Hacker Remix

Show HN: Bayleaf – Building a low-profile wireless split keyboard

722 points by sgraz 5 days ago | 245 comments

Hey HN,

I built a wireless, split, ultra-low profile keyboard from scratch called Bayleaf. As a beginner I learned all things electronics, PCB-building, designing for manufacturing, and many other hardware-related skills to put this together.

This case study dives into the build process and of course the final result, hope you enjoy!

jacobevelyn 5 days ago

This is the keyboard I’ve been hoping Apple would make for years! I’ve currently got a UHK 60 but would probably switch to this if I could buy it. Especially if it had the standard Apple keyboard layout so my fingers don’t need to relearn things like arrow keys and cmd like when I switch between the UHK and MacBook keyboard.

robenkleene 4 days ago

For folks looking for something similar that you can buy https://hellonuio.com/

arcastroe 3 days ago

woah, ty! I've been using the Dygma Defy [1], but I might switch to this!

[1] https://dygma.com/products/dygma-defy

jacobevelyn 4 days ago

Ooh thank you for the tip!

nextos 5 days ago

Me too. Low-profile is nicer to joints in the long run, and also makes me type faster due to low-travel switches. Bonus points if it has a little bit of negative tilt.

In this regard, I found Microsoft Sculpt really good because it ticked all ergonomic boxes but it didn't require relearning. However, quality was a bit subpar, it didn't offer a wired option, and it has been discontinued. The entire lineup was actually sold to Incase, who are releasing it again soon.

jjani 4 days ago

> Low-profile is nicer to joints in the long run

Very much so. I went through 10+ keyboards before finding one that I could type on for a day without aching joints. The only one that passed (which I'm typing this comment on), was the lowest out of all of them. Cheap Aliexpress keyboard, but it's maybe 4mm from table to keyboard. For those in a similar situation that want to try it, it's this one [1]. Definitely not "high quality", but saved me from inability to type.

Question for the keyboard-knowledgeable: it claims in this picture [2] to use scissor keys but the picture of its insides are pretty meaningless to me. Does their claim look to hold up?

[1] https://www.aliexpress.com/item/1005006265944802.html

[2] https://ae01.alicdn.com/kf/Sf4d150eda2c4441490535d5f17ebfbfc...

vladvasiliu 4 days ago

I loved th MS Sculpt (the "75%" one, with a layout similar to a laptop, only split). It started randomly failing for some reason, the only keyboard I've ever had to fail...

But I don't think that being low-profile was what was nice on the joints with that keyboard, but rather the presence of the palm rest which didn't force the wrists to bend upwards.

I now type on a "gaming" keyboard with full-size cherry switches, which is quite thick, and I don't have any wrist pain. Some 12 years ago, I used to have an Apple keyboard which was hell on my wrists, despite being quite low profile. The Apple didn't have a palm rest, the mech one does.

As to your typing speed remark, I suppose you already know that you don't have to bottom out your keys. I find I type faster on my mechanical keyboard than on laptops, and it's also easier on my fingers since there's no shock from bottoming out. It takes a while to get used to not press the keys all the way, but it's a much more pleasant experience. It's also quieter than the random HP domed keyboards we have at work or any laptop.

someothherguyy 4 days ago

> Low-profile is nicer to joints in the long run

I think low profile is worse on joints in the long run. It probably isn't a universal truth.

kalium-xyz 4 days ago

I have had 3 of which 1 has bricked itself.

roflchoppa 4 days ago

I really just want a wireless 60v2. Been debating trying to make a new PCB that uses a nice!nano.

I’m sure the devils in the details.

seemaze 5 days ago

Bravo! You have elevated a honed tool to a truly engaging artifact! I think the large challenge in design is mitigating the breaking point between ruthless efficiency and endearing novelty.

I picked up a Let's Split v2[0] when it came out years ago and never soldered it up.. maybe it's time!

[0]https://shop.beekeeb.com/product/lets-split-v2-keyboard-pcb-...

w10-1 5 days ago

For next iteration, consider integrating trackpads?

Moving to the mouse and back is pain enough that people go all-in on keyboard-only interfaces.

I velcroed a trackpad to the middle of a Kinesis Advantage. Now I use either hand for the mouse, and even stretch a finger or thumb to the pad without leaving the keys. The movement is little different from using the keyboard.

But for split keyboard, you'd need one trackpad for each side, and in dimensions not readily available. hmm.

jazzabeanie 5 days ago

I want a trackpads integrated into the keys. Under the J key a trackpad with low sensitifity, and under the K key a trackpad with high sensitivity.

mystified5016 3 days ago

Have you ever used a J-key trackpoint? I found an absolutley ancient Compaq laptop from the W3.1 era and I found the J-key mouse shockingly pleasant! Well, as pleasant as trackpoint mice get anyway.

We should bring that back, I think I like it more than the thinkpad style

wyre 4 days ago

Brilliant! Use modifier keys with your left hand for either key entry or mouse click?

sgraz 4 days ago

Definitely looking into adding trackpad/trackpoint/touchpad(?) into the V2. And finding a way to do so without sacrificing the minimalist look and feel too much.

JeremyBarbosa 5 days ago

So happy to read this because I don't see it mentioned often enough.

I have a ErgoDox EZ, and I still prefer using my Framework 13 (with Kanata![0]) because having my thumbs navigate the trackpad is so convenient even with a keyboard-driven setup.

[0] https://github.com/jtroo/kanata

yencabulator 4 days ago

I have a touchpad in between the halves of my Ergodox EZ. It's not exactly as easy to reach as a laptop touchpad, but it's worlds better than moving your shoulder to reach a mouse.

mkl 5 days ago

I put a trackball (Kensington Expert Mouse) in between the keyboard halves. With tenting it can nestle in quite close.

danielvaughn 5 days ago

Personal opinion, but I really don’t get low-profile keyboards. I always need a foam cushion for my palms, which means that a normal profile always feels best for me. Low feels too low with a cushion, and yet still feels too high without one.

crazygringo 5 days ago

I know what you mean, but there's also a big difference between key height and keyboard height.

There's also a common misconception that it's ergonomic to angle a keyboard "upward" (elevating the back of the keyboard), when correct ergonomics is actually to angle the keyboard downwards (elevate the front).

See if you have a long object a little shorter than your foam cushion that you can scotch tape to the bottom of the front of a low-profile Apple keyboard, so that you still use the foam cushion but the front of the keyboard is at the same level as the cushion, and then angles downwards.

And then you get the advantages of the short key travel, which just means your fingers move less and so there's less force/strain.

You may find it shockingly comfortable!

luqtas 5 days ago

keyboard upward or downward angulation depends on your wrist support, arm position/chair height... what's optimal and generic to suggest because how our upper extremity muscles are, (at least for mouse but i think it can work for keyboards too) [0] is having it sides angled from 20° up to 30°, subjective exprience also plays a role

you can easily find MX switches that have equal or +- 0.5mm travel lenght as low-profiles ones... which considering how thin PCB (and even handwired) flat keyboards are, i can't see the point unless you are supporting your arms in the same table your keyboard is! i rather have my arms floating so i can use shoulder and elbow movement for distant keys other than wrist movements but who cares :P

[0] https://www.sciencedirect.com/science/article/abs/pii/S00036...

crazygringo 5 days ago

> keyboard upward or downward angulation depends on...

It doesn't. But to clarify: it should always be downwards relative to your forearm. And your forearm should be generally be about horizontal (or a little bit downwards is OK too).

A keyboard that is angled upwards puts constant strain on the top of your wrist. There are no circumstances where this is a good thing (assuming you don't have injury/disability that requires other accommodations).

Also, your link is about mice not keyboards, and about angling mice sideways. It has no relevance to angling keyboards up/down. (But yes, "vertical" mice, that are angled in reality, are much better too.)

jwr 5 days ago

Correct. So many people get this wrong, and there are so many misconceptions around, that there is actual market pressure to deliver upwards-tilting keyboards and wrist pads.

If you want to get your keyboard and desk ergonomics correct, look at pianists. They have it nailed down. Elbows at right angle, no wrist supports, sitting straight. And guess what, the piano keyboards are not upwards-tilting! Unfortunately, most of our desks are too high relative to the chairs we sit in.

luqtas 5 days ago

various resources from universities recommend having your forearm angled between 90-120°, which if it's > 90°, an front angulation doesn't make sense at all... [0]

> Also, your link is about mice not keyboards, and about angling mice sideways

the author points out another research, i should have linked it instead of that one [1]... but it's about the ideal slanted angle considering how our muscles are structured; the author even cites that suggestions as something not taking relative preferences, which is totally fair as some people may use their computers for a short time in very awkward positions and that's fine

[0] https://www.sciencedirect.com/science/article/abs/pii/S10506...

[1] yet about mouse but please take your time to read what's was typed in the paper regarding the slant angle suggestion -- https://www.sciencedirect.com/science/article/abs/pii/S02680...

crazygringo 5 days ago

> recommend having your forearm angled between 90-120°

That's what I said ("your forearm should be generally be about horizontal (or a little bit downwards is OK too)"). Never upward.

> which if it's > 90°, an front angulation doesn't make sense at all.

If you want a 10° downwards angle on the keyboard relative to your forearm and your forearm is 5° upwards, then your keyboard still needs 5° downward relative to your desk. If your desk is so high that your forearms are angled up 20° from the horizontal, then something is extremely wrong with your chair/desk height.

And your first link supports exactly what I wrote:

> The natural position of your hands with respect to the relative vertical position at the wrist is along the plane or below it (i.e. you want the tips of your fingers to be at the same height as your wrist or preferably slightly lower). When your hand rises above this plane (making the hand signal for 'STOP'), this is called dorsiflexion or wrist extension. This greatly reduces blood flow through the wrist and can quickly cause pain, fatigue and numbness. Most keyboards have a 'foot' located at the back of the keyboard which is not desirable as it creates a positively inclined keyboarding surface. Many articulating arms offer the option of a negative inclination, which will make the entire work surface slope away from you, ensuring that your hands are not 'bent' upward at the wrist.

And I don't know why you've added another link about sideways mouse slant. That has nothing to do with up-down keyboard/arm angle. If there's some relevant sentence in it, please quote it, because I can find nothing relevant in the publicly available text.

You ask me to "please take your time to read" -- I suggest you take the time to re-read your own first link. It makes clear there is no situation where a keyboard angled upwards (the back of the keyboard elevated) is good ergonomics -- exactly what I said.

luqtas 5 days ago

you are right, i always considered the front of the keyboard what you considered the back, sorry for this (and by the way, i changed the link to a more generic one citing the variation of 90° -> 120° i typed of one easily finding by surfing the internet)

on the 2° paper, > Among the five tested mice, the 25° or 30° slanted mice caused lower muscle activity and more neutral working postures for ECU, Trap and PT muscles.

these muscles are also used during keyboard usage. slant angle is always suggested on ergonomic research, as our arms muscles are tensioned/twisted when using a flat keyboard/mouse... you can easily find research pointing slant angle on ergonomic keyboards but the research i pointed out is interesting as the author actually studied which is the optimal angle based on our anatomy (there's tension/forces on greater slant angles to maintain the posture, as there aren't surfaces to support our hand) and not on individual preference

crazygringo 5 days ago

No worries! Just trying to save people from wrist pain. :)

And yes -- slanting the keyboard that way is usually called keyboard tenting. Also highly recommended, but difficult to implement in practice because it requires an even lower base for the keyboard, and there aren't many good accessories out there for it, sadly...

I'm still waiting for someone to sell a cheap kit for angling keyboards downwards and implementing tenting. It's kind of complex to do both at once...

yencabulator 4 days ago

Ergodox has a design where the keyboard can sit on 3 adjustable legs, and you can get both a downward angle and tenting out of that.

The downside is that you really need to get your desk way low for your arms to naturally "droop" down over that. I have a motorized sit-stand desk I bought specifically after comparing specs on how low the desk surface can go, and the position that's low enough while seated is basically as low as my knees will let it go. A separate keyboard tray would be great but they rarely have room for the footprint and height of a tented keyboard with a touchpad in the middle.

Standing up, it's a lot easier to find a good height.

https://ergodox-ez.com/

crazygringo 4 days ago

Those legs look amazing.

I want to be able to buy them as an adhesive add-on for any (split) keyboard!

yencabulator 4 days ago

I don't think glue will hold them well enough. There's a metal thread in the keyboard for a reason. The angle is adjustable and tightening the nut locks the foot at that angle.

jjani 4 days ago

I'd love an accessory that makes a keyboard completely flush with the surface it's on. I get this is hard because dimensions are different for each keyboard, but something that e.g. a Macbook fits into would be nice, given its standard size. Everyone's different but for me personally, I feel that would be a big ergonomic benefit. Supported by the fact that my current winner is the one coming closest to this ideal, being extremely low, maybe ~0.4cm from the surface.

Been toying with the idea of getting something 3d-printed, basically something just as high as a Macbook, that you can attach to the front of it, giving a flush surface. Even better would be carving a Macbook-bottom-shaped or external keyboard-shaped "recess" into a table to place it into.

Ever seen or heard of anything like this?

apocalyptic0n3 4 days ago

I've tried numerous keyboards and the conclusion I've come to is that there's low profile and then there's low profile. The bulk of the low profiles I've tried (NuPhy, Keychron K8, the mechanical Logitech, a few others) are definitely low profile compared to the Logitech Pro X TKL I use for my gaming PC. However, they're still tall. Most of the gains are from a shorter switch and keycap, but the body is still quite high.

If you compare it to the Apple Magic Keyboard I'm typing on now – and that seems like a definite inspiration for the Bayleaf – it's a stark contrast. The K3, for example, is more than twice as tall (10.9mm vs 22mm backrow). The Magic Keyboard feels fine to type on without any sort of wrist support and I never feel any strain. But on the K3, even with a support (tried both their wooden support and a similarly sized foam one), I would feel strain after an hour or two.

Most low-profiles are really just a middle ground between the two sides. And, at least in my experience, you get the downsides of both without any of the positives of either.

lsaferite 5 days ago

As a low-profile keyboard zealot, I'm trying to imagine why you'd need a foam pad under your palm. But maybe I'm odd. My entire forearm rests on my desk and supports my arm weight so nothing is pushing down on my hands. Then my hands flow directly over the LP keyboard without any appreciable bend in my wrist.

makeitdouble 5 days ago

My take is that finger muscle are really different from person to person, to a degree that can be surprising.

I might be at the other extreme end, but even typing on glass doesn't bother me much, and laptop keyboards are a good compromise to have just enough travel and not too much.

I tried a nuphy low profile and it was tiring after a while. The thinkpad standalone trackpoint keyboard has been my go to for a while, and tgis keyboard also looks great to me.

hinkley 5 days ago

Squishy keyboards make the tendons on the backs of my hand burn after a while. After Apple stopped producing their wired keyboard, which I used for everything, I bought a bunch of surplus ones and burned through them. I tried a few knockoffs, several of which were close, and now I’m on the wireless ones.

I was surprised as anyone when an Apple keyboard became my gaming keyboard.