A Keyboard in Twenty-Four Tweets

13 Nov 2021

Following on from my macropad practice run, I commenced on my first proper build — a Corne Classic 40% true split keybaord. I chronicled the highs and lows of the process in real time on a Twitter thread which I've reproduced here for posterity.


6:23 PM Sep 12, 2021 — So it begins...

9:27 PM Sep 18, 2021 — Progress: all LEDs (key and underglow) soldered. Once I've done the diodes and controller I'll find out if they actually work.

1:39 PM Sep 22, 2021 — Some lunch break soldering and the diodes are done — through-hole in this kit, so a lot more straightforward.

12:55 PM Sep 25, 2021 — Headers, TRRS Jack and reset button — no shorting pins with tweezers for me! Ready to add the controller and OLED and see if it all works.

1:32 PM Sep 25, 2021 — Success! And failure! The controller and OLED are working well, and all the keys register correctly. Unfortunately, none of the LEDs light up. This could be a firmware thing, but more likely I messed up the first LED in the sequence (which was also the first one I soldered).

1:36 PM Sep 25, 2021

<Googles "how to desolder SMD LED" while wishing he'd thought to do them in reverse order>

3:38 PM Sep 25, 2021 — It *was* firmware! I just needed to set RGBLIGHT_ENABLE (...and link-time optimisation so that the resulting binary would fit into the 32K RAM the controller) but it looks like it my shoddy soldering is off the hook. Looks like it's time to properly read up on QMK.

6:13 PM Sep 25, 2021 — Some more firmware tweaking and the per-key LEDs are in Tastefully Understated mode. Looks like I have one broken or badly soldered one (top of the fourth column) — will redo it tomorrow and see if that fixes everything downstream as well.

1:33 PM Sep 26, 2021 — Fixed — the connections looked OK as far as I could tell, but I suspect that I damaged the LED itself by applying too much heat. Fortunately, the kit came with a few spares, and once I'd replaced it everything is working as it should. Bonus: I can now use a solder sucker.

1:45 PM Sep 26, 2021 — Next step is to get the other board (the left hand bit) assembled to the same point — once both halves are working I'll move on to the switches.

12:20 PM Oct 2, 2021 — Bit of a setback on the second board. One duff LED again (the second this time), but it really didn't want to come off and I've made a mess of the pads in the process. I *hope* there's enough left to make a good connection with the replacement; what do you think my chances are?

7:54 PM Oct 2, 2021 — Having done a bit of research, a fix may be beyond my abilities, at least at present. There are no convenient vias to wire to, and I don't want to hack away at the top layer too much for fear of damaging the very close and as-yet-working circuitry for the keys themselves.

7:56 PM Oct 2, 2021 — One option may be to add jumper wires to the other LEDs in the sequence, or maybe even skip that one entirely and modify the firmware to account for the altered sequence.

7:59 PM Oct 2, 2021 — (As a side note, it turns out I'd misunderstood the sequence for the under lighting LEDs — they go by row rather than column. Which means the LED may have actually been fine, and there was no need to remove it in the first place. 🤦‍♂️)

8:46 PM Oct 2, 2021 — This is going to end up with me writing Tcl, isn't it?

4:36 PM Oct 10, 2021 — The damaged pads (and the previous LED, which was also not working) successfully bypassed; sequence passes to the per-key LEDs, which are the ones I might actually use. Still some work to do there, but progress.

6:16 PM Oct 16, 2021 — It turned out that only one of the key LEDs was a problem; replacing that fixed the sequence. Flushed with success (and not having to conserve spare LEDs) I fit a new one on the empty-but-intact pads for the under lighting. That worked too.

6:16 PM Oct 16, 2021 — So, I decided to go for broke, and use my newfound familiarity with the board to wire in my sole remaining LED into the sequence directly, anchored to the last undamaged pad...

6:16 PM Oct 16, 2021 — ...and it worked! IT WORKED! I am disproportionately happy with this.

6:17 PM Oct 16, 2021 — So I now have the full complement of LEDs, and both sides working together. Next step, adding the switches. Not trivial, but certainly feels like I'm on the home stretch.

8:49 PM Oct 19, 2021 — Doing a mechanical bit while cooking tea.

9:23 PM Oct 19, 2021 — Switches done ( not yet soldered).

4:34 PM Oct 23, 2021 — Took a couple of rounds of testing and resoldering, but all the switches are now connected and working consistently. With the addition of key caps (just some simple blank DSAs from eBay), and it's complete:

4:36 PM Oct 23, 2021 — Now all I need to do is figure out a layout, and learn to type on it. And turn off those damn lights.

This site is maintained by me, Rob Hague. The opinions here are my own, and not those of my employer or anyone else. You can mail me at rob@rho.org.uk, and I'm robhague on Twitter. The site has a full-text RSS feed if you're so inclined.

Body text is set in Georgia or the nearest equivalent. Headings and other non-body text is set in Cooper Hewitt Light. The latter is © 2014 Cooper Hewitt Smithsonian Design Museum, and used under the SIL Open Font License.

All content © Rob Hague 2002-2019, except where otherwise noted.