Author Topic: IBM M15 Clone - in Cherry  (Read 3767 times)

0 Members and 1 Guest are viewing this topic.

Offline 0100010

  • Thread Starter
  • Posts: 1127
  • Location: DFW, TX, US
  • Not Sure
IBM M15 Clone - in Cherry
« on: Fri, 21 November 2014, 09:02:37 »
http://www.keyboard-layout-editor.com/#/layouts/4444e5d1f5cb71213628bde67409cf83





I've got a coworker who currently uses an old rubber dome Microsoft Natural keyboard  >:D , that I am dragging into the mechanical keyboard world  :thumb: .  After letting him review the myriad or ergonomic keyboards out there online, he has decided he would like a board that has the layout of the IBM M15  ;D - but he doesn't want the buckling spring noise that would come with one  :'( .  After letting him test out the various Cherry switches, I think he has settled on MX Clears.  I mentioned we could draw up a set of plates in the layout he wants, and build the case in layers much like an Ergodox.  Also like an Ergodox, it could have completely separate halves (which could be tented as desired independently).  Since I expect this will be a one off keyboard it will be hand wired matrix with a Teensy.

But - what I need help with, is how to connect the two halves of the keyboard.  Would the Ergodox solution of a TRRS cable and I/O expander work?  Would it be as simple as using the Ergodox firmware and updating the matrix?  Or would we need a whole new firmware?

Software in this space is not my strong suit - any assistance here would be appreciated.
  Quoting me causes a posting error that you need to ignore.

Offline yasuo

  • Posts: 978
  • Location: ID
  • spanengan puyeng newbie
Re: IBM M15 Clone - in Cherry
« Reply #1 on: Fri, 21 November 2014, 09:10:46 »
sym sgg would be great :D
« Last Edit: Fri, 21 November 2014, 09:15:37 by yasuo »
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 0100010

  • Thread Starter
  • Posts: 1127
  • Location: DFW, TX, US
  • Not Sure
Re: IBM M15 Clone - in Cherry
« Reply #2 on: Fri, 21 November 2014, 16:24:07 »
No clue what that means.
  Quoting me causes a posting error that you need to ignore.

Offline jacobolus

  • Posts: 3661
  • Location: San Francisco, CA
Re: IBM M15 Clone - in Cherry
« Reply #3 on: Fri, 21 November 2014, 16:31:15 »
sym sgg is yasuo’s abbreviation of “symmetric stagger”. I’m not sure why he always says it that way, since almost no one understands. :-)

He means something like this:

Offline 0100010

  • Thread Starter
  • Posts: 1127
  • Location: DFW, TX, US
  • Not Sure
Re: IBM M15 Clone - in Cherry
« Reply #4 on: Fri, 21 November 2014, 17:10:55 »
Ah - discussed that with my coworker.  He prefers to keep the standard layout, just split the halves // just like the M15 is:

« Last Edit: Fri, 21 November 2014, 17:15:32 by 0100010 »
  Quoting me causes a posting error that you need to ignore.

Offline Puddsy

  • nice
  • * Elated Elder
  • Posts: 12275
  • Location: RSTLN E
  • "Do you shovel to survive, or survive to shovel?"
Re: IBM M15 Clone - in Cherry
« Reply #5 on: Fri, 21 November 2014, 22:51:45 »
I've been thinking about a sym sgg (I'm gonna start saying it so it catches on), but I don't like the ergodox. I'd like to see where this goes.
QFR | MJ2 TKL | "Bulgogiboard" (Keycon 104) | ctrl.alt x GON 60% | TGR Alice | Mira SE #29 | Mira SE #34 | Revo One | z | Keycult No. 1 | AIS65 | First CW87 prototype | Mech27v1 | Camp C225 | Duck Orion V1 | LZ CLS sxh | Geon Frog TKL | Hiney TKL One | Geon Glare TKL



"Everything is worse, but in a barely perceptible and indefinable way" -dollartacos, after I came back from a break | "Is Linkshine our Nixon?" -NAV | "Puddsy is the Puddsy of keebs" -ns90

Offline dorkvader

  • Posts: 6288
  • Location: Boston area
  • all about the "hack" in "geekhack"
Re: IBM M15 Clone - in Cherry
« Reply #6 on: Sat, 22 November 2014, 00:31:36 »
Ah - discussed that with my coworker.  He prefers to keep the standard layout, just split the halves // just like
I still think playing with sym stg is a good idea. You can mock up keyboard designs in cardboard to see how they feel
I've been thinking about a sym sgg (I'm gonna start saying it so it catches on), but I don't like the ergodox. I'd like to see where this goes.
I'm gonna start calling it sym stg and hope mine catches on instead.

Offline vivalarevolución

  • Posts: 2146
  • Location: Naptown, Indiana, USA
  • Keep it real b/c any other way is too stressful
Re: IBM M15 Clone - in Cherry
« Reply #7 on: Sat, 22 November 2014, 09:47:02 »
it doesn't matter what he wants.  Convert him over to the light of matrix or symmetrical stagger.  Once you go forth, you can never go back.
Wish I had some gif or quote for this space, but I got nothing

Offline 0100010

  • Thread Starter
  • Posts: 1127
  • Location: DFW, TX, US
  • Not Sure
Re: IBM M15 Clone - in Cherry
« Reply #8 on: Sat, 22 November 2014, 11:03:11 »
Maybe symmetric stagger is an option - but I won't have an opportunity to check until December.

For now, just trying to learn about Teensy,  I/O expander, TRRS cable to connect the halves.
  Quoting me causes a posting error that you need to ignore.

Offline jacobolus

  • Posts: 3661
  • Location: San Francisco, CA
Re: IBM M15 Clone - in Cherry
« Reply #9 on: Sat, 22 November 2014, 14:47:21 »
By the way, if you decide at some point that making something is going to be a pain, this is also a pretty good alternative: http://matias.ca/ergopro/pc/

Offline jacobolus

  • Posts: 3661
  • Location: San Francisco, CA
Re: IBM M15 Clone - in Cherry
« Reply #10 on: Sat, 22 November 2014, 14:50:52 »
But - what I need help with, is how to connect the two halves of the keyboard.  Would the Ergodox solution of a TRRS cable and I/O expander work?  Would it be as simple as using the Ergodox firmware and updating the matrix?  Or would we need a whole new firmware?
Yes, TRRS + I/O expander would work, or I would recommend using a 4P4C connector (like a phone plug) instead of TRRS: the cords and connectors are cheaper, easier to find, and a bit less flaky.

Easier from a construction/firmware perspective though (especially if you hand-wire the matrix), is to just send a ribbon cable across to directly connect the matrix from one half to the microcontroller pins on the other half.
« Last Edit: Sat, 22 November 2014, 14:52:36 by jacobolus »

Offline 0100010

  • Thread Starter
  • Posts: 1127
  • Location: DFW, TX, US
  • Not Sure
Re: IBM M15 Clone - in Cherry
« Reply #11 on: Sat, 22 November 2014, 18:53:47 »
Yeah - I think I've settled on just using a 16 pin ribbon cable to connect the right half of the board (8 columns, 7 rows) to the left half (8 columns, 6 rows).  Will have to see about sleeving or heat shrinking the cable to make it look nicer.
  Quoting me causes a posting error that you need to ignore.

Offline dorkvader

  • Posts: 6288
  • Location: Boston area
  • all about the "hack" in "geekhack"
Re: IBM M15 Clone - in Cherry
« Reply #12 on: Sat, 22 November 2014, 21:01:12 »
I had heard that using an IO expander is probably a worse option than others.

I'm not expert in these things but I do know that the mcp23018 and i2s really limit the offhand of the ergodox when compared with SPI or others.

Offline 0100010

  • Thread Starter
  • Posts: 1127
  • Location: DFW, TX, US
  • Not Sure
Re: IBM M15 Clone - in Cherry
« Reply #13 on: Tue, 25 November 2014, 10:35:48 »
By the way, if you decide at some point that making something is going to be a pain, this is also a pretty good alternative: http://matias.ca/ergopro/pc/

Whoa - missed your suggestion.  Right now he has an old Microsoft Natural rubber dome board which he likes fine.  He uses part of a cardboard box over the numpad area for his mouse.  He also has a Comfort ErgoMagic Keyboard (sitting on my table at the moment) waiting on a couple replacement keys that I ordered for him on ebay for super cheap (even if he hates it, it's only $30 bucks).

Your suggestion of the Matias Ergo Pro looks good - will have to run it by him to see what he thinks. 
  Quoting me causes a posting error that you need to ignore.