Author Topic: Re-Create the DataHand - Thumb cluster under development. Project 75% done.  (Read 182602 times)

0 Members and 1 Guest are viewing this topic.

Offline OldDataHands

  • Thread Starter
  • Posts: 277
  • Location: Michigan
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #250 on: Wed, 26 February 2014, 20:30:18 »
I would really love to be able to mouse without leaving the DH. [...] For me the best two options to put a trackpoint-style 'nubbin' would probably be ...

I quite agree with you about the desirability of incorporating some type of pointing
device into the design. It is already supported in a couple different ways by the PCB,
but there is no final or perfect solution identified.

Along with the couple options Turbinia has proposed with his case design
( http://geekhack.org/index.php?topic=41422.msg1060497#msg1060497 ) ,
I quite like what gator456 did here: http://geekhack.org/index.php?topic=12212.0

I do not believe that packaging will allow your #1 position, though I think it would
be comfortable. I'm not personally a fan of your #2 position, as it seems awkward to
me. I'm also not sure that the parts will be able to fit in that spot.

If you used your middle finger in the same way as you're using your index finger in #1,
I think that could be supportable, and comfortable. ( my understanding is that you're
imagining the "nubbin" to be sticking out of the front edge of the palm support ).

Please note that I have not been concerned with being mechanically compatible with
the existing DataHand PCBs. You might be able to achieve that by tweaking my OpenSCAD
dimensions, but I have not been targetting that. I am targetting a complete replacement,
and therefore new PCBs too.  Along with that will come complete control over your layout
and key assignments - the joy of a custom keyboard.

The feel of the center switch (and the others for that matter) is greatly affected by the shape
and mass of steel in the clips to which the magnet is drawn. If you wanted to increase the
force required to release the center key, you could choose to make clips out of 0.020" material
instead of 0.015" material and make them 0.050" wide rather than 0.045" or anywhere
inbetween.  This is a wide range of release force. I bet you could be happy somewhere in there.
I have not considered using a cherry switch in the center because 1) MX are too large, 2) I want
to maintain the feel of the magnetic release on the switches.

I was wondering if this design still uses optical switches (which have the disadvantage of potentially not working in direct sunlight)
If it is so, why? Is there a viable alternative?

I am primarily a coder. I work best in the dark, far from windows. I've only ever had a
direct-sun-on-my-datahand experience one time in my 15 years of use, and while I
found it initially confusing, I just had to slightly adjust the horizontal blinds and the
"problem" was solved.

Anyhow, the real answer is that I couldn't come up with a better solution.
This approach allows me to create a reliable switching mechanism in a
funny, small, custom package without the need of making super-high-tolerance
electrical switch contacts. I'm an EE by degree, and a software guy by trade.
I'm sure that the ME's (and people with access to Pro-E!) look at my stuff
here and mutter, but I'm only concerned with getting something that works.
A tangible, working design that will hopefully be improved upon, but can be
used as-is.

However, more to your point: 3D printing allows some easy improvements to
the original in terms of shielding. I have no data to prove it, but I expect that
I have improved on the resistance to ambient and IR light through being able
to have the IR phototransistor snap partially into a box which shields its front
completely and does not expose the eye from above at all. Should be better.
 



Offline Hashable

  • Posts: 4
  • Location: Ottawa, Canada
  • KAFP PHE TP U K
    • Ted Morin
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #251 on: Wed, 26 February 2014, 21:59:34 »
However, more to your point: 3D printing allows some easy improvements to
the original in terms of shielding. I have no data to prove it, but I expect that
I have improved on the resistance to ambient and IR light through being able
to have the IR phototransistor snap partially into a box which shields its front
completely and does not expose the eye from above at all. Should be better.

Thanks for the input. I really hope that your work is successful and that I can contribute if you decide to GB or Crowdfund.

In addition, I believe that the middle-finger mouse-nib is a good option. I know a lot of people who use the IBM/Lenovo nibs with their middle-finger so I think it would work, certainly better than a thumb!

Offline eviltobz

  • Posts: 95
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #252 on: Thu, 27 February 2014, 03:57:18 »
...I am primarily a coder. I work best in the dark, far from windows...
:)

Offline Stone

  • Posts: 22
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #253 on: Thu, 27 February 2014, 11:00:17 »
I quite like what gator456 did here: http://geekhack.org/index.php?topic=12212.0
I like that approach too, though I'm right handed! Unfortunately it wouldn't work as well for me as I like to have the finger wells dialled quite far forward and I'm not sure if there's enough space left over. It's a very tidy mod, though.

I do not believe that packaging will allow your #1 position, though I think it would
be comfortable. I'm not personally a fan of your #2 position, as it seems awkward to
me. I'm also not sure that the parts will be able to fit in that spot.

If you used your middle finger in the same way as you're using your index finger in #1,
I think that could be supportable, and comfortable. ( my understanding is that you're
imagining the "nubbin" to be sticking out of the front edge of the palm support ).
That's a fair point and a good idea. I used to have a Toshiba Libretto 70ct which had a very wide/flat top to the trackpoint-style mouse  - if such a device could be mounted in the front of the palm support (top facing the keywells) for use by the middle finger, perhaps it could be flanked by two small rectangular tactile switches for left/right click?

Please note that I have not been concerned with being mechanically compatible with
the existing DataHand PCBs. You might be able to achieve that by tweaking my OpenSCAD
dimensions, but I have not been targetting that. I am targetting a complete replacement,
and therefore new PCBs too.  Along with that will come complete control over your layout
and key assignments - the joy of a custom keyboard.[/ quote]
Yeah, I understand that - the best thing about opening it up like you have is that I could go as far as building a new unit from scratch or mod your case designs to fit my existing DH parts. It's great to have the flexibility!

The feel of the center switch (and the others for that matter) is greatly affected by the shape
and mass of steel in the clips to which the magnet is drawn. If you wanted to increase the
force required to release the center key, you could choose to make clips out of 0.020" material
instead of 0.015" material and make them 0.050" wide rather than 0.045" or anywhere
inbetween.  This is a wide range of release force. I bet you could be happy somewhere in there.
I have not considered using a cherry switch in the center because 1) MX are too large, 2) I want
to maintain the feel of the magnetic release on the switches.
That's interesting, I might experiment with mine at some point to see if I can firm it up a little. They're still my favourite input method but there's always room for personal taste :)

As for the sunlight thing, I work in a fairly bright open-plan office and never had a problem. They're not as badly affected as everyone assumes!

Stone

Offline Turbinia

  • Posts: 64
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #254 on: Thu, 27 February 2014, 15:33:01 »
I haven't worked on the track-point in a while. Hit a bit of a stumbling block with my easypoint module, was getting the interrupts for the button but no xy. I will probably try in a month or two to recompile with the C libraries instead of the arduino ones for I2C and see what that does. Really any option will have sourcing issues, the easypoint module is discontinued in a retail package and there is no easy way to source ibm style trackpoints either without ripping up keyboards. As for the mount on the palm rest, there looks like there would be room, but there really isn't. Did some testing with some mock ups and it just didn't fit the pcb for the pointers.
| Dolch | KBT ONE | QFR w/PBT | Poker II |

Offline OldDataHands

  • Thread Starter
  • Posts: 277
  • Location: Michigan
56918-0

Starting to put together the thumb cluster model.

Still need to implement the "press-down" switch
for the thumbs. I think I can make something like
the original 2-stage DataHand switch, so will give
that a go next.

Offline OldDataHands

  • Thread Starter
  • Posts: 277
  • Location: Michigan
Re: Re-Create the DataHand - work on finger switches nearly complete
« Reply #256 on: Thu, 13 March 2014, 23:07:50 »
Thanks to damorgue for the input. With
that, some serious redrawing, then redlining
by an M.E. friend, then more redrawing
I now have a drawing to get quoted.
I'll pass along the info once they get
back to me.
...

Well, the shop that I wanted to use to
manufacture a batch of clips declined saying
"Unfortunately it is just too small for us to work with."

I'm a bit annoyed and disappointed, but
will eventually find some way to get them
made other than my crude bending jig.

Anyone have good experiences with any particular
website for soliciting quotes on metal fabrication work?

Offline damorgue

  • Posts: 1176
  • Location: Sweden
    • Personal portfolio
I know a Swedish manufacturer called Lesjöfors which makes clips, springs and other various small stamped metal contact parts. It might not be ideal but perhaps their site can give you some help in finding a similar manufacturer locally.

Offline OldDataHands

  • Thread Starter
  • Posts: 277
  • Location: Michigan
59229-0

So, this is the guts of the thumb down double switch.
It still needs to be equipped to connect to the keycap,
but this is the first mechanical design of the switching
part.

Essentially, the key-cap will press off-center on the bar
(show below the box it lives in). This will force one end
of the bar to move first, only causing the second end
of the bar to move with greater force applied.
« Last Edit: Mon, 31 March 2014, 00:44:20 by OldDataHands »

Offline psyclops

  • Posts: 1
  • Location: San Francisco, CA
Firstly, let me say a big THANK YOU for working on this... recreating the DataHand has been at the back of my mind for a while, as I always knew that my two DataHands would eventually wear out, and I figured I had the mechanical fabrication chops to rebuild the thing from scratch at this point.  But, to find that someone else is already working on it is a boon!  My second is now showing glitches and a little googling brought me here.  I have extensive 3D design experience (more on the mechanical than electrical) if you need any help...

BTW, I have a great sheet metail shop in the SF Bay Area who have done small run clips for me in the past.  Small sheet metal runs are not cheap however... but these guys' fabrication and more importantly QA is second to none, I have never received a bad part from them.  The company is called Vander-Bend and I can PM you the contact info for my guy over there.

cheers,

nick.

Offline OldDataHands

  • Thread Starter
  • Posts: 277
  • Location: Michigan
Unfortunately, Vander-Bend declined to quote making the clips,
indicating that this part was not a good fit for them. That's the
second shop which really looked well-placed (to my untrained eyes)
and decided that they weren't interested.

If at first you don't succeed...


Offline OverKill

  • Posts: 108
  • Location: Arizona, USA
At my old shop we used to get a lot of stuff done at Bokers. www.bokers.com
Not sure how small of a part they can do but they seem to do some pretty decent stuff. The only thing is I don't know what their MOQ is or if they even have one.

Offline Zekromtor

  • Posts: 229
    • My Setup
I would love a thumb section of the datahand combined with my kinesis.
Great work.

Offline nclu

  • Posts: 3
I miss seeing these updates in the morning. Has the project moved? Is it still going?

Offline OldDataHands

  • Thread Starter
  • Posts: 277
  • Location: Michigan
Real life is interfering. I expect to be able to work on this again this fall or winter. probably not before.

Offline eviltobz

  • Posts: 95
that's ok, it's just more time to save money for an eventual group buy. that's ok, it's just more time to save money for an eventual group buy. that's ok, it's just more time to save money for an eventual group buy. that's ok, it's just more time to save money for an eventual group buy. that's ok, it's just more time to save money for an eventual group buy. that's ok, it's just more time to save money for an eventual group buy. that's ok, it's just more time to save money for an eventual group buy. that's ok, it's just more time to save money for an eventual group buy.

/sobs

;)

Offline jeep

  • Posts: 33
  • Location: Hillsboro
Sorry for the spam, but this keeps falling off my radar, so I'm posting here to keep tabs on this project. I am looking forward to seeing what happens with it.

Offline wolfv

  • Posts: 235
Burro Volando has a easy improvement for DataHand:
Quote
I think the keys were designed erroneously with too large a “dish” in the center and with the outer ridge too close to the side keys. I would have made the “dish” about one-quarter or at most 5/16th of an inch in diameter.
from http://deskthority.net/photos-f62/datahand-t2384-90.html

Offline OldDataHands

  • Thread Starter
  • Posts: 277
  • Location: Michigan
Well, BV's complaint isn't one that I share, but
happily for them, it'll be easy (and inexpensive)
for them to print out a set of smaller buttons if
that's what they want.

in other news: Still likely to be fall before I have
time to work on this again - just isn't possible
at the moment.

Offline OldDataHands

  • Thread Starter
  • Posts: 277
  • Location: Michigan
https://github.com/dodohand/dodohand

just a PSA: my sources are all up on github.

Offline bieniekmat

  • Posts: 1
Thanks for doing this. I have RSI and I hope to find something like that one day on the mark. Have you ever thought of starting a kickstarter project? I'd be glad to donate some money and see such project coming to fruition.

Many thanks, Mat

Offline LaPoune

  • Posts: 16
Really cool project. With the recent popularity and countless "ergo" cherry designs I was wondering why no one attempted to make a modern version of the datahand. I'll take 10! :)
Realforce 87U AE | Realforce 87U | Realforce 86U | HHKB JP | HHKB PRO | Filco Tenkeyless blue | Deck 82 Toxic (blue switches) | Crappy Gigabyte black cherries | Corsair K60 | Dell Model M

Offline wolfv

  • Posts: 235
Please lend your broken DataHand
« Reply #272 on: Sun, 31 August 2014, 22:25:39 »
This is an awesome project.
OldDataHands has uploaded quality work to https://github.com/dodohand/dodohand

The thumb cluster still needs to be modeled; it is the most complicated part on DataHand.
People that have DataHand don't need to make one, and people that need DataHand can't model it.  A Catch-22.

I would model the thumb cluster if I had access to a DataHand.
If a generous soul could lend me a broken DataHand they are not using, I could model the thumb cluster and write the firmware.
That would complete the DodoHand project and return the DataHand.
Even broken DataHands are valuable, so an escrow could be arranged.
I live in the USA.

I know I can do this because of my experience from making a small split keyboard.
I made the split keyboard small so that I could use it without moving my hands.
I modeled the keyboard and Shapeways 3D-printed the mount plates.
I assembled, soldered, and wrote the firmware for the Teensy & MCP23018 IO expander.
After a month of using the small keyboard, my arms felt great, but my fingers hurt.
Now I am desperate for a DataHand or something like it.

My motivation is selfish, but the DodoHand project is open source so many others can benefit as well.
Please PM me if you have a broken DataHand you could lend.

Thank you.
« Last Edit: Sun, 31 August 2014, 22:43:02 by wolfv »

Offline wolfv

  • Posts: 235
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #273 on: Wed, 10 September 2014, 02:19:17 »
Now that I think of it, I don't need much hardware to develop the software.
I can develop 99% of the software on an 8-key breadboard.
That way the firmware will be ready by the time OldDataHands finishes the thumb-cluster modeling :).

Please answer two questions so I can get started on the firmware:
 1) Is the space between the emitter and optic sensor normally blocked or unblocked?
 2) What is a good example firmware for implementing optical switches?
   scantest.cpp
   led_test.cpp

Offline OldDataHands

  • Thread Starter
  • Posts: 277
  • Location: Michigan
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #274 on: Wed, 10 September 2014, 23:16:03 »
The IR LED is normally blocked from shining onto the phototransistor.
This helps to reduce power consumption (no current through phototransistor).

led_test.c scans the finger switches for both hands and activates the status
LEDs on both sides. This is the file that should be referred to for how to read
the switch states (well, this, and scanlh.c and scanrh.c which do the actual
scanning of the switch matrices).

This ought to be a good resource for datahand firmware, and a working
example of the datahand modifiers. It is my intention to leverage this project
to the greatest extent possible:
https://github.com/imarko/dhteensy

I also think it would be worthwhile to look at Soarer's and TMK's firmware
to see if it makes sense to present the key switch status in a manner which
makes it possible to leverage one of those projects.


Offline wolfv

  • Posts: 235
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #275 on: Thu, 11 September 2014, 11:05:57 »
Thanks OldDataHands.  I will study the dhteensy working example of the datahand modifiers.

I did a quick survey of some keyboard firmwares and if they use an io expander.

These firmwares do not contain the word "expander" in the source or documentation:

These firmwares use io expanders:
And two questions, if you don't mind:

In the DodoHand firmware, what does "trig" mean? e.g.
   pca9655e_trig_scanrh()
   pca9655e_trig_cyc()
   pca9655e_trig_init()

In the DodoHand Bill of Materials, what are the 22 "test point" parts for?:
 http://www.digikey.com/product-detail/en/1040/1040K-ND/315157

Offline eviltobz

  • Posts: 95
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #276 on: Thu, 11 September 2014, 11:26:26 »
...
These firmwares do not contain the word "expander" in the source or documentation:

These firmwares use io expanders:

There is a TMK ErgoDox port here https://github.com/cub-uanic/tmk_keyboard so you could use that as a starting point if you want the TMK feature set.

Offline wolfv

  • Posts: 235
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #277 on: Thu, 11 September 2014, 18:19:07 »
Thanks eviltobz; its good to know all the options :thumb:.
The TMK feature set does seem to be a bit much.
More about the TMK ErgoDox port here http://geekhack.org/index.php?topic=48106.0

Offline OldDataHands

  • Thread Starter
  • Posts: 277
  • Location: Michigan
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #278 on: Fri, 12 September 2014, 23:16:49 »
Thanks OldDataHands.  I will study the dhteensy working example of the datahand modifiers.

I did a quick survey of some keyboard firmwares and if they use an io expander.

These firmwares do not contain the word "expander" in the source or documentation:

These firmwares use io expanders:
And two questions, if you don't mind:

In the DodoHand firmware, what does "trig" mean? e.g.
   pca9655e_trig_scanrh()
   pca9655e_trig_cyc()
   pca9655e_trig_init()

In the DodoHand Bill of Materials, what are the 22 "test point" parts for?:
 http://www.digikey.com/product-detail/en/1040/1040K-ND/315157
trig is short for trigger.
trig_scanrh() triggers a scan of the IO expander (IOX) switch matrix.
trig_init() configures IOX inputs and outputs.
trig_cyc() will trigger a single read/write cycle from/to the IOX.

The test points allow you to securely attach a probe to a signal. only for debugging.

just as background info: I think the communications with the expander are
working nicely, as is the scanning of the finger switches.

Offline wolfv

  • Posts: 235
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #279 on: Fri, 12 September 2014, 23:52:49 »
OldDataHands,

Since you already tested the PCB, would the test points be of any use to me?
(I have a multimeter and know basic electronics).
« Last Edit: Sat, 13 September 2014, 01:04:46 by wolfv »

Offline OldDataHands

  • Thread Starter
  • Posts: 277
  • Location: Michigan
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #280 on: Sat, 13 September 2014, 10:46:07 »
Depends on what parts of the software you modify.
these guys:
https://www.saleae.com/
make a nice, inexpensive tool which you might want to use if you modify the matrix scanning code. easy to make a bug. easier to fix if you can see the electrical signals.

Offline wolfv

  • Posts: 235
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #281 on: Sat, 13 September 2014, 20:48:02 »
Thank OldDataHands.  I never heard of a logic analyzers, and that looks like a nice one.

Offline rosingle

  • Posts: 1
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #282 on: Sun, 28 September 2014, 16:41:23 »
great project... Suffering from dystonia it would be nice if datahand could be available again...

Offline caseyandgina

  • Posts: 54
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #283 on: Tue, 30 September 2014, 11:24:29 »
I hope this comes to fruition.  When complete, what will it cost?

Offline OldDataHands

  • Thread Starter
  • Posts: 277
  • Location: Michigan
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #284 on: Wed, 01 October 2014, 10:20:45 »
Just wanted to note that I am starting to work on this again, a bit.

When complete, what will it cost?

I recall thinking that, depending on case design, I could make it for ~$600, not counting time or tools of course.

Offline vivalarevolución

  • Posts: 2148
  • Location: Naptown, Indiana, USA
  • Keep it real b/c any other way is too stressful
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #285 on: Thu, 02 October 2014, 08:16:15 »
Just wanted to note that I am starting to work on this again, a bit.

When complete, what will it cost?

I recall thinking that, depending on case design, I could make it for ~$600, not counting time or tools of course.

I think if you make it available to receive donations for your efforts, some of us will be more than willing to send you compensation.
Wish I had some gif or quote for this space, but I got nothing

Offline lohapuk

  • Posts: 1
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #286 on: Sun, 05 October 2014, 04:56:11 »
Maybe it's worth setting up a Kickstarter campaign, I'm more than willing to donate. 

Offline OldDataHands

  • Thread Starter
  • Posts: 277
  • Location: Michigan
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #287 on: Sun, 05 October 2014, 22:53:07 »
I think if you make it available to receive donations for your efforts, some of us will be more than willing to send you compensation.
Maybe it's worth setting up a Kickstarter campaign, I'm more than willing to donate.

Thanks guys, but time is the bigger challenge. Maybe I'll consider trying to raise money
if it comes to needing to purchase some real tooling, but right now, SLS nylon is affordable!

Offline yasuo

  • Posts: 976
  • Location: ID
  • spanengan puyeng newbie
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #288 on: Sun, 05 October 2014, 22:55:40 »
nice:thumb:
Logitech MK220 Colemak DH
SplitSyml by Moz BlacksMx fuk blacks

2/3 8.5pm                                          in de la my september month ya da all get my fukka "fake message"

Offline OldDataHands

  • Thread Starter
  • Posts: 277
  • Location: Michigan
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #289 on: Mon, 06 October 2014, 00:04:32 »
Rough shape of the thumb down double switch is complete!
78973-0
only a few more details to add to the model.
78975-1

Offline OldDataHands

  • Thread Starter
  • Posts: 277
  • Location: Michigan
Re: Re-Create the DataHand - Thumb switches under development. Project 70% done.
« Reply #290 on: Sat, 11 October 2014, 23:36:24 »
Starting to finalize the arrangement of the whole thumb cluster.
Here are a few pics of where it stands:
79455-0

79457-1

79459-2
There are still a few tweaks required, but I'm starting to get excited
about placing my next parts order with Shapeways so that I can
see how these thumb switch mechanisms will work!

Offline OldDataHands

  • Thread Starter
  • Posts: 277
  • Location: Michigan
Re: Re-Create the DataHand - Thumb cluster under development. Project 75% done.
« Reply #291 on: Thu, 16 October 2014, 22:47:47 »
Working on the PCB footprints for the thumb switches now.
Finally getting back to incorporating regack's solder bridge science
into the project.
79975-0
This is two of these solder bridges, on opposite sides of the board,
connected together by a single PTH / via.

Hope to have a thumb PCB designed this month.

Offline hillel369

  • Posts: 2
Re: Re-Create the DataHand - Thumb cluster under development. Project 75% done.
« Reply #292 on: Sat, 18 October 2014, 13:19:28 »
Just another long time DataHand user ready to purchase once they're ready.

Buying my DataHand saved my career in programming many years ago.

Keep up the great work!!

Offline OldDataHands

  • Thread Starter
  • Posts: 277
  • Location: Michigan
Re: Re-Create the DataHand - Thumb cluster under development. Project 75% done.
« Reply #293 on: Tue, 21 October 2014, 00:18:06 »
Some progress made on the thumb PCB:
80344-0

Offline hweller

  • Posts: 25
  • Location: UK
Re: Re-Create the DataHand - Thumb cluster under development. Project 75% done.
« Reply #294 on: Tue, 21 October 2014, 04:21:02 »
I have been following this excellent thread and work closely and had printed a finger cluster using and ABS filament with partial success.  It looks like a filament-based printer could be used but it would need PVA support and probably using nylon rather than ABS.  When the local printing company has a dual-head machine with nylon/PVA we will have another go.

In the meantime I purchased an old DH200-6 in order to test my plan to build a 2 1/4" finger trackball into the DodoHand.  I am using the ADNS9800 laser sensor for the ball and build the prototype cup out of polymorph.  The first problem I hit was how old the electronics in the DH200-6 is and decided to replace it entirely with a Teensy microcontroller.  I chose the Teensy-3.1 because I had one in stock and wanted a project to play with it.  The prototype electronics is on veroboard for now, if all goes well I will look into creating a PCB.

One big headache was the extent of the thumb-cluster PCB which pushed the trackball too far from a comfortable position.  In order to trim this board I had to move the up-switch inwards, re-mold the leaver and re-wire underneath.  With a lot of trimming and messing about I managed to get the ball into a good position and trimmed the case appropriately to expose the ball.  If others are interested in putting a 2 1/4" trackball into the DodoHand we may need to look carefully at the extend of the thumb-cluster PCB and possible relocation of the up-switch closer-in and appropriate changes to the leaver.

I have written the code for right-hand and have the ball working, mouse buttons (mapped to thumb switches), key matrix (Dvorak layout) and it is all working really well.  I am now working on IO expander firmware for the left-hand unit with an I2C connection to the right-hand unit.  In the longer-term I might put a trackball in the left-hand unit also.

I have  attached to photos of the right-hand unit.  I would be happy to provide more detail and discuss this project if there is general interest in a track-ball option for the DodoHand.

P.S.  This is my first post here and I am not sure how the images will be processed, sorry if it doesn't come out right.

Offline dorkvader

  • Posts: 6282
  • Location: Boston area
  • all about the "hack" in "geekhack"
Re: Re-Create the DataHand - Thumb cluster under development. Project 75% done.
« Reply #295 on: Tue, 21 October 2014, 11:03:15 »
I have been following this excellent thread and work closely and had printed a finger cluster using and ABS filament with partial success.  It looks like a filament-based printer could be used but it would need PVA support and probably using nylon rather than ABS.  When the local printing company has a dual-head machine with nylon/PVA we will have another go.

In the meantime I purchased an old DH200-6 in order to test my plan to build a 2 1/4" finger trackball into the DodoHand.  I am using the ADNS9800 laser sensor for the ball and build the prototype cup out of polymorph.  The first problem I hit was how old the electronics in the DH200-6 is and decided to replace it entirely with a Teensy microcontroller.  I chose the Teensy-3.1 because I had one in stock and wanted a project to play with it.  The prototype electronics is on veroboard for now, if all goes well I will look into creating a PCB.

One big headache was the extent of the thumb-cluster PCB which pushed the trackball too far from a comfortable position.  In order to trim this board I had to move the up-switch inwards, re-mold the leaver and re-wire underneath.  With a lot of trimming and messing about I managed to get the ball into a good position and trimmed the case appropriately to expose the ball.  If others are interested in putting a 2 1/4" trackball into the DodoHand we may need to look carefully at the extend of the thumb-cluster PCB and possible relocation of the up-switch closer-in and appropriate changes to the leaver.

I have written the code for right-hand and have the ball working, mouse buttons (mapped to thumb switches), key matrix (Dvorak layout) and it is all working really well.  I am now working on IO expander firmware for the left-hand unit with an I2C connection to the right-hand unit.  In the longer-term I might put a trackball in the left-hand unit also.

I have  attached to photos of the right-hand unit.  I would be happy to provide more detail and discuss this project if there is general interest in a track-ball option for the DodoHand.

P.S.  This is my first post here and I am not sure how the images will be processed, sorry if it doesn't come out right.

Oh my goodness that's awesome! One heck of a first post, thanks for sharing. That mod is really very nicely done.

About getting the LH running, you can adapt the firmware from the ergodox, which uses I2C and an MCP23018 IO expander. Still I have heard that it's not the best choice cue to the low speed of I2C. You might want to use some shift registers (like many vintage keyboards and bpiphany's costar replacement controllers (two 74HC42's))

Offline hweller

  • Posts: 25
  • Location: UK
Re: Re-Create the DataHand - Thumb cluster under development. Project 75% done.
« Reply #296 on: Tue, 21 October 2014, 13:04:52 »
I have a MCP23018 connected to a Teensy-3.1 on a piece of breadboard and playing with the code from the ergodox.  I would rather use the Wire library than the lower-level twi library so I am unravelling the protocol from the ergodox code and writing a MCP23018 class.  I should have this finished this evening and tomorrow I will connect the thumb and finger clusters from the LH unit to the breadboard and finish the programming.    I hope to have the LH unit finished by the end of the week.

I will release all of the details and code on GitHub as soon as it is ready.

Offline OldDataHands

  • Thread Starter
  • Posts: 277
  • Location: Michigan
Re: Re-Create the DataHand - Thumb cluster under development. Project 75% done.
« Reply #297 on: Tue, 21 October 2014, 22:33:24 »
hweller - that is a thing of beauty. Welcome to GeekHack, and what an entrance!

I will be acutely interested in hearing how you like that trackball positioning.
It could be an amazing improvement. I will have my ears/eyes open for your
updates.  Please feel free to put your updates in this thread - It'd help me to
keep track of them, but your project certainly deserves a thread of its own.
Excellent work.

I'll have to adjust the layout of my finger and thumb boards to help make
room for that. It looks very promising.

My update for the night is the OSH Park rendering of the first rev of the thumb
PCB:

80407-0

80409-1

Not yet ordered, as I want to do some double-checking of footprints and such first.
It is on the order of 3 inches by 2 inches, so $30 from OSH Park for 3 copies.

Offline hweller

  • Posts: 25
  • Location: UK
Re: Re-Create the DataHand - Thumb cluster under development. Project 75% done.
« Reply #298 on: Wed, 22 October 2014, 04:32:15 »
I am glad there is interest in my idea of building a trackball into the DodoHand, I really think this will be a good option.  I am more than happy to post my progress to this thread as I see it as a prototype option for the DodoHand rather than an upgrade to the obsolete DataHand.  I will also release all details and code publicly in case anyone else is interested in cannibalizing a DH200 in the same way but I must stress that it is far from trivial and would be easier to start from the DodoHand components when they are complete.

I have been using the right-hand unit as a trackball for a week now and find it surprisingly comfortable and convenient.  In mouse-mode I have mapped the thumb down key as the left mouse button, the second level down switch and left double-click, the thumb pad key as the middle mouse button and the thumb nail key as the right mouse button.  This arrangement feels similar to Logitech TrackMan Marble FX which I used from the late 90s until I fitted a Kensington Slimblade ball unit into the middle of my Kinesis Advantage keyboard.  While this is a good arrangement and saves on arm movement the original Logitech TrackMan is the best trackball I have used and I am hoping the trackball fitted into the DodoHand will be even better.

One minor issue I have with the current hardware I am using is that the Kicklighter breakout board for the ADNS9800 laser sensor:

https://www.kickstarter.com/projects/1034145369/high-speed-laser-optical-sensor
https://www.tindie.com/products/jkicklighter/adns-9800-optical-laser-sensor/

is quite large and needs to be mounted opposite the point at which the fingers move the ball which is hard to achieve as it hits the base of the case.  However John Kicklighter has a smaller version of the breakout board which if he releases would solve this problem completely.

Offline vivalarevolución

  • Posts: 2148
  • Location: Naptown, Indiana, USA
  • Keep it real b/c any other way is too stressful
Re: Re-Create the DataHand - Thumb cluster under development. Project 75% done.
« Reply #299 on: Wed, 22 October 2014, 14:15:32 »
When you say Dodohand, what are you referring to?  Just need some clarification.
Wish I had some gif or quote for this space, but I got nothing