Saltar al contenido principal

Repair information for non-Apple computer keyboards.

Preguntas 157 Ver todo

How to stop "short" between 2 keys

Hello !

Once upon a time, a friend gave me this keyboard, that was not working properly since day 1, and he got it replaced but they did not asked the failing keyboard back.

So I'd like to repair it to give it to a friend with a lame keyboard.

This is a mecanical keyboard, here is all the references I can give you :

Block Image

There is some kind of short circuit between key " 3 " and " 4 ", meaning that when I type either of theses, both fires.

So far I tried to de-solder both and re-solder them and checked with a continuity test that there was a indeed short between the keys.

There was no visual problem with the cherry MX switches that I re-soldered and no visual problems on the other side of the board either. But I may be wrong (?).

Here is a picture of the logicboard centered on keys " 3 " and " 4 ".

Block Image

Here is a close up shot of the traces of the PCB on failing keys :

Block Image

And there is a close up shot of the traces of the PCB on normal keys :

Block Image

I have solder and electrical tools but I'm "short" ( ;) ) on ideas on how to fix it.

Thank you for your time !

Edit 07/03/17 :

I tried to de-solder the diode you asked me to test and I'm having difficulties to pry it out. It feels like I would need to heat it more but I am afraid I might kill the diode in the process. Still I can confirm that it's soldered on a pad and not in a " hole " since I can observe some other places where there is no key and where there's still a key slot with all the pads and mounting holes.

How to you feel about it ?

Also the more I think about it, the more I wonder why we should do that.

The only anomaly we found yet, never seen on other keys is on the 2 big pads on this picture :

Block Image

Contesta esta pregunta Yo también tengo este problema

Es esta una buena pregunta?

Puntuación 0


Hi @whitelight ,

Have you tested the diodes (D20, D21) to make sure that one of them is not s/c?

- de

I just did that test :

- No conductivity between the 2 top sides ( black ring ) of D20 and D21

- conductivity between the 2 bottom sides of D20 and D21

I then tested others properly functioning keys :

- No conductivity between the 2 top sides

- No conductivity between the 2 bottom sides

We can then deduct that there is a symptom here.

Can I provide any further details that may help you ?

- de

Hi @whitelight ,

Cannot see your last comment made a few minutes ago.

Nothing is showing up.

try it again

- de

Agregar un comentario

1 Respuesta

Respuesta Más Útil

Hi @whitelight ,

Prove that there is conductivity in one direction only through the diodes i.e. top to bottom in each diode. If none reverse the meter leads and test again as diodes only conduct in one direction.

Compare good ones as you did before to make sure it is the same.

I'd carefully remove the diodes as the tracks seem to go under them both and check for solder bridges or some other connection between the 2 diodes at the bottom end.

Check if there is still a connection between 3 and 4 with the diodes removed.

Update (03/07/2018)

Hi @whitelight ,

What is it that is highlighted in the green circle?

Is it a solder splash across two tracks?

Block Image

(click on image to enlarge for better viewing)

Fue útil esta respuesta?

Puntuación 2


( reposting, my answer disapered through edits )

( edit : actualy it got flagged and deleted for some reason ?? if a mod comes around )

I did some new test and it seems that I made some mistakes with my tests, maybe some clear coat on the board prevented the continuity, maybe i'm dumb. :p

Anyway, here is what I can tell you now :

There is continuity by pair between keys

ig : 1-2 , 3-4, 7-8, 9-0

You might notice there is no 5-6 in the list, and that is indeed the only pair of num keys that have no continuity. I scratched the solder with sandpaper to double check it.

Can that have something to do with my 3-4 short ?

About the other side of the PCB :

To get acces to this side, I would need to de-solder every single key to remove the top shell of the keyboard.

Is there some test you would like me to perform before attempting this ?

- de

Hi @whitelight ,

Hold off the desoldering forthe moment until all possible options have been tried.

Just wondering if there is continuity between 4 - 5 and 6 - 7. If there is it may have something to do with how they matrix the keys for ID location by the controller, i.e grouping the grid arrangements to reduce the number of wires to the controller Not sure about this though.

Did you test the diodes at all as suggested. Also what happens if you remove just 1 diode say d20 what happens. Perhaps you may get more problems as you may have opened a main pathway but this is still easier to try than removing zillions of keys. ;-)

it's late here so good luck catch up tomorrow maybe Cheers

- de

Also, I did continuity test through the diodes, both sides, and none of the diodes on the PCB are positive

But they are positive by pair on the bottom part.

Checked your questions :

We have continuity between 3 (D20 , 4 ( D21 ) and 5 ( D22 )

We don't have continuity between 6 ( D23 ) and 7 ( D24 ), but 6 is positive with other keys such as F8 ( D9 ), F9 ( D10 ), minus and plus symbols ( D28 a,d D29 )

Key 7 has same type of matching.

( all tested on bottom of diodes )

Thanks a lot ! See you tomorrow :p

- de

" @jayeff : Is it a solder splash across two tracks? "

No there is nothing, that might have been some dust or something while I was taking the shot

- de

Hi @whitelight ,

What is on the other side of the board of those two pads that are 'connected'? (as shown by the last picture that you sent with the red arrows.

This side looks like the 'through hole' track connection solder pads of a component lead coming from the other side of the board.

- de

Mostrar 3 comentarios más

Agregar un comentario

Añadir tu respuesta

White estará eternamente agradecido.
Ver Estadísticas:

Ultimas 24 horas: 0

Ultimos 7 días: 2

Ultimos 30 días: 14

Todo El Tiempo: 628