1/20/2026 7 PM - Set-Up and Installed Libraries

Set-up the project and installed all the required KiCad libraries. I also made a bit of a plan for what I will add to the project.

I had a bit of trouble installing the libraries, although managed to work it out eventually :D

1/20/2026 8:20 PM - Added the key matrix

I finished adding the matrix for the keys! I've included diodes and switches for half the keyboard (the other half will automatically duplicate from it)

1/20/2026 8:27 PM - Added XIAO circuitry

I've now added the main microcontroller and battery circuitry! I'm using the XIAO nRF52840, which is a good small and wireless microcontroller, with enough inputs for the keys matrix.

1/20/2026 9 PM - Tidied Up Schematic

I mostly just tied everything up, fixed a few issues and added some mounting points and made everything look nicer! Next up I'll assign footprints and start designing the PCB :D

1/20/2026 11 PM - Assigned Footprints

I've spent this time doing some research to decide which parts to use, and assigned footprints to everything. I had a bit of an issue trying to install the marbastlib footprint library, which took ~20-30mins to sort out, but i got it sorted in the end!

1/21/2026 11 AM - Designed the PCB

I've now designed and routed the PCB! I spent a little while deciding on and optimising the layout, and then I also routed everything. It also took me a little while to get the board outline I wanted. Next I will be deciding what to add to make mine a little unique!

1/21/2026 8 PM - Added LEDs

This took way too long... 😭

Basically I added a bunch of RGB backlight LEDs to :

But to do that I had to fix up some power stuff to both protect the LiPo and regulate a 5V output for the LEDs to run happily. So trying to get my head around the whole buck-boost convertor thing was kinda slow... But i got there in the end!

1/22/2026 2 PM - Layout and Re-Routed PCB

So after adding the LEDs and BMS and stuff, I have now redone the layout (most stuff stayed the same, I just had to shuffle the BMS and 5V regulator in), and then re-routed accordingly.

I've put a GND pour on both sides with stitching vias to help with current management and EMI.

^^ Some 3d views of the PCB

So now all I have left to do is work out how to mirror the PCB...

1/22/2026 9 PM - Layout Left Side

Ok so now I've been working on laying out the opposite side of the keyboard. This time I knew what I was doing, although I did have to be careful to make sure the layout matched up with the side I already designed.

I also begun routing some of the critical power areas, like the BMS and 5V boost converter.

1/22/2026 11 PM - Finished Routing

I've now finished routing the entire PCB!!! I tried to do similar things to the other half, although tried a new power distribution method, which made routing a lot easier.

Here are some 3d renders of the PCB:

Just need to do a couple of last checks, and then I'll be ready to ship!

1/23/2026 11 AM - DRC Fixes and Panelisation

I just did a quick fix of everything based on the DRC checker, then spent a little too long trying to make the mousebites work, cause I couldn't make the board outline 'close' properly. Turns out there's a tool called 'heal shapes' which fixes everything automatically!

1/23/2026 9 PM - Worked on Sourcing

I spent this time working on finding suppliers for the parts. I've worked quite hard to find other potential sources which might be cheaper, and think I've done decently.

I still need to compare LCSC with other options like mouser though:

1/24/2026 1 PM - Switched to Nice!Nano

Redesigned the board to be compatible with the Nice!Nano, which is substantially cheaper than the XIAO. It unfortunately did make the board a bit bigger, but all in all its not too bad.

1/24/2026 9 PM - Optimised BOM

So, having revised the PCB a bit, I redid the BOM, with as many optimisations as I could, to bring it down by ~$35 to just under $100! Next up is working on the firmware!

1/25/2026 2 PM - Created ZMK Shield

i've been researhcing ZMK (mostly from their docs), and used that information to create a custom 'shield' for my keyboard. I haven't gone too detailed into config, but focused on overall setting up keymaps, etc. THen when I've physically made the PCB, I can add extra features and test!

Unfortunately the build seems to be failing rn, so my next step is to do some troubleshooting.

1/25/2026 4 PM - Fixed Firmware

I finally managed to fix the build issues - it was mostly an issue with the keymap definition. While I was at it, I also made some significant improvements to the keymap, and fixed some other config. Next up is the case!!

1/25/2026 5 PM - Add 3D Models

As a quick preparation for making the case, I added in some 3d models of the Nice!Nano and its headers, which should make modelling the case a bit easier. I also manually combined it with some headers for better positioning.

1/25/2026 11 PM - Added Keycaps

So I kinda got a little bit sidetracked here... But basically I spent a little while adding the switches into the sockets, and then choosing some nice looking keycaps for the board! I also did a little render (not the best quality ik), cause why not.

1/26/2026 3 PM - Designed Case

I've now finished designing the case!!! (which is basically the last key part of this project). I designed the two halves so they are almost identical to mirror, so I'll only show one half here:

I'll tidy up the repo, and then be ready to submit.

1/26/2026 5 PM - Wrote README

Basically, I just wrote up a README for the project, with a couple renders and screenshots.

I tried to detail each part of the project, as well as include a BOM.

4/6/2026 3 PM - Soldered the Left Diodes

I got started on the soldering, after some gear I bought finally arrived.

I hadn't done any surface mount soldering before, so I began with some simple diodes. As you can see, my first attempt was a bit wonky, and had a bit too much solder.

But at least my multimeter showed that it had good joints.

After some more practice, my later joints were significantly better!

Here's the board with all the diodes:

4/6/2026 5:13 PM - Soldered the other Passives

Next, I soldered the rest of the passives on the left-side:

I managed to improve, with better control of solder flow, getting some nicely rounded joints:

4/6/2026 5:58 PM - Soldered ICs

This took me a while to manage, but I've soldered up the charge IC circuitry. The ICs are really small, and I found that I needed a magnifying glass to even see the pins, which took some getting used to.

I then also had a bridge, which took some time to fix, using braided solder wick, and resoldering it.

In the end, they came out pretty well though:

I also soldered a capacitor while i was at it:

4/6/2026 8 PM - Soldered Power Circuit

This one took a while...

I soldered up the power circuit on the front!

The inductor proved quite annoying to deal with, cause it doesn't have leads, so was hard to get soldered well. And then, I ended up putting it the wrong way around 😭

So after some kapton tape and painful desoldering, I got it off, so I could put it back on!! And then, I realised I had scratched the board, and had a bridge between two pads....

Fortunately, though, they were connected to the same net, so it should be ok!!

4/7/2026 8 AM - Researched Print Settings

Pretty quick one here, but I just did a little bit of research around the best print settings for the keycaps. It seems that a 45degree orientation, with very low layer height is the best, along with super slow perimeter speeds.

4/7/2026 9 AM - Soldered Left LEDs

Now, I've soldered up all the LEDs on the left hand side! I think I did a pretty decent job with them, and the practice from the other components has definitely helped.

The fact that they countersink into the cutouts in the board makes alignment so much easier as well.

4/7/2026 10 AM - Soldered nice!nano

Now, I soldered the nice!nano onto the left side of the keyboard. This was pretty simple, although I found (after soldering them on), that the debug pins were in the wrong place on my aliexpress clone board.

Fortunately, these pins used in the design, so I could desolder them and ignore the disparity in the design, which only took a few extra minutes.

4/7/2026 12 PM - Added the battery & troubleshooted

We have liftoff!! After soldering up the battery and connectors (via a socket terminal, so it can be disconnected), and fixing a few solder joints, I managed to get it to turn on, and appear on the computer!!

Next step, is fixing up the firmware, and flashing!

4/7/2026 4 PM - Bodged.

So, soon after getting it working, I found that the battery was not working. At all. The status LEDs were doing weird flashes that suggested some sort of fault, meanwhile the battery seemed to work perfectly fine when using a different charger.

After a ton of cross checking, testing, and digging up schematics and datasheets, I discovered that the aliexpress nice!nano clone I had was in fact using a different charger IC, which meant that the status LED didn't follow the patterns of the genuine nice!nano documentation.

So, hopeful, I continued, with the expectation that all would be ok. However, I then discovered (through some fiddly multimeter probing), that this charger chip didn't play nice with the BMS chip I included, because it seemed to handle all this itself. That meant that a bodge was needed, to bypass the chip.

Looking at the PCB, I identified some large vias, which could be abused for this purpose:

I then scratched up the ground pour around this area, to make a makeshift pad:

Which I fairly successfully soldered to.

(As you can see, I was pretty happy when my testing showed that it had worked!)
Video of it working!!!

4/7/2026 10 PM - Soldered the keysockets

With the bodge successful, I next soldered up all the hot-swap sockets. These were fairly easy, as once again, they countersunk into the PCB, so I could align them really quickly!

4/9/2026 12 PM - Soldered Right Power Circuit

With the left side done, I turned to soldering up the right hand side. I decided that to make testing easier, down the track, the first thing I would solder, would be the power circuitry on the front side.

I soldered it fairly easily, although as always the inductor proved troublesome, due to its huge thermal mass.

After soldering, I hooked it up to a power supply, and tested that the 5V output was stable, which it was!

4/9/2026 2 PM - Soldered Lipo BMS

So I soldered up the BMS on the right hand side of the keyboard, and, as before, the mistake in the PCB design meant that a bodge was necessary. This time, knowing it in advance, I did some quick calculations, and realised that my enamelled copper wire would take it. This made bodging much easier, cause it is smaller and stiffer.

Here's the successful bodge:

4/9/2026 3:15 PM - Soldered Power leads

Quick one this time - basically I just soldered on the two power leads, and added a dab of glue as strain relief, to stop them breaking of the pads. It's a bit messy, but it works ig.

4/9/2026 3:39 PM - Soldered Diodes

Again, quite a quick one (i really am getting faster lol). But I soldered up the diodes no the right side of the keyboard. It all went pretty straightforward-ly:

4/9/2026 4:16 PM - Soldered Capacitors

Next up was the right side capacitors, which again went fairly smoothly. I had some minor trouble with the big electrolytic capacitor, specifically getting the solder to flow and wet under the very small legs, to give it strength.

4/9/2026 4:39 PM - LEDs!!

After the capacitors, I've soldered the LEDs up! These were actually really quite quick to do, thanks their countersunk design, and fairly large pads.

4/10/2026 7 PM - Soldered keysockets

Hopefully my final soldering!! I've soldered up the hotswap sockets on the right hand side. They were fairly easy to do (only two pads per each), although they did take a little time to heat up, cause of the large chunk of metal that the pad connects to.

4/10/2026 8 PM - Tested & Debugged

With the soldering out of the way, I inserted all of the switches into the sockets, and loaded on the firmware. At first, only the left side would work. However, after turning to the ZMK docs, it turned out that a reset image needed to be used, to remove all config from the board so they would pair. I did this, and....

...it worked!!

However, I found that a few keys weren't working. After probing with my multimeter, I discovered that a couple of diodes weren't working (presumably burnt), and a whole row wasn't working (because I forgot to solder a pin on the controller 🤦). Fortunately, a few joints and new diodes later, everything was working!!!

4/10/2026 10 PM - Improved Case & Sliced

Now that I've got the firmware working, I turned to the case. Whilst I had already designed this, certain assembly matters (like bodges and battery placement) meant that I had to redesign certain aspects. I also optimised it to remove around 10g of unnecessary supports.

Once I had done this for both sides, I imported it to PrusaSlicer, and optimised the print settings.

4/11/2026 5 PM - Assembled Case

Basically I just put the case on. I had to pull all the switches out, and put them back in to make the plate fit properly, which was somewhat tedious. I also had to trim all the header pins and fix up an alignment issue with the battery terminals.

Apart from that it was all fine!

(the elastic bands are just to make sure it doesn't fall apart before I properly secure it.

4/11/2026 7 PM - Improved firmware

So while I waited for my keycaps to print out, I decided to make some improvements to the firmware. The biggest change I made was improving the keymaps, and adding another layer, for navigation arrows, etc.

I also fixed up the battery indicators, which hadn't been working. Basically, the default configuration had involved the wrong ADC pins. This took way too long... 😭 (Turns out some weird internal ZMK error was meaning that the config was being overwritten.)

However, worked in the end!

4/13/2026 - Added Keycaps!

So, after my keycaps had finished printing

I removed all their supports (yes, there were a lot. and yes, this was tedious...)

And then cleaned up their edges, where they had been stuck to a brim. This was challenging, as the brim left a nasty bit on the edges it separated from, although fortunately these edges are primarily out of sight when the keyboard is assembled, so it was okay that I didn't do a perfect job with this.

Finally, I put the keycaps onto the switches (the fun part!), and tbh the keyboard is looking really cool!

4/14/2026 - NEW DEMO VIDEO LINK

Because of the under 16s social media ban in Australia, my prior video was taken down, so I've reuploaded it to GitHub instead (which is allowed). The link is https://github.com/MadAvidCoder/Split-Keyboard/releases/tag/Build_Demo

(I was advised to use a GitHub link instead in #blueprint-support here)