring the resistance of the switch vs. the other switches in the row when it is pressed. If the resistance is wonky you could be looking at a simple switch teardown or replacement, as it could be the contacts or something.

[nevin]

Oooh, i didn't even think about that... good point. Even if it is making a connection, if the reading is way off it could be throwing off the reading/scan.

Ordered some parts last evening for a numberpad i want to do a switch swap on so i'll have some extra parts if needed.

Thanks again.

[Wodan]

Could be broken diode as well!

[nevin]

Did a quick test, switch is reading the same as others (aprox .2ohm when pressed, all tests done while keyboard is unplugged) Guess i'll have to wait until my parts come & try some cleaning/swapping...

Diode looks in good shape and is acting as the others when tested with meter. Correct me if i'm wrong, but don't the diodes just help with multiple key presses (ghosting)? When trying to get the key to work i am only pressing that one key. Not that the diode couldn't be broken, but it just shouldn't affect my troubleshooting.

[Matt_]

Can you reflow the switch solder joints? Sometimes they look good on the outside but have problems (cold joints). Heat them, and add a bit of solder (the basic hobbyist rosin core solder which contains flux, so that the solder flow well).

[nevin]

yeah, i can try that.
thanks

[nevin]

swapped the switch & still not being recognized. shorting pins/pads on pcb gets no response either. guessing it has to be something in the firmware at this point. being a non-programmable board, i don't know if there's a way to fix it, if that's where the problem lies.

any suggestions?

[nevin]

ended up replacing PCB with a GH60 board i picked up from techkeys. problem solved, and now fully programmable. yay!

[Wodan]

Nice move, congrats