Author Topic: Broken PCB Column, Need Help Diagnosing  (Read 4627 times)

0 Members and 1 Guest are viewing this topic.

Offline cono

  • Thread Starter
  • Posts: 2
Broken PCB Column, Need Help Diagnosing
« on: Wed, 24 June 2020, 21:46:32 »
SOLVED! Thanks to those who helped out. Turned out to be a short from an LED pin to an unused pad from some leftover solder residue.

Hey everyone!

So I've been working on building my very first custom keyboard over the past few days, including lubing and modding switches/stabs. I'm building a Tofu60 with a DZ60 and Gateron Black Inks, lubed with Trib 3204. I put together the board the other night and had everything soldered and all of the switches were working properly. But I put in a few keycaps and noticed that the spacebar felt mushy. After taking the board out of the case I noticed the stab wire had fallen out.. no problem, I'll just have to desolder a few switches and get it back in place with tweezers. Which I was able to do.

However, in the process I must have broken a trace or lifted a pad because after putting the switches back in I realized that an entire column of the pcb has stopped working. Now.. I'm still really new to all of this so I'm learning as I go but I've done some testing and need some help to move forward.

So, the switches I desoldered to fix the stab were the spacebar "c" "v" "b" "n" "m" and ",". After soldering the switches back on and testing them, an entire column wasn't working. This column is made up of the following keys; "6" "t" "g" "b" and spacebar. You can see in the DZ60 circuit diagram below that those keys are all in the same column in the keyboard matrix. So it makes sense that they could all break at once.

245725-0

Next I wanted to find out what exactly isn't working for each switch, and if I could just create bridges to nearby switches in order to make it work. First I'll fill you in on how the pcb is laid out. In the image below I have labeled which switches are the ones that aren't working. As well as their pins; the pins labeled A are the vertical pins responsible for calling which column is being triggered. The pins labeled B are the horizontal pins responsible for triggering a particular row on the keyboard.

245727-1
245729-2

Through my testing of the pins by shorting them with other pins on the board, I found a few things.

1) All of the B pins are completely functional and can be used with all of the other A pins on the board (other than the A pins in this column)
2) All of the A pins don't seem to be sending any signal at all EXCEPT for the spacebar (read the next point for more about the spacebar)
3) The spacebars A pin has a strange behaviour when shorted with another columns' A pin. It will trigger that entire column at once. None of the other keys on the keyboard seem to have this behaviour.

So that's it.. I'm not sure how much info this gives you to help me figure this out.. but I'd love any ideas or suggestions. I'd like to note that this ENTIRE column isn't working. This includes the other set of pads directly next to the current spacebar and "b" positions.

I think the issue is with the spacebar or a trace near it since that is the only key to have any effect from it's A pin and it has that strange behaviour. I'll leave you with an image of the entire board and non marked-up photos of the areas of interest. I appreciate any help with this, as I'd love to be able to make my first keyboard work.

Thanks!
« Last Edit: Thu, 25 June 2020, 10:26:19 by cono »

Offline suicidal_orange

  • * Global Moderator
  • Posts: 4771
  • Location: England
Re: Broken PCB Column, Need Help Diagnosing
« Reply #1 on: Thu, 25 June 2020, 03:35:24 »
Welcome to geekhack, and can I commend you on that very helpful 'help!' post :thumb:

Looks like there's a short between the bottom left pin of the RGB LED and some unused Space switch pads.  So first thing to do is clean up that stray solder and see if that fixes it, if not you might need to sacrifice your Caps Lock LED or practice your soldering until you're happy to attach a jumper wire to the tiny pin on the controller chip.

Hopefully I'm wrong about that, I just woke up and have a stupid illness (literally) and can't get my head round the implications of that short.
120/100g linear Zealio R1  
GMK Hyperfuse
'Split everything' perfection  
MX Clear
SA Hack'd by Geeks     
EasyAVR mod

Offline yui

  • Posts: 1082
  • Location: 127.0.0.1 (in azerty)
Re: Broken PCB Column, Need Help Diagnosing
« Reply #2 on: Thu, 25 June 2020, 03:47:00 »
for it to trigger he whole row i would agree with orange that you have a short to a power somewhere and for the lack of triggering on the whole row usually i would expect the trace between the space and B switches to be broken, although i only ever fixed membrane keyboards with no rgb so my experience is low that, if you have a multimeter you could try to test continuity on the closest contacts that you have soldered.
vi vi vi - the roman number of the beast (Plan9 fortune)

Offline cono

  • Thread Starter
  • Posts: 2
Re: Broken PCB Column, Need Help Diagnosing
« Reply #3 on: Thu, 25 June 2020, 10:01:03 »
Hey guys, thanks for your replies! I had noticed that the solder residue next to the LED but figured it wouldn't have any effect, and was a little scared to scrape it off in fear of breaking a trace underneath or something. But.. after about 30 seconds with tweezers it all came off easily in one chunk. AND IT WORKED. Thanks for the suggestion, I now have a completely functional custom mech! I'm so excited, thanks guys!

To anyone else who sees this thread, don't worry we're all good!