title: DIno keyboard
author: @Magic (slack)
description: A completely custom keyboard!
created_at: 21.05.2025
total time: 34 hours

Day 1 (19.05.2025) ~1hour

Created a folder for the project, and added symbols & footprints for both xiao and raspi microcontrollers. Watched a couple videos on how to make a key matrix, and not too long later, i started making the matrix

Day 2 (21.05.2025) ~5hours

Today i finished the key matrix, and researched on how demanding would led matrixes be. due to gpio limitations, i switched to the raspi pico (originally used seeed rp2040 dip) Afterward i researched a bit on how i can limit the number if gpio pins needed to create a led matrix. however i found no satisfactory one, so i just decided on a screen. But lcd are ugly, i hate the tint, slow speed, and small viewing angles. I found e-ink screens, and decided on using that one. i added it onto the scematics, alongside a handful of led.

After that i placed most of the keys, only leaving the top row (known as the f keys). i left the eink display and potentiometer out of the pcb (it will be in an alternate position). I also will add another pico, since the first one has to little gpio pins. (im writing this paragraph the day after)

Day 3 (22.05.25) ~3 hours

Now i have routed and placed everything, it was a big pain in the ass. The routing is very ugly, but it works, and its all that matters. I have also added the extra pico to the pcb. finally the pcb is finished.

Day 4 (23.05.25) ~3 hours

After making a almost complete 3d model for it, i realised the kays are slightly tilted, so i went back and deleted it, and remade it.

Day 5 (24.05.25) ~5 hours

Today i made the keyboard in fusion, creating the case, adding the top layer, pasting the pcb, and adding the rest of the parts. I also added 2x rotary encoders, 1x slider, and changed from the eink screen to the ssd1306. The reason? easier coding & wiring and alright contrast. I also decided on a slightly different design, not the classic 75%, but with more keys, with massively different positioning.

Day 6 (25.05.25) ~5 hours

Made an attempt at using qmk configurator, and python. both ended in a loss. Now im setting up QMK, and am reading the docs, and watching videos and reading over templates, while making a draft of my own.

Day 7 (30.06.25) ~8 hours

F me im still doing qmk. Today i coded like 5 version, all of which failed. At first i even created a nicely drawn keyboard, and made 4 layers, but noooo. Got errors till i gave up, and stated deleting problematic things. In times of desperation, i turned to reddit, and after 200 views, im yet to get a response. Anyway, after 9 hours of debugging, sorry, trying to debug i decided on polishing the cad instead. So while waiting for my answer from reddit or #code, i smoothed out the edges, and fixed tons of small 0.1 mm errors. Note: i will continue working on the firmware, once i make the keyboard, since my matrix is pretty messed up, so even if qmk compiles, im not sure if it'd work correctly. this is the most simple, cut down version of QMK i now have, and it still doesnt work. And scratch the b10k, i don't have the correct pins necessary.

Day 8 (02.07.25) ~4 hours

At last, after help from some fellow redditors, i managed to get it to work. I also made the animation for the ssd1306, hich is a running dino, from the chrome easter egg.