Guess what, a custom ergo!

[Sagii]

Yo, guys! It's been a looong time since I've posted or done anything actively on here, but damn I've missed this place )

So, I got an old poker 2 with ergo clears where I may or may not have fried its PCB while desoldering the switches. Anywhoo, I plan on salvaging the switches, and then make my own custom ergo keyboard! I'm lucky enough that I have access to both 3D printers and laser cutters (only Ultimaker 2, and a laser cutter that mainly takes wood and acrylic, but good enough!). I plan on using a Teensy 2.0, some laser cut acrylic and wood, and my old ergo clears.

The only real requirement I have at this point is that I don't want it to be too big. Wether it's gonna be a solid piece, two pieces, tented, not tented, and all that, has yet to be decided. As my daily driver I use an HHKB, and I find 60% to be more than enough to cover all my needs. In other words, going as far down as to 40%, layout wise, is a possibility, if that would help me save some space (duh). As it is atm I'm considering using an atreus layout as base, and work off of that. The only real kink I got with the atreus is that I'd like to utilise my thumbs a lil' more (AKA possibly more thumbkeys).

Right now I'm in the phase of designing my layout, and was wondering if there's anything in particular I should be thinking of when making it? I'm new to the whole ergo scene, and was wondering if there are any newbie mistakes I definitely should avoid? Or just general advice, for that matter!

Cheers, Sagii


EDIT: At second thought, this might be more fitting in the ergo sub? Lemme know if it is!

[jeffgran]

My advice is to make plenty of prototypes and try them out with real switches and keycaps to see how they feel. I can't tell you how many things I thought I would like but actually didn't, after trying it out.

If you're going 60% style, and especially if you're going 40% style, think very hard about your soft layout in addition to the physical layout. It may partially depend on what you do in your daily keyboard usage. For example, if you have your Fn keys on a separate layer, and you often need to hit the "Alt-F4" combo, make sure it's actually going to be possible/practical/comfortable to hold down the Fn-layer key and the Alt key at the same time! Keyboard shortcuts are the biggest issue. Things like Ctrl-+, alt-tab, alt-`, etc. can be a problem.

If your board is going to be flat, it's very difficult if not impossible to get more than 4 thumb keys per hand that are comfortable/practical/easy to hit. Surf around the projects that have been done before on this forum and you'll see many attempts, but few if any successes. The Keyboardio board went through 20 or so prototypes and came out with a board that has 4 keys per thumb (and.... a palm key...?), so you could steal...ahem take inspiration from something like that.

Good luck. Have perseverance. Things will go wrong but if you keep at it you can get it done.

[Sagii]

My advice is to make plenty of prototypes and try them out with real switches and keycaps to see how they feel. I can't tell you how many things I thought I would like but actually didn't, after trying it out.

If you're going 60% style, and especially if you're going 40% style, think very hard about your soft layout in addition to the physical layout. It may partially depend on what you do in your daily keyboard usage. For example, if you have your Fn keys on a separate layer, and you often need to hit the "Alt-F4" combo, make sure it's actually going to be possible/practical/comfortable to hold down the Fn-layer key and the Alt key at the same time! Keyboard shortcuts are the biggest issue. Things like Ctrl-+, alt-tab, alt-`, etc. can be a problem.

If your board is going to be flat, it's very difficult if not impossible to get more than 4 thumb keys per hand that are comfortable/practical/easy to hit. Surf around the projects that have been done before on this forum and you'll see many attempts, but few if any successes. The Keyboardio board went through 20 or so prototypes and came out with a board that has 4 keys per thumb (and.... a palm key...?), so you could steal...ahem take inspiration from something like that.

Good luck. Have perseverance. Things will go wrong but if you keep at it you can get it done.

That's some very good advice! I'll definitely do as you say. And I believe the tips of making bunch o' prototypes will be very beneficial. I've already tried with the ergodox layout, and figured out that wasn't something for me. Regarding soft layout, thanks for the tip, hadn't really thought too much about that, but you have a very good point

[Oobly]

I spent a lot of time planning and thinking about my design, both from a physical and character layout perspective.

In the end I still had to protoype and test the thumb clusters a lot. I ended up changing their physical design a number of times until I eventually had the current setup (and it's still not quite perfect, I think I'll have to switch to lower profile switches to get the thumb positions exactly right). I also had to adjust the stagger for the pinkie, by a large amount. It really is a lot shorter than your other fingers and I've seen a few builds where this needed altering afterwards. The ErgoDox has very little stagger for the pinkie and I think this is one of its bigger negatives.

Don't have too many pinkie keys and don't add keys in positions you need to stretch for, but rather go for layers that you can access with your thumb clusters (I broke this rule for my Esc and Del, but I had other reasons for including them).

4 keys per thumb is the perfect number IMHO. Especially if your physical layout allows you to press them in combinations easily. This was a crucial driving factor in my thumb cluster design. It allows any pair except diagonals to be pressed simultaneously (front 2, back 2, top 2, bottom 2, all four at the same time) and that creates a lot of options for the "soft" layout. Taking the load off pinkies by using thumbs is probably one of the biggest benefits of my Ergo board for me. I consider Shift to be a layer key, so adding another layer key in line with it allows me to access 3 layers with just 2 keys and one "finger" in one position, just changing angle when pressing: Shift, Layer, Shift+Layer.

I went for dedicated arrow keys and I'm glad I did. After using 60% boards without them I really miss having them.

Adjustable tenting and split angle are good for prototyping, but both can be fixed for the final build. Going split introduces a number of complications, not least of which is the method of connecting them electronically. Encoding the one side like the ErgoDox does can increase response times and reduces the possible complexity of your firmware, since you're taking cycles to decode the signals from the 2nd half. Split angle is important, but it can be fixed since you can move the board closer / further away to match it to your shoulder width / inward arm angle. Split designs do allow more positioning options, though, so it's a compromise you have to decide yourself.

Also, tenting is not crucial, but if you're going for an ergonomic design, it really does help a lot. It doesn't have to be extreme, I find from 12 degrees or so upwards eases the torsion on the wrists nicely and all the way up to even 60 degrees can be comfortable. IMO the Yogitype takes it too far at 90 degrees.

Most important features IMO:
1. Well designed thumb clusters
2. Vertically staggered columns
2.5 dedicated keys for important functions. I have arrow and edit keys, but this is a personal one. I also put Esc and Del into the extreme top corners since I have some very ingrained muscle memory for those and it's nice to have them there. It makes for a nice Ctrl-Alt-Del gesture with my right thumb and pinkie only   My arrows actually become part of the layered number pad, too.
3. Tenting and separation angle

[henz]

Looking forward to see your results

[Sagii]

Thanks for good suggestions and tips Oobly, I'm definitely gonna take it all into consideration! A good thumb cluster, vertically staggered and dedicated keys for my most important functions are all things I've been thinking about a lot. But what exactly do you think of an ergo that is not tented and/or separated (read, atreus)? One of the things I'd like is to be able to move it around a lot. Not taking it on the train move, but around the house move.

With a separated and tented keyboard I'm imagining that'll be a bit cumbersome. Got any input on that? Considering you got a split one yourself
I should also mention that my goal with this build is something that can become my daily driver, and substitute my HHKB. I first considering making a new 60%, but I figured since I'm gonna make it from scratch I might as well make an ergo. As it is atm, making a "fully" ergo that is "perfect" ain't too much of a requirement to me. Something that is just better than a plain keyboard is good enough.

Maybe it's dumb going in with that mindset, should I rather go head first into this build with the plan of making a fully ergonomic keyboard? ._.

So many options!

[technomancy]

But what exactly do you think of an ergo that is not tented and/or separated (read, atreus)? One of the things I'd like is to be able to move it around a lot. Not taking it on the train move, but around the house move.

Build two! Start with a one-piece more portable one (because it's a lot simpler) that you can take around the house. Use the experience you gain from that to build a tented, separated one that is optimized for stays-on-a-desk use.

That's what I would do if I hadn't started with an Ergodox anyway.

[Sagii]

But what exactly do you think of an ergo that is not tented and/or separated (read, atreus)? One of the things I'd like is to be able to move it around a lot. Not taking it on the train move, but around the house move.

Build two! Start with a one-piece more portable one (because it's a lot simpler) that you can take around the house. Use the experience you gain from that to build a tented, separated one that is optimized for stays-on-a-desk use.

That's what I would do if I hadn't started with an Ergodox anyway.

That's actually a very good idea..  Saying that, I might go with a slightly modified and bigger design of the Atreus. Adding one or two extra keys for the thumbs. From what I've gathered so far it seems the only real way to know what you want is to simply make it and use it for a while.. One thing is having a prototype layed out on cardboard, another thing is actually getting used to it and using it for a couple of months!

[Zustiur]

From what I've gathered so far it seems the only real way to know what you want is to simply make it and use it for a while.. One thing is having a prototype layed out on cardboard, another thing is actually getting used to it and using it for a couple of months!

This. A thousand times this.
Not only will you discover things that you could not have known at the start but you're requirements will actually change over time anyway.

Start simple. Add complexity one board at a time.

[bocahgundul]

looking forward to it 

[Sagii]

looking forward to it 

Cheers! Stuff won't really start coming together before after Christmas, unfortunately :/ Got exams coming up. After Christmas I'll fire all engines up and start as soon as possible

[Sagii]

Holy hell. Christmas sure was an optimistic timeframe.

Finally updating this thread. Been bogged down with uni (and not knowing how I wanted to make the keyboard) until this summer. I've now finally made the keyboard I've also lost most of my progress photos in the process (I'm still trying to find them).

After countless prototypes out of cardboard I finally settled on a layout that was pretty much copy-paste from the Atreus, with the small exception of moving the pinky column further down, and the innermost column slightly up. I also took the two halves and moved them slightly further from each other, so that the position when typing felt a little more comfortable to me. Other than that I added three holes in the middle, for indication LEDs (different layouts etc.). Not decided entirely I want to use them for yet.

I settled on simple 4mm plywood, which I then coated with a very nice and dark wood finisher (don't know what it's called tbh). The thickness seemed appropriate because that's what I had at hand, and it doesn't flex much, due to the relative short width of the keyboard. The switches are clears with 65g springs, unfortunately not yet lubed (I'll have to order some first). The switches definitely feel "sticky", due to the light springs and no lube, causing them to not always spring back as quick as I'd like them to.

The controller's a teensy 2.0 running TMK, and for now running almost the standard layout technomancy has on his site.
What's left for this project is making a case, which will be made out of 6mm acrylic, with a plywood bottom. Also considering putting some LEDs inside the case, so that it shines through the acrylic, but haven't decided yet. I also want to make the acrylic stick out about 2-3mm from under the plate itself, so that I can make a lid to put over. That too will be out of plywood, and maybe some acrylic.

All in all, I'm extremely happy with the outcome, and genuinely surprised at how good it actually looks! (If I can say so, myself)
I'd like to thank technomancy for making such an awesome and good looking keyboard! Really, this is the most aesthetically pleasing keyboard I've seen to date.

Here's the album so far (again, I have to find out where the heck I put all the pictures):

http://imgur.com/a/rKTNi


EDIT: Found the pictures! At least some of them.
MODS: Is it possible to change the title to something like "A hand wired Atreus, with minor tweaks". Really don't feel like it's fair to call it a custom ergo without mentioning Atreus anymore.