geekhack Projects => Making Stuff Together! => Topic started by: KnowArt on Tue, 14 September 2021, 03:34:13

Title: Split 30% ergo as first build a bad idea?
Post by: KnowArt on Tue, 14 September 2021, 03:34:13
Hi there,

I'm very very new to this and totally lost in the rabbit hole that is keyboards. Now I decided I want to build something that I think is called a split 30%
This is what I came up with: (

26 keys for letters and 6 for space, backspace and 4 layer actuators.
layer 0: lowercase
layer 1: uppercase. Actuated by sticky shift
layer 2: punctuation and things like windows keys. Actuated by another sticky shift-like key
layer 3: numbers and related characters. Actuated by a key that's sticky till the next space, not just the next character (halp, how do I do that?)
layer 4: key combo layer and macros (as I plan on using a custom layout, but want crtl+z and such to stay somewhat similar). Active when held.

This layout would let me move my fingers minimally and not require too many extra keystrokes, as everything is reachable within two non-stretched-strokes.
They idea behind the key placement is that sideways motion is the devil. Except for the pinky finger, as it is opposed to the thumb. And the key on the inside of the index finger is there because getting rid of two letters in the alphabet is too purist for society to handle... (but honestly, who should get rid of Q and C) I plan on putting the nerd letters there. And at the pinky of course as it's weak and small.

If possible I would love it to be wireless, but I don't know if that's a dumb idea for my first build.

I have no idea of how I should go about all this.
One is making the actual keyboard. I saw that designing and ordering a custom PCB is an option?! Seems like a contender. Where would I do that?
Otherwise maybe laser-cut wood/acrylic and some wires?

And then there's also the key-layout. Should I just make a variation upon Workman for example, or should I go from the ground up as the design is so different? How should I do that? There are so many variables and I'm no prodigy with AI or even just coding in general.

I'm dying for your feedback and ideas.

Kind regards,
Title: Re: Split 30% ergo as first build a bad idea?
Post by: fpazos on Tue, 14 September 2021, 03:59:32
Hi. You could use keyboard-layout-editor for the layout. In my opinion. If you try to do a handwire for a new keeb is the most complex project for a newbie.
You could start building a corne keyboard that is almost the same that you want to do.
Title: Re: Split 30% ergo as first build a bad idea?
Post by: Findecanor on Tue, 14 September 2021, 05:32:06
Welcome to Geekhack!

If I were you, I would buy kit parts for a Corne keyboard (also known as "HelioDox")  and build it, just leaving some key positions out to get a layout like in the image you posted.
Then build my own case and/or plate part to cover up the empty key positions.

These stores in the EU should have parts for Corne (but not always in stock at the same time):
... and there are more stores internationally.

Corne's PCB layout is free and on Github (
There are also many other keyboards with similarities, such as the Kyria.
Title: Re: Split 30% ergo as first build a bad idea?
Post by: nevin on Tue, 14 September 2021, 08:07:13
yeah, for a 1st board, i'd suggest some form of a kit, even if it's just a set of pcb's

check these two sites, large lists of split keyboard projects. ....your layout may already exist.

or, if your really set on doing a one off, there's a handful of tools that will make some of it a little easier.
- definitely start with keyboard layout editor (print out your layouts and see what you think)
- you can then take the KLE and dump it into a plate builder and have a plate cut from your KLE
- you may want a fancier case, but start with an inexpensive spaced plate/skeleton style build (or 3d print if you have a printer)
- then it's wiring & firmware time
- most common controller setup for these type boards is two promicro's (or similar, there are many special adaptations based on the promicro like the postage board, etc...)

here's a handful of links to help (lots of good info)
KLE & plate builders

Some good handwiring articles:

Qmk docs
Title: Re: Split 30% ergo as first build a bad idea?
Post by: _rubik on Tue, 14 September 2021, 10:49:33
Wow -- yes this is a very ambitious first project. Others have listed helpful resources, but I have to ask... do you have any experience with pcb design, electrical engineering, or working with micro-controllers? What's your background?

Maybe start with a fun 40%(and smaller) split ergo kits if not?

Either way, very intrigued to see where you take this. I'll be sitting front row if you deliver :thumb:
Title: Re: Split 30% ergo as first build a bad idea?
Post by: KnowArt on Tue, 14 September 2021, 15:41:06
Thanks for the fast, kind and helpful replies. I was pleasantly surprised!

The recommendations were really good and a think I'll settle for the elehpant42 DIY soldering kit, as it has the exact keys I want + just a couple more which I can easily cover up within a self-made case (or just cut off?) if I really want to.


Now there's just the key layout left to crack. And potentially a wireless upgrade?

When building a case, is it important to have a precision cut layer supporting the switches, or can I just do whatever, as long as it supports the pcb and covers it if I ought that necessary? It has hotswap thingymebobs if that matters.

As one of you asked: my background is in music and animation. So not officially technical, but I can do some rudimentary soldering, coding and 3d software. I often call myself a digilliterate in disguise. I can act as if I know, but it's surface level. Luckily I do have two brothers in tech who can sometimes help me out. My maker-skills (as far as they go) are mostly woodworking, which I hope to use for the case. I'd like to create something that can 'tent', but that seems kinda tricky.
Title: Re: Split 30% ergo as first build a bad idea?
Post by: nevin on Tue, 14 September 2021, 16:45:26
the git page has what looks like a couple different case files (plates)
as well as a pretty detailed build guide

for wireless look at the bluemicro's or some nicenano's

here's a larger list as well as a link to the zmk repo (the wireless QMK)

starting point would be a plate, you'll need one anyway if it's hotswap. then you can add a bit of extra material to either do stacked acrylic or something or make a 3d printed bottom case.
- yes, the hot swaps should be supported from underneath so you don't pull them off the board when inserting switches.

and here's an interesting read for alternative layouts
you can really go down the rabbit hole when digging into alternative layouts...

here's the other one i was looking for...

and another focusing on home row modifiers
Title: Re: Split 30% ergo as first build a bad idea?
Post by: _rubik on Tue, 14 September 2021, 17:15:34
think I'll settle for the elehpant42 DIY soldering kit

That's the exact kit I was going to recommend! I've heard lots of great reviews. I might suggest using it in its stock form before chopping it up (just to get a sense of what you do and don't like). Either way, good look with any modifications!
Title: Re: Split 30% ergo as first build a bad idea?
Post by: jamster on Wed, 15 September 2021, 11:05:52
From that photo, not sure if you've already sourced yet. Otherwise also check out for inspiration or parts. And consider getting an off the shelf Atreus (not split, but tiny and well built).