geekhack Community > Keyboard Keycaps
Making exotic wood artisan keycaps
failKeys:
I hope this is the right place to put this. I want to make a little thread about my process making wood and corian and other types of keycaps. More than just "heres some caps, pay me!". I want to detail the process and update on progress. Thanks for looking.
---
So about 6-8 months ago I got the idea to make some keycaps from wood. I have a CNC mill that I built from scratch and got underway.
Some initial designs...
Although, the first thing I discovered was that any and every 3d model reference you could find online of a keycap, be it cherry, oem or otherwise is completely wrong. Not usable at all. No stem offset, bad shape, heights, etc.
So I wound up needing to design my own "cherry" profile.
The next issue was the stem. Most wood keycaps, (and many artisan in general), use some sort of cat epoxy resin stem. Oof. Not ideal because they tend not to hold their shapoe and stay snug.. I decided i needed something a lot better. i tried 3d printed them in nylon.
That went poorly as nylon is super flexible. Gummy stems! Hana. I tried 3d printed stainless steel. That was also a fail for the opposite reason. Zero flex and no real ability to fine tune the fit. The tolerances on the prints were quite bad. Finally i just decided to machine them. I got some really tiny bits and milled them from delrin (POM). These turned out great. Super snug and dont lose their shape even after removing and reinstalling 100 times.
On to the caps then. I did a number of trials and tests, and eventually ahd to give up on my cherry profile caps for now, because it was difficult to find a way to fixture them securely.
I switched to a square sided keycap design to get something going. We will revisit cherry later, but for now, I could do r4 escape caps and low profile style caps.
In the mean time though, we need some graphics. There are not really many choices for graphics on wood. You can use a decal, which will wear off in short order. You can use laser to engrave some thing but it will tend not to be very readable except on light woods. Last, you can inlay. I chose to engrave pockets and then inlay epoxy paste to create a double shot (and triple shot) style cap in wood.
With that sorted, I could start on "production". It is not too hard to make 1 keycap. It is another thing to try and make 50 efficiently and cost effectively. I made a variety of fixtures for holding the caps, and some for sanding them, and one for settng in then stems.
I could now set about making a bundle of caps in different woods and inlay designs.
I should add in here, gluing delrin stems was complicated. Delrin doesn't really like to stick to anything. In the end 3M sent me a care package of things that managed to let me glue the delrin to the wood.
Next on the list was finishing. Inevitably anything you put on these will wear off, so I wanted something the user could easily touch up and repair themselves. I wound up sanding them to 600-1500 grit (depends on the wood) and putting osmo polyx oil on them. This is a wax oil that acts as a barrier to moisture, and a tiny $25 can will last 1000s of keycaps. Heavy users could touch up the keys once or twice a year and keep them looking brand new indefinitely.
Last on the to do list was packaging. When you only need a handful of boxes, it is surprisingly expensive to package a keycap. Little printer boxes with foam cost as much as $15 each, and off the shelf jewelery boxes while cheap, increase shipping costs $5-$15 due to the size. i decided i would just make my own packaging. Oof. Nightmare, although in the end, i think it was worth the effort. I decided to make blister packs. Went through a number of tests and iterations for vacuum forming and printing.
That's it for now. I will keep this thread going with updates on new designs, cherry profiles, different key positions (spacebars etc) and new processes. Please ask any questions you like about the process. If you like the caps, there are a handful in my etsy store at the moment.
failKeys:
Doh. Looks like i botched all the image links... please stand by :)
Axiom_:
--- Quote from: failKeys on Mon, 20 November 2023, 11:00:57 ---I hope this is the right place to put this. I want to make a little thread about my process making wood and corian and other types of keycaps. More than just "heres some caps, pay me!". I want to detail the process and update on progress. Thanks for looking.
--- End quote ---
This is nice and well documented. Keep the posts coming.
Sometimes we make things for no other reason than to satisfy our desire to create.
The trial and error behind the product's development is what makes it interesting.
--- Quote from: failKeys on Mon, 20 November 2023, 11:00:57 ---Although, the first thing I discovered was that any and every 3d model reference you could find online of a keycap, be it cherry, oem or otherwise is completely wrong. Not usable at all. No stem offset, bad shape, heights, etc.
--- End quote ---
I too encountered the same frustration and had to resort to 3D scanning to get the right measurements.
--- Quote from: failKeys on Mon, 20 November 2023, 11:00:57 ---
On to the caps then. I did a number of trials and tests, and eventually ahd to give up on my cherry profile caps for now, because it was difficult to find a way to fixture them securely.
--- End quote ---
Have you considered retaining some sort of "stem" for the keycaps to aid work holding? The machining will need to be carried out in 3 steps.
1) Rough out the bottom cavity and external profile while retaining a "stem"
2) Turn over and machine the top surface
3) Turn over again, using a vacuum fixture or soft fixture for work holding and remove the "stem"
The reasoning behind this is that wood works best in compression and you'll need something much thicker than the typical mx stem for secure work holding.
Due to the number of contours on a cherry keycap, trying to machine it on a 3 axis CNC will require many bit changes.
Otherwise, it's next to impossible to get nice and smooth shapes efficiently unless you're using a 5 axis machine IMO.
failKeys:
Thanks
The contours needs to be done with a ball nose really. the sides of a cherry or OEM keycap are not a plane (except r4), so you are kinda stuck. My machine is precise and very fast, so this is not a big deal to surface... if the keycap stays put. I have entertained the vacuum fixture idea, but for a few reasons this is problematic. 1 being there is you only get about 10lbs of holding force, and 2, if there is any change in the wood shape or deviation of the surface, it wont sit flat. Did also look at adding an extension to the cap to allow something to grip, but removing it is problematic (vac comment). There are 2 things I see done with metal caps. One is to leave a square ledge on the bottom of 1.6mm for a vise to grip but I don't like that look and it is a bit of a weak area to grip. Another is to thread the stem, but we cant do that with wood. I did have one idea to mill the sides from the bottom with a side cutter. Undercut the part. Then you flip it over and grab it with a dovetail vise jaw. Downside here is you need to spit the keys off individually, and make a jaw set for each different keycap.
I will revisit this in a few weeks. Fixing my CNC on Tuesday (servo died) and ill be making a number of new caps inlays in different woods ands corian and a case.
Axiom_:
--- Quote from: failKeys on Sat, 25 November 2023, 03:45:45 ---Thanks
The contours needs to be done with a ball nose really. the sides of a cherry or OEM keycap are not a plane (except r4), so you are kinda stuck. My machine is precise and very fast, so this is not a big deal to surface... if the keycap stays put. I have entertained the vacuum fixture idea, but for a few reasons this is problematic. 1 being there is you only get about 10lbs of holding force, and 2, if there is any change in the wood shape or deviation of the surface, it wont sit flat. Did also look at adding an extension to the cap to allow something to grip, but removing it is problematic (vac comment). There are 2 things I see done with metal caps. One is to leave a square ledge on the bottom of 1.6mm for a vise to grip but I don't like that look and it is a bit of a weak area to grip. Another is to thread the stem, but we cant do that with wood. I did have one idea to mill the sides from the bottom with a side cutter. Undercut the part. Then you flip it over and grab it with a dovetail vise jaw. Downside here is you need to spit the keys off individually, and make a jaw set for each different keycap.
I will revisit this in a few weeks. Fixing my CNC on Tuesday (servo died) and ill be making a number of new caps inlays in different woods ands corian and a case.
Show Image
--- End quote ---
Wood is certainly a tricky material to work with when you're machining small items like keycaps. The approaches used to machine metal keycaps seldom apply since wood does not have the same structural integrity as metal (especially if you're looking at features as thin as 1.6mm), its likely to chip and break if speeds are too high. You'll definitely need to lower your feeds and speeds for the intricate parts.
You mentioned having to make different jaws for each keycap, does that mean you're considering making keycaps for the different rows? Most manufacturers just stick to the function row for simplicity's sake.
For machinability, the timber I would recommend is Jelutong (more accessible in my part of the world). It has few pores and imperfections, you can stain it if you don't like the look and it has good dimensional stability when exposed to moisture.
Corian is definitely worth investigating. It machines nicely, is isotropic (unlike wood) and offers more finishing options.
Navigation
[0] Message Index
[#] Next page
Go to full version