Ergodox SW2:13 Not actuating

[Magas]

Hello! I'm having issues with getting a specific key to register on this ergodox that I'm working on for a customer. It was sent to me with Alps switches that I desoldered and then resoldered back in some Holy Pandas with the new switch plates. I made the cardinal sin of not testing the lead before I resoldered the board together as I didn't get a trrs cable until today as the customer didn't send one over with it.

What can I do to make sure that this switch works? The other thing is.. Why if this switch isn't working is SW3:13 and SW1:13 still working? How can one switch in the middle of the line not function? Where could the short be?


Any and all assistance is Greatly appreciated! Ergodox PCB.pdf (249.01 kB - downloaded 146 times.)

 Ergodox PCB.pdf (249.01 kB - downloaded 146 times.)

[nevin]

check continuity on the switch, make sure it's not the switch.
then, if switch is good, check around the switch, switch leg to common on column (above/below), switch leg to diode. if one of them is out, make a short jumper (wire) to fix the broken trace.

should be pretty straight forward. feel free to ask for more help/explanation if this doesn't make enough sense.

[Magas]

Doesn't matter what switch I put in there it doesn't work. I can understand the travel pattern to a degree but I'm no professional in that regard (yet). Where exactly am I looking to bridge the travel from?

The switches I've used worked just fine in every other spot. I'm right there with you in regards to the broken trace being the main possible culprit I just don't know where I would jump to.

I will desolder the switch again tonight and pull out my tweezers to see if I can't get a spot to trip the switch.

Any and all help would be greatly appreciated and i would very much like to learn more.

[Magas]

Also... What's the best guage wire to work with in regards to this?

[nevin]

wire size.... anything will do, doesn't have to be anything substantial. use the smallest you have that you can work with.

see attached pic. check continuity between these first. all the green should be the same and all the purple/pink should be the same.  if everything is good there, check the red, you can see the trace on the board connecting the other end of the diode to the column.
- if it's something in the green is broken, run a jumper between the green pins.
- if it's something in the purple is broken, run a jumper between the purple pins.
- if it's something in the red is broken, run a jumper between the red pins.

[Magas]

I tried bridging that connection a million different ways to no avail. I even swapped out a diode to test and see if that was the issue and that also led me nowhere... for some reason it's just this switch that's completely inoperable and I have NO idea as to why. Do you happen to have any more ideas Kind stranger? I will be honest I have no ideas how the red pins work. So I wasn't exactly sure how I should have been connecting those.

I'm just so confused on this.

[nevin]

do you have a multimeter?
does it have a continuity tester?
usually indicated by a music note or similar.
this will help you find where the break is without soldering anything until you find the break.
(do a google search for "multimeter with continuity")

what i was saying was

- any of the pads within the green line should all be the same and should ring/sound when you test between any pads within the green line.
   - could jump between two pads in the green outline say if you lifted a pad or broke a trace when removing a switch

- any of the pads within the pink/purple line should all be the same and should ring/sound when you test between any pads within the pink/purple line.
   - could jump between two pads in the pink/purple outline say if you lifted a pad or broke a trace when removing a switch

- any of the pads within the redline should all be the same and should ring/sound when you test between any pads within the red line.
   - could jump between two pads in the red outline say if the column trace is broken.



if you didn't lift a pad, and the break is further away from the switch legs, it might be within the red outline (column) or it might be and is more probable that it's in the row trace, since it's the last switch in the row. i can't see the row traces because they are actually on the other side of the pcb and i can't see the switches to the left because they are not included in the photos you provided. i tried to look at other pics online but i didn't find one with the date that's on your board or they were white pcbs and hard to see the traces.

can i see a pic of the whole pcb? this same side is fine (backside, not switch side)
- in the pics you sent, i can't get an orientation. if it's in the bottom of the left side or close to the top of the right half....

did you change the (reflash) firmware/keymap when you changed switches?
was it working before the switches were changed?

[Magas]

I tested all of the highlighted locations with a continuity test on a multimeter and they all checked out just fine. Nothing wrong at all..

Any other ideas?

I'm still at a loss.

[suicidal_orange]

Either the green or pink section needs to connect to the bottom of the diode, the figure 8 of pads below the S in the SW2:13 label (I'm tired and can't work out which!) and the other end of the diode needs to be connected to the red section.  Looks like there could be pad damage on the red end from removing a through hole diode...

If all are connected are you sure the firmware is right?  An unassigned key appears not to work.

Also as Nevin said pics of the back of both boards would be good as that shows both sides.

[nevin]

ok. if all others checked out, then it's probably the row then.
i couldn't see the row traces because they are on the other side of the board.

you can see by the schematic of the board below how everything is wired and flows.


test these two pins for continuity. it should ring/make noise. if it doesn't, then put a jumper here.

[Magas]

That was it!!!!


Thank you so much! That didn't trace!

Oh man finally!!! Thank you so much!

[nevin]

no problem. glad you got it fixed. sorry i didn't suggest that portion sooner. i was not familiar with the schematic of that board and didn't want to suggest something that was wrong. took me a bit to find the info i was looking for without having other photos.

[Magas]

That's actually why I included the actual PDF of the PCB and everything it was an attachment down at the bottom of the original post but that might not have had all the diagrams needed necessarily. By the way you guys were a big big help and I super appreciate it! My customer will too!

[nevin]

Sorry, must have missed that

[Magas]

No need to be sorry at all. It works flawlessly now!

THANK YOU AGAIN SO MUCH FOR EVERYTHING!

[nevin]

no problem.