Author Topic: ManuForm: custom contoured 50% ergo board (Read 109134 times)

jeffgran · « **on:** Sat, 20 July 2013, 18:23:08 »

UPDATE

ManuForm, the Hand-Shaped Keyboard

"Manu-" = "Hand"
"-Form" = "Shape"

This project changed a lot from conception to completion, as I guess most product designs do.

The general design goals are:

- ergonomic keyboard
- 6 keys per thumb
- all 6 thumb keys are ACTUALLY reachable ergonomically
- all keys for all fingers are reachable without having to move the hands/wrists, thus losing their "home" (major culprits on a normal keyboard for this are ESC, backspace, number keys especially 6)
- to achieve the above point we use a 40%/60% style where there are multiple layers triggered by function keys (in this case these are some of thumb keys)

Original goal was to make it foldable/travelable too. That goal was scrapped, mainly because in all my prototyping I couldn't figure out a way to achieve the above goals without making the board contoured and "3D", and that basically makes it impossible to make it foldable too.

This is the final product:

Typing demo:

Key Layout (subject to change )

Note that to get to the FN layer, you just hold down both of the "Symbol layer" keys at the same time.

/* *************** */
/* Layer 0: qwerty */
/* *************** */
/* _________________________________________ _________________________________________ */
/* | | | | | | | | | | | | | | */
/* | SHFT | Q | W | E | R | T | | Y | U | I | O | P | SHFT | */
/* |______|______|______|______|______|______| |______|______|______|______|______|______| */
/* | | | | | | | | | | | | | | */
/* | CTRL | A | S | D | F | G | | H | J | K | L | ; | CMD | */
/* |______|______|______|______|______|______| |______|______|______|______|______|______| */
/* | | | | | | | | | | | | | | */
/* | ALT | Z | X | C | V | B | | N | M | , | . | / | ALT | */
/* |______|______|______|______|______|______|_ _|______|______|______|______|______|______| */
/* | | | | | | */
/* | BSPC | CMD | | CTRL | SPC | */
/* | | | | | | */
/* |______|______|_ _|______|______| */
/* | | | | | | */
/* | TAB | SYM | | SYM | ENT | */
/* |______|_LYR__|_ _|_LYR__|______| */
/* | | | | | | */
/* | ESC | NUM | | AROW | ALT | */
/* |______|_LYR__| |__LYR_|______| */
/* */
/* */
/* */
/* **************** */
/* Layer 1: symbols */
/* **************** */
/* _________________________________________ _________________________________________ */
/* | | | | | | | | | | | | | | */
/* | SHFT | ~ | ( | ) | + | > | | : | & | * | | | | SHFT | */
/* |______|______|______|______|______|______| |______|______|______|______|______|______| */
/* | | | | | | | | | | | | | | */
/* | CTRL | ` | { | } | = | ' | | " | $ | % | ^ | \ | CMD | */
/* |______|______|______|______|______|______| |______|______|______|______|______|______| */
/* | | | | | | | | | | | | | | */
/* | ALT | | [ | ] | - | < | | / | ! | @ | # | ? | ALT | */
/* |______|______|______|______|______|______|_ _|______|______|______|______|______|______| */
/* | | | | | | */
/* | DEL | | | | _ | */
/* | | | | | | */
/* |______|______|_ _|______|______| */
/* | | | | | | */
/* | | FN | | FN | | */
/* |______|_LYR__|_ _|_LYR__|______| */
/* | | | | | | */
/* | | | | | | */
/* |______|______| |______|______| */
/* */
/* */
/* */
/* ************************* */
/* Layer 2: numbers/movement */
/* ************************* */
/* _________________________________________ _________________________________________ */
/* | | | | | | | | | | | | | | */
/* | SHFT | | HOME | UP | END | PGUP | | | 7 | 8 | 9 | | SHFT | */
/* |______|______|______|______|______|______| |______|______|______|______|______|______| */
/* | | | | | | | | | | | | | | */
/* | CTRL | | <- | | -> | ESC | | | 4 | 5 | 6 | | CMD | */
/* |______|______|______|______|______|______| |______|______|______|______|______|______| */
/* | | | | | | | | | | | | | | */
/* | ALT | | | DOWN | | PGDN | | | 1 | 2 | 3 | | ALT | */
/* |______|______|______|______|______|______|_ _|______|______|______|______|______|______| */
/* | | | | | | */
/* | | | | | 0 | */
/* | | | | | | */
/* |______|______|_ _|______|______| */
/* | | | | | | */
/* | | | | | | */
/* |______|______|_ _|______|______| */
/* | | | | | | */
/* | | | | | | */
/* |______|______| |______|______| */
/* */
/* */
/* */
/* ************** */
/* Layer 3: FKeys */
/* ************** */
/* _________________________________________ _________________________________________ */
/* | | | | | | | | | | | | | | */
/* | SHFT | | | | VOLUP| | | | F7 | F8 | F9 | F10 | SHFT | */
/* |______|______|______|______|______|______| |______|______|______|______|______|______| */
/* | | | | | | | | | | | | | | */
/* | CTRL | | | | MUTE | | | | F4 | F5 | F6 | F11 | CMD | */
/* |______|______|______|______|______|______| |______|______|______|______|______|______| */
/* | | | | | | | | | | | | | | */
/* | ALT | | | | VOLDN| | | | F1 | F2 | F3 | F12 | ALT | */
/* |______|______|______|______|______|______|_ _|______|______|______|______|______|______| */
/* | | | | | | */
/* | | | | | | */
/* | | | | | | */
/* |______|______|_ _|______|______| */
/* | | | | | | */
/* | | | | | JGSL | */
/* |______|______|_ _|______|______| */
/* | | | | | | */
/* | | | | | | */
/* |______|______| |______|______| */
/* */
/* */
/* */

Make Your Own

Not sure if anyone will be interested, but I decided to make this project open source.

The one warning I'll give is that the column stagger is pretty aggressive. After making some prototypes with less stagger, I found that, for my hand at least, with my short pinkies, this worked best. If your hands are shaped differently this might not fit the "ManuForm" (Hand-shaped) moniker for you. I wish I had the skills to make the model fully parametric so you could adjust it and print a variation with different column stagger and thumb-length, etc.

Anyway, if your hand is shaped kinda like this, and you like 40%-style (layered) ergonomic keyboards, you might love this board:

So if you want to make one just like mine, here's what you'll need:

Print the Case

The case is also the plate the holds the MX keys. You can print it ("buy" it) at shapeways here: http://www.shapeways.com/product/6BNKXTXTA/manuform-keyboard?key=9d07b99823474e036abe511d064ed4b0. Full disclosure, there is a slight markup to buy that model, just to try to recoup some of the costs (probably close to $1000) of making all the prototypes and stuff that I did. You could get around that by using the original design files and uploading the model to your own shapeways account.

Wire it up

You can see my pictures of how I wired it below. The controller is a Teensy 2.0 with header pins. The board that holds it is this one from ebay: http://www.ebay.com/itm/10-Pcs-DIY-Single-Side-Tinned-Prototyping-PCB-Circuit-Universal-Board-5cmx5cm-/231275000224?ssPageName=ADME:L:OU:US:1120

There are some wiring schematics I made here: http://imgur.com/a/QVU3J

I would have loved to use this instead: https://geekhack.org/index.php?topic=45587.msg955666#msg955666 to house the teensy, but I didn't see that in time before the design was finalized.

Paint it

This is the paint I used. Two (light) coats of the color, and then one of the clear coat to protect it:

Close it up

The case is designed for M3x10mm bolts and nuts.

* Bolts: http://www.grainger.com/product/APPROVED-VENDOR-Skt-Cap-Screw-22TY49?cm_mmc=EMT-_-ShipConfirm-_-GcomShipConfirm-_-sku&RIID=137151846&GID=&mid=ShipConfirmation&orderNumber=&rfe=5169c25bb51177fd6d0b14b8e8e74f029fa7562ceec35793fdcdb3043d428be6&origin=prod
* Nuts: http://www.grainger.com/product/APPROVED-VENDOR-Hex-Nut-22UK74?cm_mmc=EMT-_-ShipConfirm-_-GcomShipConfirm-_-sku&RIID=137151846&GID=&mid=ShipConfirmation&orderNumber=&rfe=5169c25bb51177fd6d0b14b8e8e74f029fa7562ceec35793fdcdb3043d428be6&origin=prod

Firmware

I used @hasu's amazing TMK firmware. My fork (only difference is the addition of my project directory with the setup for my matrix) is here: https://github.com/jeffgran/tmk_keyboard

3D Model Source Files

You'll only need this if you're going to change the shape, but all of the source code and (editable) 3D models are on github, here: https://github.com/jeffgran/ManuForm

--------------------------------------------------------------------------------

Original Post

The idea is a little portable ergo layout with lots of thumb buttons. After several design iterations I'm down to just 56 keys. When in use it is about 6x13 inches, when folded about 6x7 inches.

The big knobs on the bottom allow for a little tenting when you set it up, but would probably be a little annoying when trying to pack this in your bag...

There'd have to be a pin to stick down the underside of the spine, or maybe a latch of some kind, to keep the tent from collapsing under its own weight.

What do you think the chances are I could actually get something like this made?

Of course, it'd probably be a lot easier to do with without the tenting but this was the idea I dreamed up.

Working draft of a layout:

The bottom:

CPTBadAss · « **Reply #1 on:** Sat, 20 July 2013, 18:26:01 »

This reminds of a foldable Catboard.

Pretty cool idea, interested in seeing it completed. What do you mean when you ask "What are my chances of getting it completed"? There's plenty of information on GH as to how you can build a custom keyboard. Are you motivated enough to find the info that you want and actually build it? If so I don't see why it can't be done.

gnubag · « **Reply #2 on:** Sat, 20 July 2013, 18:27:38 »

pretty sure that this is going to be very expensive and probably heavy when made out of metal.
3d printing is not yet very stable enough to be able to thrown into a bag on a daily basis.

you could use magnets to have it proped up and keep it closed.

Photekq · « **Reply #3 on:** Sat, 20 July 2013, 18:31:16 »

A few questions :

-PCB or direct wiring?
-PCB or plate mounting?
-What kinds of materials do you have in mind? Bare in mind that machining is very expensive and metal costs quite a lot too.
-How will you do the folding mechanism? Just simple hinges?
-How will you connect the two sections wire-wise?

CPTBadAss · « **Reply #4 on:** Sat, 20 July 2013, 18:37:07 »

Wonder if this would better it were direct wired. Would be much easier IMO and you could concentrate on the case more.

MOZ · « **Reply #5 on:** Sun, 21 July 2013, 00:10:41 »

For a prototype or one-off, wiring is th best IMHO. If you want mass production, then PCB is the way to go.

I thought of something similar for my upcoming Ergodox, but what is the point of the split then, huh?

Acrylic could work, but it would be a little thicker than a metal case.

Simple hinge should work well for this.

jeffgran · « **Reply #6 on:** Sun, 21 July 2013, 14:13:40 »

Quote from: CPTBadAss on Sat, 20 July 2013, 18:26:01

This reminds of a foldable Catboard.

Pretty cool idea, interested in seeing it completed. What do you mean when you ask "What are my chances of getting it completed"? There's plenty of information on GH as to how you can build a custom keyboard. Are you motivated enough to find the info that you want and actually build it? If so I don't see why it can't be done.

1.Thanks for that link, I hadn't seen that catboard thread before. It does have a simlar shape -- that's a cool personal board he built! So custom -- doubt anyone but him would really want it like that.

2.Yeah that was a poorly worded question. What I mean is: Are there any obvious design flaws or problems on the horizon? Like say, "that would only work if it was X material and that is going to cost you thousands of dollars" or something?

I used to do 3d graphics for a living but I know next to nothing about building things in real life, so any guidance is appreciated.

Quote from: gnubag on Sat, 20 July 2013, 18:27:38

pretty sure that this is going to be very expensive and probably heavy when made out of metal.
3d printing is not yet very stable enough to be able to thrown into a bag on a daily basis.

you could use magnets to have it proped up and keep it closed.

Magnets are a good idea, I hadn't thought of that. I guess you'd have to find that balance between being strong enough to keep it standing but not so strong you can't pull it back apart...

Quote from: Photekq on Sat, 20 July 2013, 18:31:16

A few questions :

-PCB or direct wiring?
-PCB or plate mounting?
-What kinds of materials do you have in mind? Bare in mind that machining is very expensive and metal costs quite a lot too.
-How will you do the folding mechanism? Just simple hinges?
-How will you connect the two sections wire-wise?

-- I am thinking plate-mounted and direct-wired, just because that seems easiest in terms of making custom parts. I just have to design the plate for the switches to "snap in" to, and then wire the back up, right? Am I right that they just sort of snap in?

- I was picturing it as solid milled aluminum ("milled" is what you say when it's made in the CNC machine?) LOL i am such a noob. Aluminum would be lighter than steel, right?

Maybe it could be CNC-milled acrylic? Seems like that oddly shaped bottom would have to be built from a 3d block. the plate and the top rim could be a 2d design.

- Hmm. I'm not sure how that hinge will work. Again with no experience with these product design issues, all I can keep picturing is like a cabinetry hinge that would be screwed on there somehow. If that piece was milled out of metal it could made in hinge shape like a door hinge and I'd just have to get a rod to stick in there as an axle.

- Again, good question. I picture is as like a headphone/telephone cord with the curls that stretch when you open it up like a spring. I need to study how the ergodox did it a little more to figure out how that has to work. I think the main question there is about how many individual leads you've got going across from one side to the other?

Quote from: CPTBadAss on Sat, 20 July 2013, 18:37:07

Wonder if this would better it were direct wired. Would be much easier IMO and you could concentrate on the case more.

Quote from: mohitgarg on Sun, 21 July 2013, 00:10:41

For a prototype or one-off, wiring is th best IMHO. If you want mass production, then PCB is the way to go.

I thought of something similar for my upcoming Ergodox, but what is the point of the split then, huh?

Acrylic could work, but it would be a little thicker than a metal case.

Simple hinge should work well for this.

Thanks for your thoughts. I am leaning towards direct wiring since this will likely be a one-off. I'll open source anyhing I make though in case anyone wants to use it. I've learned so much already from all the stuff people here have built.

Photekq · « **Reply #7 on:** Sun, 21 July 2013, 14:30:41 »

You could have it CNC milled (machined) but as I've said it's really quite expensive. And yes you're correct about platemounting. Good luck, you've got a lot of things to find solutions to if you want this project to happen! It's an ambitous project, but a good idea.

CPTBadAss · « **Reply #8 on:** Sun, 21 July 2013, 14:58:51 »

If you're going to have a cabinetry hinge, my worries are that when it folds in, the switches at the bottom will interfere with the keyboard properly folding. Wonder what the best solution for this would be...

Yes, the plates will allow the switches to snap into it.

Aluminum is lighter than steel, yes. If you're going to make this out of acrylic, I'm wondering if the acrylic is too brittle for this motion. I think you'd want to stick with the aluminum since there will be a lot of folding and unfolding.

MOZ · « **Reply #9 on:** Sun, 21 July 2013, 15:03:55 »

I would suggest a door hinge, not a cabinet hinge

jeffgran · « **Reply #10 on:** Wed, 24 July 2013, 18:14:00 »

I switched to openSCAD to build the CAD models for this. Being a programmer, it appeals to my sensibilites. I can plug in different angles and recompile and see different variations with the different parameters.

I put the code for this on github: https://github.com/jeffgran/ergo56-hardware

Like I said it's open source and my aim was/is to create re-usable pieces (so far: a dcs key with cutout below, a matrix of keys in a key plate, etc.)

Next step: start building out the case design in openSCAD and try to figure out the hinge mechanism, how much space I'll need for the controller & wiring, how to construct the "rim" around the top of the case, etc.

tufty · « **Reply #11 on:** Thu, 25 July 2013, 01:45:42 »

If it were me, I'd take a partial cue from the "stowaway" folding keyboards. Rather than having one hinge, I'd have 2, and a rectangular piece in the middle. This way you can make the whole deal fold inwards; this keeps your keys inside (and protected from damage) when folded. In addition, the rectangular piece can be thick enough to provide your tenting whilst simplifying your keyboard "wings", the additional width will give better hand separation, and leaves you with a handy space for placing your controller (and maybe even cable storage).

Just a thought. Keep at it!

[edit] Ah, you are going for inward folding. my bad. Still, it seems to me that a middle piece will simplify your "wings" enormously.

jeffgran · « **Reply #12 on:** Fri, 26 July 2013, 12:33:00 »

@tufty

Thanks for the tip, I had to do a search -- I had never seen the "stowaway" keyboards you're talking about. Pretty cool! Someone needs to make an ergnomic version and then this custom project would be obsolete.

The two hinge idea is very intriguing. You're right, the middle piece could extend down and provide the tenting and the base, so the "knobs" in my original design on the bottom wouldn't be needed. And better hand separation, as you point out. I'm definitely going to play with that idea!

jeffgran · « **Reply #13 on:** Sat, 16 November 2013, 19:35:31 »

After a long hiatus due to buying a new house, moving, building a new workbench in the new garage, etc., I'm back thinking about this project again.

I decided I need to test out a prototype of the board with real switches and keys, so I ordered a 3D printed plate with a rim around the bottom for it to stand on, from shapeways.

Now just have to wait a few weeks for them to ship it to me.

Here's another pic even though it's basically the same as what I showed before, I think. But without any of the tilt/tent/foldable madness yet:

Can't wait to get it and put some keys on there and pretend to type on it!

bearcat · « **Reply #14 on:** Sat, 16 November 2013, 19:53:28 »

Yahoo, another openScadder! I did my first version on open scad as well, really liked it. I did the next version in openJsCad, actually...

I think your case idea is totally doable! If you're not set on it all being a single piece, then the high shared walls in the middle could be a vertical piece.

If you haven't already shipped your order to shapeways, you might consider getting them laser cut from Ponoko. Probably ~$25 for a set.

Also, check out Upverter for doing the PCB, it's much easier to use than kicad!

Love the design, can't wait to see it!

jeffgran · « **Reply #15 on:** Sat, 30 November 2013, 00:17:11 »

@bearcat: thanks dude. Yeah, I know javascript pretty well and it's obviously a lot more capable of a language than openSCAD is, so I've migrated everything to openjscad. It's sweet being able to tweak the design/parameters and rebuild it at will.

I got the box from shapeways last week and put all the switches in there to try it out. I like it but I definitely want to do another design revision and try it again. Will post pictures and update soon.

I think I'm going to just wire this up with wires instead of a pcb -- it'll skip the PCB design step and it'll allow me to play with different angled surfaces too.

jeffgran · « **Reply #16 on:** Sat, 30 November 2013, 17:27:11 »

Here's some pics of how the prototype turned out. I only had one side printed (left hand), and of course ignore the actual legends on the keycaps. I just stuck on what I had laying around. I like it but it has some problems that I want to address in the next iteration:

1. The top two thumb buttons (F3 and page down in the pic) are too hard to hit, just like the top thumb buttons on ergodox, because of having to reach past the lower row and then press down. I think i want to try to put them on an angle so the thumb sort of "pokes" at them instead of pressing straight down on them. Because they are within easy reach of my thumb but just at the wrong angle.

2. The stagger on the columns is not severe enough for my hands. My forefinger and pinky are like WAY shorter than my middle and ring finger, so that if I put all 4 fingers at a natural curl, they end up resting on a different row. So I'm going to try making the pinky and forefinger columns shifted down about 3/4 of a key unit from the middle two columns, to see how that feels. It'll be specific to my hand, because I know other people's fingers are not all the same length relative to each other, but I guess that's the point here, to make it custom to my preferences. I have a feeling that if someone were ever to make a truly ergonomic keyboard it would HAVE to be custom fit to each person's hands. There's no one-size fits all ergonomic.

I already have done some work on the next design iteration. Will post back when I have something presentable.

bearcat · « **Reply #17 on:** Sun, 08 December 2013, 03:06:04 »

if it makes you feel any better, i came to the same conclusion about needing waaaay more pinkie stagger!

jeffgran · « **Reply #18 on:** Sun, 08 December 2013, 18:06:43 »

@bearcat yeah I just saw a thread from suka at deskthority where he said he had a 1/2 -3/4 key unit stagger on the columns too. Seems like everyone who goes full bore with the staggered columns comes to that realization...

---

So I have v2 of the basic plate/case finished. Below are pictures!

Of note:
- I decided to angle the thumb cluster a la the Kinesis (but more angled) and Oobly's board (mine is maybe less angled), and also do a sort of "double angle" on the top row of thumb keys. You can see in the pictures what I mean hopefully.
- I changed the stagger of the columns to what I think will fit my hands better and allow all my fingers to reach all 3 rows of keys easily without having to move my wrists.
- I also noticed in the first version that the plastic was flexible enough that it bowed in the middle a little bit when you press down on it, so I added some support "rails" on the underneath side to hopefully strengthen it.
- I was trying to design it so that a standard key set could be used to populate it, but it's just too hard. The standard set of keys is just too specific to a "normal" keyboard and doesn't adapt well to other designs. For example there are two 1.5x keys in a normal set but they are different profiles because they are on different rows (delete and tab), and all the 1.25x mods are the heavily angled R4 profile which wouldn't work well to turn sideways and use as a thumb key.

Gonna order just the right hand side this time. If this turns out to be a good layout, I'll have to finish off the design with a bottom piece and a place for the controller board to sit, figure out the hinge mechanism (I think it'll have to fold backwards now, with this shape), figure out a cover for when it's folded up so you can throw it in a backpack, and then onto the electronics. Man, what a project I've gotten myself into!

One nice thing about using openjscad for this is that I have a pretty well parameterized model (needs a little work to remove some remaining hard-coded numbers) so I can tweak things like the stagger amounts, the various angles for the thumb piece, etc. so this could actually be a general design that people could tweak to their specific needs. It would be possible to make a little web UI to play with different settings, but I'm not sure if I'll have the motivation to get that up and running.

bearcat · « **Reply #19 on:** Sun, 08 December 2013, 19:42:34 »

Love it! Crazy steep thumb cluster, too

My hypothesis on everyone under-staggering the pinkie is that it looks like a half a key should be SO MUCH, but as soon as you put your hands on it, you find out it's not... another reason for everyone to prototype

Looking forward to seeing your build!

Puddsy · « **Reply #20 on:** Sun, 08 December 2013, 19:56:22 »

I'd totally use one of these. It'd probably be a little easier to lug around than my pure pro.

kurplop · « **Reply #21 on:** Sun, 08 December 2013, 22:45:29 »

I had prototyped your multi angled thumb cluster idea last summer and initially liked the idea, although the arrangement does require precise placement for it to work best. Eventually I decided to just use taller caps with a sloped face on the back keys instead for several reasons.

-The design does further complicate the construction process, especially if you use PCB's.
-The taller keycap route seem to be a bit more flexible for different sized hands.
-I noticed the thumb action for the bottom thumb keys are a side strike, where the upper key require more of a thrust of the entire hand.

This is not to say that I'm suggesting you abandon the idea. I am just saying I couldn't work through some of the drawbacks I noticed in my design and decided to go in a different direction.

I appreciate those of you who have furthered the Key64/ErgoDox type design. We all seem to be working through many of the same issues and it's helpful and encouraging to me to see your progress. Keep posting!

jeffgran · « **Reply #22 on:** Sun, 08 December 2013, 23:59:32 »

Thanks for the comments, guys.

@kurplop, thanks for sharing your experience with that. The kinesis uses the taller keycaps in back and it does work OK, but I thought this would be similar but an improvement. I guess I'll find out when this next print comes and I try it out.

You're right that it will be a bit more of a "stab" or "poke" motion as opposed to a downward/inward press, but I'm hoping that it won't require wrist movement. We'll see.

BTW, I would like to second your sentiment that it is very inspiring to see what everybody else is/was working on. The key64, Kinesis, the ErgoDox (which, after building it, I thought "hey if I can build this from a kit then I can make a custom one too"), oobly's board, bearcat's board, kurplop's experiments, suka's experiments, have all been my inspiration and I have to count myself lucky that I live in a time where I can see the projects people are working on around the globe in a place like this.

yasuo · « **Reply #23 on:** Mon, 09 December 2013, 02:16:16 »

Jeff,this your project also??
http://deskthority.net/workshop-f7/an-other-diy-keyboard-project-t4560.html

jeffgran · « **Reply #24 on:** Mon, 09 December 2013, 12:50:31 »

Quote from: yasuo on Mon, 09 December 2013, 02:16:16

Jeff,this your project also??
http://deskthority.net/workshop-f7/an-other-diy-keyboard-project-t4560.html

Nope, different Jeff. From the looks of it that guy is crazy.

lkong · « **Reply #25 on:** Wed, 11 December 2013, 21:58:19 »

Is it possible to make two pieces detachable?
Since we are folding backwards, hinges are really not necessary. Instead, why not make some kind of jack and slot to make them snap together.

jeffgran · « **Reply #26 on:** Thu, 12 December 2013, 20:21:01 »

Quote from: lkong on Wed, 11 December 2013, 21:58:19

Is it possible to make two pieces detachable?
Since we are folding backwards, hinges are really not necessary. Instead, why not make some kind of jack and slot to make them snap together.

It's a good question and an interesting idea. The purpose of making it fold in on itself is to make it easy to throw into a bag without worrying about the keys/caps, and to have it take less space. I'm definitely not sure what the final case should look like at this point.

Oobly · « **Reply #27 on:** Fri, 13 December 2013, 02:13:29 »

It's hard to come up with a good way to make this kind of board portable. I'm in the same boat. I've considered making the bases able to clip together so they fit together with the keys outwards and the cables fit between them, but that leaves the keys vulnerable and it's still bulky.

I'm considering a fold-flat design which can be folded "inwards" to protect the caps and key switches, but it's a PITA to design something that can be made without a lot of effort.

For now I just put both halves in a box like this:

kurplop · « **Reply #28 on:** Fri, 13 December 2013, 05:39:15 »

I've also spent a lot of time considering the folding/flattening options and got nowhere. My conclusion is that I'd be better off making a custom travel case or shell. This could double as a dust cover if designed for that. I think that, considering the dimensions, folding your units in half, while having the benefit of protecting the keys, would be harder to handle than a flatter package.

Keep up the good work.

jeffgran · « **Reply #29 on:** Sat, 12 September 2015, 14:21:26 »

ZOMBIE PROJECT IS BACK FROM THE DEAD...

Believe it or not I've been working on this off and on in my spare time over the last two years (damn, has it really been that long?)

After a few more prototypes, and one that I THOUGHT was the final print but turned out to have some spacing and other little issues, I finally have the final shape printed and I'm currently working on the hand-wiring.

As you can see I've completely given up on the "foldable" idea for this keyboard. maybe someday if I build another one I'll try a more complicated design to get that done. For this project I decided to prioritize comfort in typing over portability. I did a lot of experimentation with the thumb key positions to try to figure out how to get 6 keys per thumb that could all be pressed without moving my hand, shifting my wrists, etc. This is what I came up with.

Also I renamed the project to "ManuForm" -- a backwards-latinization of "hand-shaped".

Pics:

All the prototypes so you can see the progression. The red one is the one I thought was the final iteration. After it was discovered to be wrong, I painted it with spraypaint as a test to see whether it would work on this material.

A couple of more views:

The "cap" that goes over the little board that I'm attaching the teensy to. The little hole is positioned to be where the usb input is on the teensy.

Painting in slate grey/blue:

Okay, off to the garage to do some more soldering. Will have more pics Soon™ (it's getting exciting now so hopefully I will make quicker progress)

kurplop · « **Reply #30 on:** Sat, 12 September 2015, 14:34:51 »

Very impressive. It's good to see it coming together. I like the direction it is taking. Just don't wait 22 months for the next progress report.

suicidal_orange · « **Reply #31 on:** Sat, 12 September 2015, 14:40:56 »

I just saw this thread for the first time - the original idea looked very interesting but certainly no shortage of challenges.

The new single piece looks much more manageable and it looks like you're nearly there - exciting indeed!

heroinbob · « **Reply #32 on:** Sat, 12 September 2015, 15:01:06 »

Wow!! incredible.
Can't wait to see final.
Paul (HB)

vivalarevolución · « **Reply #33 on:** Sat, 12 September 2015, 17:06:25 »

Wow, really interesting. Great to see that you are still working on this.

pr0ximity · « **Reply #34 on:** Sat, 12 September 2015, 20:56:40 »

Wow, super cool to see you followed through with this! The new single-piece design looks great, looking forward to seeing the finished product and hear how you like it!

Is the final version also Shapeways-manufactured? What switches are you putting in it?

jeffgran · « **Reply #35 on:** Sat, 12 September 2015, 23:21:43 »

Thanks for the comments guys. @pr0ximity yeah, all Shapeways, their default "strong and flexible plastic" material. Kinda expensive, but short of buying a 3D printer myself I couldn't come up with any other options for this. Putting blue switches in there that I ripped off (desoldered from) a Cherry keyboard.

Man, doing all this soldering for the hand wiring is really a pain in the ass. I can't believe so many people have done this. Even hand-soldering those little teeny SMD diodes on the ErgoDox PCB was easier in my estimation.

jeffgran · « **Reply #36 on:** Tue, 03 November 2015, 23:35:31 »

Soldering is finished:

I put some stuffing from an old pillow in there to try to help dampen the echo-y noise from the switches inside the big hollow case. I have no idea what material this is so I tested it with my multimeter to make sure it doesn't conduct electricity first:

Naked:

All dressed up:

However:

I configured and flashed tmk firmware onto the teensy, and it is working... except for the top row on each side. None of the top-row keys work (including the two thumb keys that are part of the "top row" on each side). The rows/columns are actually columns/rows in tmk, I guess that's how it has to work because of how I did my diodes? I don't really know, but that's the only way I was able to get it to work. But I can't figure out what's wrong and I've run out of ideas on what to test... multimeter confirms that current flows from the teensy on that pin through a switch.

It's the white wires in the picture, on both sides that don't work. Does anybody with more experience than me have any ideas? Did I mess up the wiring somehow?

jeffgran · « **Reply #37 on:** Wed, 04 November 2015, 18:53:07 »

aww ****.

I think I know what I did wrong. I wired my diodes in series instead of in parallel. Put another way, I the solder point should have been AFTER the diode as it comes out of the switch, not before. Crap. back to the workbench. since I only have two diodes per "column" maybe there's an easy way to fix this...

EDIT: yep, confirmed by jumpering with alligator clips across one of the diodes between the top/middle "column", and that made the top key on that "row" work. Can anyone explain to me why this doesn't work with them in series? Is it just because with two diodes in series there is too much resistance? I thought that two diodes both oriented the same way would be fine in terms of the circuit completing.

PJE · « **Reply #38 on:** Thu, 05 November 2015, 05:54:17 »

There's a voltage drop of typically 0.7V across each diode. Depending on the supply voltage you're dropping the voltage supplied to the micro into the undefined region between 0 & 1. what type of diode were you using?

I'd measure the voltage at the input to ensure it's in the valid range.

CPTBadAss · « **Reply #39 on:** Thu, 05 November 2015, 07:36:17 »

Damn....what a great resurrection! How do you like the dampening you get from the pillow stuffing? I typically use drawer liner.

ideus · « **Reply #40 on:** Thu, 05 November 2015, 08:03:09 »

If this would be a portable it should be light, PCB mounted switches makes more sense than plate mounting.

jeffgran · « **Reply #41 on:** Sun, 08 November 2015, 16:41:04 »

Quote from: PJE on Thu, 05 November 2015, 05:54:17

There's a voltage drop of typically 0.7V across each diode. Depending on the supply voltage you're dropping the voltage supplied to the micro into the undefined region between 0 & 1. what type of diode were you using?

I'd measure the voltage at the input to ensure it's in the valid range.

The diodes I'm using are the 1N4148, here:http://www.amazon.com/gp/product/B00UXPVLEG

I tried measuring the voltage drop from the input on the teensy across all 3 diodes, the number was above 1, but I'm not sure if I was really doing it right.

Anyway, I think I'm going to re-solder all the diodes so they're in parallel, one per switch.

jeffgran · « **Reply #42 on:** Sun, 08 November 2015, 16:42:59 »

Quote from: CPTBadAss on Thu, 05 November 2015, 07:36:17

Damn....what a great resurrection! How do you like the dampening you get from the pillow stuffing? I typically use drawer liner.

Thanks, just hope I can get it actually working so I can try using it.

The pillow stuffing does OK, definitely better than nothing, but not that great either. Maybe I need to stuff more in there, or find something denser. It's a weird shape to try to fill.

jeffgran · « **Reply #43 on:** Fri, 13 November 2015, 00:22:16 »

Yeeehaw! Just have to post my long-awaited "I'm typing this post on my new keyboard" post! So excited to have it working. I'm sure I'll be spending a lot of time tweaking the layout.

Hasu if you're listening, THANK YOU for your tmk firmware, it is so amazing! I feel like there's nothing it can't do, and I can't wait to keep tweaking my layouts and layers to my liking.

And of course thank you to everyone else who came before me and led the way -- this forum has been like my bible on this project as I struggle through this project. Thank you all for being such a kind and helpful and ambitious and skillful community. God, I know, I sound so cheesy... but I can't help it, I'm just so excited!

ursulache · « **Reply #44 on:** Fri, 13 November 2015, 03:43:41 »

Congratulations

vvp · « **Reply #45 on:** Fri, 13 November 2015, 04:18:56 »

Quote from: jeffgran on Sat, 12 September 2015, 23:21:43

Man, doing all this soldering for the hand wiring is really a pain in the ass. I can't believe so many people have done this. Even hand-soldering those little teeny SMD diodes on the ErgoDox PCB was easier in my estimation.

Use Ø 0.2 mm enameled (magnet) wire instead of a wire with a plastics insulation. You will save time stripping insulation at the right places. Soldering iron (at abut 400 °C) will just burn through the enamel at the joint place. The enamel will stay insulating the wires on the other places.
Check the last picture in this post for an example.

Nice project. Keep going on.

kurplop · « **Reply #46 on:** Fri, 13 November 2015, 05:40:45 »

Congratulations. You did a really nice job on this. I can definitely understand the thrill. Whenever I fire up a new board for the first time and it works, I have an uncontrollable urge to giggle. I'm all by myself and I'm giggling. Go figure.

So how does it feel using it? Do you think it will take long to get use to the physical layout? Any concerns?

tufty · « **Reply #47 on:** Fri, 13 November 2015, 05:42:53 »

Excellent news! It looks proper "space age", too. Well done.

suicidal_orange · « **Reply #48 on:** Sat, 14 November 2015, 08:24:57 »

Glad to see this working, but aren't the diodes wrong? They appear to be wired with the black end of the top switch connected to the clear end of the one below, where usually all the black ends are connected.

If you press the three top middle corner switches (Y U and H?) at the same time does it detect a press of J?

Mad_Maxx · « **Reply #49 on:** Sat, 14 November 2015, 12:33:44 »

This is some next-level space age stuff.
Great project.


120/100g linear Zealio R1 GMK Hyperfuse 'Split everything' perfection	MX Clear SA Hack'd by Geeks EasyAVR mod

News:

Author Topic: ManuForm: custom contoured 50% ergo board (Read 109134 times)