victorlucachi / void_ergo

A handwired split keyboard, running QMK firmware
221 stars 14 forks source link

i think i did something wrong #17

Open andrwui opened 1 month ago

andrwui commented 1 month ago

I started assembling the keyboard today, I already assembled the right half, the left one is still not finished.

I plugged the assembled half of the keyboard to the computer, and entered vial, to test everything was ok.

The keyboard worked, but it showed the left keyboard being pressed, which i thought it was normal because the firmware uses split detection, but also the keys were mirrored, so if i pressed the most upright left key, it will show the most upright right key.

I think i messed up soldering the diodes:

For the right half, the diodes are facing < and for the left half diodes are facing > (they are mirrored, facing to center)

Are the keys being mirrored because of this? Or is it that i haven't plugged both halves?

If I did mess up, is there a way through firmware I can tell qmk that diodes are the wrong way? I know I can just mirror the layout in vial, but I just wanted to know if there is a fix firmware-wise.

Also, desoldering is not an option i think.

Thanks for the awesome keyboard!!!

victorlucachi commented 1 month ago

Pictures of the diodes/soldering would help paint a better picture, but as far as I can tell from what you're describing, it shouldn't be an issue (diodes can be installed either row2col or col2row, and you can set their direction in firmware, but they have to be soldered the same way across both halves).

Are you sure that your column wiring to the controller is not mirrored on the right half? I've seen that happen before.

andrwui commented 1 month ago

Diodes for the right half are soldered this way:

image

And the diodes for the left half are soldered this way:

image

The columns are wired to the controller from left to right on both halves, I think I got that right.

Thank you for the answer!

victorlucachi commented 1 month ago

Diodes seem fine, it doesn't matter which way their ends are pointing, electrically speaking the circuit is the same. But yeah, as long as your controller wiring checks out, everything should be good (you could take a look at this wiring diagram for the column order and compare it with yours).

andrwui commented 1 month ago

I was indeed mirroring the columns on the slave...

Now i'm having another problem...

I got them both to correctly work separately. But when i connect the trrs cable to connect them, the slave does not work, i see it's getting powered, but keystrokes do nothing.

The master works just fine, but the slave does not.

I thought it would be a problem with the trrs port wiring, but I don't think it's the case.

For context, here's pictures of the wiring:

Master: image

Slave: image

For a little bit more of context, the firmware i'm flashing the boards with is the vial qmk fork you have linked in the readme, i just disabled the encoders.

victorlucachi commented 1 month ago

disabling the encoders in config and building the firmware from my fork, or just flashing the precompiled vial hex that's uploaded to this repo? please try the latter too, if you haven't already.

FedeArre commented 1 month ago

disabling the encoders in config and building the firmware from my fork, or just flashing the precompiled vial hex that's uploaded to this repo? please try the latter too, if you haven't already.

@andrwui forgot to post the fix, the issue was with the 3.5mm connection not fully inserting because of the 90° input. Trimming a bit of the protective rubber allowed it to fit perfectly, and the slave started working without issues."

victorlucachi commented 1 month ago

that's great news, I'm glad it's working