Author Topic: GH60 prototype betatesting [Call for layouts, GUI testing]  (Read 115818 times)

0 Members and 1 Guest are viewing this topic.

Offline JPG

  • Posts: 1124
  • Location: Canada (Beloeil, near Montreal)
  • Model F is my new passion!
Re: GH60 prototype betatesting
« Reply #100 on: Wed, 16 January 2013, 15:43:41 »
I recorded two separate prototype soldering, one with hot air only and with iron only.

watching your video(s) right now while i'm working :)

Same here, but more like fast forward style  ;D
IBM F122, IBM XT F X2, IBM AT F (all Soarer converted), Filco Camo TKL Browns

Offline __red__

  • Posts: 194
Re: GH60 prototype betatesting
« Reply #101 on: Wed, 16 January 2013, 20:08:38 »
So,
My oscilloscope shows me that my crystal is oscillating at 50khz as well as 16mhz.  The amplitude of the 50khz is greater than the other by quite a margin.

That's bad.

Also, I borrowed a logic analyzer with built-in protocol decoder and the host attempts to sync but the atmega never responds.

Replacement discretes should arrive tomorrow.  Will post again when I'm working.

Offline mkawa

  •  No Marketplace Access
  • Posts: 6562
  • (ツ)@@@. crankypants
Re: GH60 prototype betatesting
« Reply #102 on: Wed, 16 January 2013, 22:22:43 »
damnit i want a laser-cutter :(

to all the brilliant friends who have left us, and all the students who climb on their shoulders.

Offline komar007

  • Thread Starter
  • Posts: 712
  • Location: Poland
    • komar's blog
Re: GH60 prototype betatesting
« Reply #103 on: Thu, 17 January 2013, 12:49:52 »
A little teaser of the GUI:
11930-0
GH60 rev. B w/ ali's case|Cherry G80-3000 HFU/05|IBM Model M (51G8572)
Check out the GH60 project! | How to make a keyboard

Offline jdcarpe

  • * Curator
  • Posts: 8856
  • Location: Odessa, TX
  • Live long, and prosper.
GH60 prototype betatesting
« Reply #104 on: Thu, 17 January 2013, 12:55:04 »
Loving it!
KMAC :: LZ-GH :: WASD CODE :: WASD v2 :: GH60 :: Alps64 :: JD45 :: IBM Model M :: IBM 4704 "Pingmaster"

http://jd40.info :: http://jd45.info


in memoriam

"When I was a kid, I used to take things apart and never put them back together."

Offline modulor

  • Posts: 231
  • Location: US
  • μ
Re: GH60 prototype betatesting
« Reply #105 on: Thu, 17 January 2013, 15:57:48 »
FC200RR | QFR | GH60

Offline TheProfosist

  • Posts: 3671
  • Location: Wisconsin, USA
  • Custom Layouts Only!
Re: GH60 prototype betatesting
« Reply #106 on: Thu, 17 January 2013, 16:19:35 »
komar are you using hasu's firmware?

Offline komar007

  • Thread Starter
  • Posts: 712
  • Location: Poland
    • komar's blog
Re: GH60 prototype betatesting
« Reply #107 on: Thu, 17 January 2013, 16:20:44 »
No, I have my own.
GH60 rev. B w/ ali's case|Cherry G80-3000 HFU/05|IBM Model M (51G8572)
Check out the GH60 project! | How to make a keyboard

Offline TheProfosist

  • Posts: 3671
  • Location: Wisconsin, USA
  • Custom Layouts Only!
Re: GH60 prototype betatesting
« Reply #108 on: Thu, 17 January 2013, 16:28:03 »
No, I have my own.
does it support all the same stuff that hasu's does other wise it the best option would probably be modifying his

Offline komar007

  • Thread Starter
  • Posts: 712
  • Location: Poland
    • komar's blog
Re: GH60 prototype betatesting
« Reply #109 on: Thu, 17 January 2013, 16:30:49 »
What does hasu's firmware support?
I don't know much about it.


BTW I don't think we need one "official" firmware. The more the merrier;)
« Last Edit: Thu, 17 January 2013, 16:33:53 by komar007 »
GH60 rev. B w/ ali's case|Cherry G80-3000 HFU/05|IBM Model M (51G8572)
Check out the GH60 project! | How to make a keyboard

Offline TheProfosist

  • Posts: 3671
  • Location: Wisconsin, USA
  • Custom Layouts Only!
Re: GH60 prototype betatesting
« Reply #110 on: Thu, 17 January 2013, 16:35:42 »
What does hasu's firmware support?
I don't know much about it.


BTW I don't think we need one "official" firmware. The more the merrier;)
suports media keys, mouse buttons, NKRO, multiple layer both toggle and non

Offline komar007

  • Thread Starter
  • Posts: 712
  • Location: Poland
    • komar's blog
Re: GH60 prototype betatesting
« Reply #111 on: Thu, 17 January 2013, 16:37:45 »
I don't have media keys and mouse buttons yet, but I was planning on doing it anyway.
Mine supports uploading new layout in less than a second, directly from the GUI;)
GH60 rev. B w/ ali's case|Cherry G80-3000 HFU/05|IBM Model M (51G8572)
Check out the GH60 project! | How to make a keyboard

Offline mkawa

  •  No Marketplace Access
  • Posts: 6562
  • (ツ)@@@. crankypants
Re: GH60 prototype betatesting
« Reply #112 on: Thu, 17 January 2013, 17:20:41 »

to all the brilliant friends who have left us, and all the students who climb on their shoulders.

Offline Batmann

  • Posts: 531
  • Location: France
GH60 prototype betatesting
« Reply #113 on: Thu, 17 January 2013, 17:34:54 »
There are some highly skilled people here, this is very inspiring
Keep it up komar! :)

Offline regack

  • Posts: 660
  • Location: Thessia
Re: GH60 prototype betatesting
« Reply #114 on: Thu, 17 January 2013, 17:58:15 »

BTW I don't think we need one "official" firmware. The more the merrier;)

Yes, indeed, the more the merrier!

Offline TheProfosist

  • Posts: 3671
  • Location: Wisconsin, USA
  • Custom Layouts Only!
Re: GH60 prototype betatesting
« Reply #115 on: Thu, 17 January 2013, 19:01:39 »
I don't have media keys and mouse buttons yet, but I was planning on doing it anyway.
Mine supports uploading new layout in less than a second, directly from the GUI;)
yea just though you could focus on the GUI if you used a firmware thats already been developed

Offline JPG

  • Posts: 1124
  • Location: Canada (Beloeil, near Montreal)
  • Model F is my new passion!
Re: GH60 prototype betatesting
« Reply #116 on: Thu, 17 January 2013, 19:26:45 »
A little teaser of the GUI:
(Attachment Link)

Got teased here  :D

I also have 2 questions:

1. Will it be possible to, for exemple, press the capslock to go to a specific layer on the first press, stay on that layer, and come back to the original layer on the second capslock keypress?

2. What do you use for coding language? Will you share the code or make it open source or something like that at some point?

Thx a lot for doing this by the way!!!
IBM F122, IBM XT F X2, IBM AT F (all Soarer converted), Filco Camo TKL Browns

Offline TheProfosist

  • Posts: 3671
  • Location: Wisconsin, USA
  • Custom Layouts Only!
Re: GH60 prototype betatesting
« Reply #117 on: Thu, 17 January 2013, 19:34:56 »
A little teaser of the GUI:
(Attachment Link)

Got teased here  :D

I also have 2 questions:

1. Will it be possible to, for exemple, press the capslock to go to a specific layer on the first press, stay on that layer, and come back to the original layer on the second capslock keypress?

2. What do you use for coding language? Will you share the code or make it open source or something like that at some point?

Thx a lot for doing this by the way!!!
so far everyone that has programmed for that teensy or that chip at least in our community has shared their code I dont know where exactly it is though...

Offline alaricljs

  • I be WOT'ing all day...
  • ** Moderator Emeritus
  • Posts: 3715
  • Location: NE US
Re: GH60 prototype betatesting
« Reply #118 on: Thu, 17 January 2013, 20:13:53 »
Not everyone... Soarer's converter is still closed, and I think there may be a couple others that have very limited exposure that have also chosen not to release.  Several people host on github tho.
Filco w/ Imsto thick PBT
Ducky 1087XM PCB+Plate, w/ Matias "Quiet Click" spring-swapped w/ XM Greens

Offline JPG

  • Posts: 1124
  • Location: Canada (Beloeil, near Montreal)
  • Model F is my new passion!
Re: GH60 prototype betatesting
« Reply #119 on: Thu, 17 January 2013, 20:35:20 »
It's ok if he prefer to keep it for himself, I understand it perfectly, but if he does make it public I would be interested to look at it (and maybe add some features  :D)
IBM F122, IBM XT F X2, IBM AT F (all Soarer converted), Filco Camo TKL Browns

Offline __red__

  • Posts: 194
Re: GH60 prototype betatesting
« Reply #120 on: Thu, 17 January 2013, 23:17:07 »
So,
My oscilloscope shows me that my crystal is oscillating at 50khz as well as 16mhz.  The amplitude of the 50khz is greater than the other by quite a margin.

That's bad.

Also, I borrowed a logic analyzer with built-in protocol decoder and the host attempts to sync but the atmega never responds.

Replacement discretes should arrive tomorrow.  Will post again when I'm working.

I replaced C1, C2, C3, C7, R5 and the Crystal.  It immediately fired up.  I was able to type a few characters so the firmware is cool too.

My 17" LCD for my Propeller Pi laptop is on its way.  Once that's measured and a frame built for it I can then find a small, maybe 7" VGA or Composite Video LCD for the propeller.  Then I can machine and build the frame to support the keyboard.

I'm stoked!

Offline alaricljs

  • I be WOT'ing all day...
  • ** Moderator Emeritus
  • Posts: 3715
  • Location: NE US
Re: GH60 prototype betatesting
« Reply #121 on: Thu, 17 January 2013, 23:22:02 »
Damn those flaky flux capacitors.  Glad you are once again in control of the flow of time.  ;)
Filco w/ Imsto thick PBT
Ducky 1087XM PCB+Plate, w/ Matias "Quiet Click" spring-swapped w/ XM Greens

Offline komar007

  • Thread Starter
  • Posts: 712
  • Location: Poland
    • komar's blog
Re: GH60 prototype betatesting
« Reply #122 on: Fri, 18 January 2013, 01:07:01 »

1. Will it be possible to, for exemple, press the capslock to go to a specific layer on the first press, stay on that layer, and come back to the original layer on the second capslock keypress?
Yes
2. What do you use for coding language? Will you share the code or make it open source or something like that at some point?

Thx a lot for doing this by the way!!!
Coded in C. Most probably yes.



I replaced C1, C2, C3, C7, R5 and the Crystal.  It immediately fired up.  I was able to type a few characters so the firmware is cool too.

My 17" LCD for my Propeller Pi laptop is on its way.  Once that's measured and a frame built for it I can then find a small, maybe 7" VGA or Composite Video LCD for the propeller.  Then I can machine and build the frame to support the keyboard.

I'm stoked!

Great!
I wonder what was wrong... We'll never know most probably.
GH60 rev. B w/ ali's case|Cherry G80-3000 HFU/05|IBM Model M (51G8572)
Check out the GH60 project! | How to make a keyboard

Offline __red__

  • Posts: 194
Re: GH60 prototype betatesting
« Reply #123 on: Fri, 18 January 2013, 05:51:47 »
If i were a betting man, I'd guess wrong values of capacitors across the crystal or a dud crystal.

My human error is more likely than a crystal failing qa imho.



Offline WhiteFireDragon

  • Posts: 2276
    • youtube
Re: GH60 prototype betatesting
« Reply #124 on: Fri, 18 January 2013, 18:26:35 »
Another batch of prototype boards when out yesterday. I'm down to the last few before I build mine, after that the real testing can begin for me.


If i were a betting man, I'd guess wrong values of capacitors across the crystal or a dud crystal.

My human error is more likely than a crystal failing qa imho.

If it was indeed it was a defective crystal, I'm just glad that one defective part whet to you haha. If it was one of the 12 that I had to solder, I wouldn't have known what to do. At least you have an oscilloscope to test it. I only have a DMM. I'm glad you got it all figured out.

Offline TheProfosist

  • Posts: 3671
  • Location: Wisconsin, USA
  • Custom Layouts Only!
Re: GH60 prototype betatesting
« Reply #125 on: Fri, 18 January 2013, 19:39:05 »
Another batch of prototype boards when out yesterday. I'm down to the last few before I build mine, after that the real testing can begin for me.


If i were a betting man, I'd guess wrong values of capacitors across the crystal or a dud crystal.

My human error is more likely than a crystal failing qa imho.

If it was indeed it was a defective crystal, I'm just glad that one defective part whet to you haha. If it was one of the 12 that I had to solder, I wouldn't have known what to do. At least you have an oscilloscope to test it. I only have a DMM. I'm glad you got it all figured out.
i dont think i paid for mine yet... could you send me an invoice?

Offline WhiteFireDragon

  • Posts: 2276
    • youtube
Re: GH60 prototype betatesting
« Reply #126 on: Sat, 19 January 2013, 13:10:40 »
New switch hole design:

   

#2 - this is the design I have on the new phantom plates and the the TKL plates GB I ran.
#3 - some switches on this PCB needs to be rotated 90, so this is just basically a 90 rotation of #2
#1 - a more crazy design that can be put on every switch that allows rotation in any direction


All these are designed to allow opening of the switches without desoldering. Most switches on the GH60 only needs #2, and some needs it to be rotated like #3. If all the switch holes were to be uniform, then #1 design allows any mounting orientation. This would be good for future PCBs to be designed without needing a specific plate to tailor to the specific PCB. If quotes are not much more for the extra details of the #1 design, then probably that design would be best. Comments/opinions before I start drawing up the different layouts?

Offline jdcarpe

  • * Curator
  • Posts: 8856
  • Location: Odessa, TX
  • Live long, and prosper.
Re: GH60 prototype betatesting
« Reply #127 on: Sat, 19 January 2013, 13:22:51 »
I like #1, for reasons you already know. :)
KMAC :: LZ-GH :: WASD CODE :: WASD v2 :: GH60 :: Alps64 :: JD45 :: IBM Model M :: IBM 4704 "Pingmaster"

http://jd40.info :: http://jd45.info


in memoriam

"When I was a kid, I used to take things apart and never put them back together."

Offline IvanIvanovich

  • Mr. Silk Underwear
  • Posts: 8199
  • Location: USA
Re: GH60 prototype betatesting
« Reply #128 on: Sat, 19 January 2013, 13:35:00 »
Have any one else consider underplate? I make some G80 mod like this with MY switch plate. Only need cut to dimension along with thin rubber mat and a few holes to allow the mounting bolts to pass through. Would give it a more sturdy feeling removing flex when use stock plastic case but without taking away from the other pros of pcb mount feel. Only would need one design too, very easy.

Offline relcc

  • Posts: 18
  • Location: Amsterdam, the Netherlands
Re: GH60 prototype betatesting
« Reply #129 on: Sun, 20 January 2013, 01:23:06 »
#1 would't work because the clip that keeps the switch in place is exactly where the rotated gaps are.

#2 and #3 give the switch less grip than holes with 2 gaps on each side.

I would recommend conventional holes with the conventional mod for top removal.

The ideal situation would also be where the plate is connected to the case, and not the PCB. That means small holes for the screws, not holes to reach for the screws that are actually on the pcb. For this board however that would probably not be feasible.
« Last Edit: Sun, 20 January 2013, 01:28:57 by relcc »

Offline WhiteFireDragon

  • Posts: 2276
    • youtube
Re: GH60 prototype betatesting
« Reply #130 on: Sun, 20 January 2013, 17:50:33 »
#2 and #3 have all 4 corners and the whole top and bottom to lock in the switch. It's not moving anywhere. These can only be mounted in 2 directions though. I have 40 TKL plates with this design.

#1 would work. It's an overlay of #2 and #3, and I specified the middle tabs to be just long enough so top and bottom of the switch can lock no matter what orientation you mount it. All 4 corners also hold the from any movement.


I'm not sure you fully understand how those hole cutouts work. What exactly do you have in mind for "conventional" holes?

Offline TheProfosist

  • Posts: 3671
  • Location: Wisconsin, USA
  • Custom Layouts Only!
Re: GH60 prototype betatesting
« Reply #131 on: Sun, 20 January 2013, 18:01:45 »
1 should work..

Offline relcc

  • Posts: 18
  • Location: Amsterdam, the Netherlands
Re: GH60 prototype betatesting
« Reply #132 on: Sun, 20 January 2013, 19:55:40 »
It's a trade-off between a situation where the key barely locks and is barely prevented from moving on one hand, and where the key is firmly locked into place with maximum support on all sides. If the lasercutting (or worse cnc) screws up a corner, the key is only supported by pcb and the plate is (for that key) useless. If you're using pcb mount keys you have some slack, but plate mounted keys leave no room for error.

Maybe you should run a prototype.

Offline regack

  • Posts: 660
  • Location: Thessia
Re: GH60 prototype betatesting
« Reply #133 on: Sun, 20 January 2013, 20:26:23 »
I like #1 - Even if the key doesn't lock into the plate, size it correctly and there will be enough friction to keep it in there very tightly. 

Offline hasu

  • Posts: 2964
  • Location: Tokyo, Japan
  • @tmk
    • tmk keyboard firmware project
Re: GH60 prototype betatesting
« Reply #134 on: Sun, 20 January 2013, 20:49:31 »
With tinkering with The_Beast's Poker Plate with MX I found MX has different width at its corner.
My cheap caliper says horizontal width is 13.80mm while 12.95mm at vertical.
This difference causes kind of wiggle when you place MX at 90deg or 270deg rotation in Poker Plate. Unfortunately Poker Plate has a bit too short width of middle tabs to hold at firm part of bottom case not only at lock spring.
Not to mention, Poker Plate works well with proper rotation of switch. This is not a complain about Poker Plate.

Though 4 corners of #1 is not enough to hold firmly, with proper middle tabs #1 would work.


#2 and #3 have all 4 corners and the whole top and bottom to lock in the switch. It's not moving anywhere. These can only be mounted in 2 directions though. I have 40 TKL plates with this design.

#1 would work. It's an overlay of #2 and #3, and I specified the middle tabs to be just long enough so top and bottom of the switch can lock no matter what orientation you mount it. All 4 corners also hold the from any movement.


I'm not sure you fully understand how those hole cutouts work. What exactly do you have in mind for "conventional" holes?
TMK products:HHKB Alt  ⌨ConvertersAlps64FC660C AltFC980C Alt

Offline WhiteFireDragon

  • Posts: 2276
    • youtube
Re: GH60 prototype betatesting
« Reply #135 on: Mon, 21 January 2013, 00:31:07 »
With tinkering with The_Beast's Poker Plate with MX I found MX has different width at its corner.
My cheap caliper says horizontal width is 13.80mm while 12.95mm at vertical.
This difference causes kind of wiggle when you place MX at 90deg or 270deg rotation in Poker Plate. Unfortunately Poker Plate has a bit too short width of middle tabs to hold at firm part of bottom case not only at lock spring.
Not to mention, Poker Plate works well with proper rotation of switch. This is not a complain about Poker Plate.

I know exactly where you measured it, and those are the wrong places. Move it up a tad. My digital caliper says 13.97mm, which is pretty much spot on with cherry's specs of 14.0mm at those areas. Also, according to the cherry's dimensions, the switch is a perfect square at the areas held in by the plate. Each side of the hole is 0.55" (14mm).

The poker plate you played around with have the exact same switch cutouts as the original phantom. You can rotate the switch 90° and it will still hold it in place fine, but you'll lose option to open up the switch. If you fully mount it where it locks down, it's sturdy and won't move. Here you can see the plate cutout that cherry specified, which is how the phantom switch and stabilizer holes were derived:

http://www.cherrycorp.com/english/switches/key/mx.htm
http://www.cherrycorp.com/english/switches/pdf/mx_cat.pdf


Offline WhiteFireDragon

  • Posts: 2276
    • youtube
Re: GH60 prototype betatesting
« Reply #136 on: Mon, 21 January 2013, 00:33:52 »
I like #1 - Even if the key doesn't lock into the plate, size it correctly and there will be enough friction to keep it in there very tightly.
It's a trade-off between a situation where the key barely locks and is barely prevented from moving on one hand, and where the key is firmly locked into place with maximum support on all sides. If the lasercutting (or worse cnc) screws up a corner, the key is only supported by pcb and the plate is (for that key) useless. If you're using pcb mount keys you have some slack, but plate mounted keys leave no room for error.

Maybe you should run a prototype.

I've looked at both laser cutting and waterjet to cut the holes. Water jet won't be able to do fine details, but laser cutting has good enough tolerances to cut the plates. The 4 corners as well as the 4 center tabs are actually touching the switch to hold it in place.

I've already prototyped and made a stack of TKL plates with hole design #2/3. I'll promise you #1 hole design will be fine :) . It's not really a question of whether it will work or not, but which design is preferred for these new plates.


Offline kmiller8

  • Banned
  •  Post Reporting Timeout
  • Posts: 1589
  • Who is that kmiller8 guy?
Re: GH60 prototype betatesting
« Reply #137 on: Mon, 21 January 2013, 00:35:54 »
With tinkering with The_Beast's Poker Plate with MX I found MX has different width at its corner.
My cheap caliper says horizontal width is 13.80mm while 12.95mm at vertical.
This difference causes kind of wiggle when you place MX at 90deg or 270deg rotation in Poker Plate. Unfortunately Poker Plate has a bit too short width of middle tabs to hold at firm part of bottom case not only at lock spring.
Not to mention, Poker Plate works well with proper rotation of switch. This is not a complain about Poker Plate.

Though 4 corners of #1 is not enough to hold firmly, with proper middle tabs #1 would work.

hasu, I think you were measuring at the wrong point, with my cheap calipers I got 13.57mm x 13.70mm :x

On the sides of the housing with clips


On the sides without the clips


I also replicated your result of 12.70mm

On the side without the clips, measured at a different point


The 12.70mm is measured from the "bottom" of the switch, while the 13.70 is measured from where the plate comes in contact with the housing, as shown below.

Where I measured 13.70mm


As long as you make that middle portion on #1 around 6.48mm, it should be able to both hold the switch in the vertical and the horizontal :)

The length of the 13.70mm portion of the switch, that won't interfere with the clips on the housing

Offline daerid

  • Posts: 4271
  • Location: Denver, CO
    • Rossipedia
Re: GH60 prototype betatesting
« Reply #138 on: Mon, 21 January 2013, 01:32:16 »
Is there an official GB thread for this?

Offline TheProfosist

  • Posts: 3671
  • Location: Wisconsin, USA
  • Custom Layouts Only!
Re: GH60 prototype betatesting
« Reply #139 on: Mon, 21 January 2013, 01:56:10 »
Is there an official GB thread for this?
no yeat the board is still in betatesting

Offline hasu

  • Posts: 2964
  • Location: Tokyo, Japan
  • @tmk
    • tmk keyboard firmware project
Re: GH60 prototype betatesting
« Reply #140 on: Mon, 21 January 2013, 03:53:43 »
My point is that four corners doesn't touch switch case perfectly and we need proper middle tabs to get better support for switch in #1 design.
You'll see small gap at bottom corner and switch case in photos below. On left side large gap is by design, of course.
On the other hand this gap is never problem in #2, #3 and Poker plate with normal rotation switch.

Left bottom corner of switch with normal rotation.


with 90deg rotation. Poker Plate's tab is a bit short and can't support rotated switch firmly.


Normal and rotated mount side by side.


[EDIT] replaced photos with a bit better one and added some photos below.

Note that corners can't hold switch sturdily in #1 design as this photo indicates.
Never mind if you already found this yourself. I just wanted to let you guys know this.

This short tab holds switch only at part of plate lock spring. This causes horizontal wiggle.
If it is a bit longer it will hold sturdily at firm part of switch case.


I think we can have a bit longer tab than Poker Plate without interfering switch open.

« Last Edit: Mon, 21 January 2013, 19:21:22 by hasu »
TMK products:HHKB Alt  ⌨ConvertersAlps64FC660C AltFC980C Alt

Offline WhiteFireDragon

  • Posts: 2276
    • youtube
Re: GH60 prototype betatesting
« Reply #141 on: Mon, 21 January 2013, 22:41:25 »
You’re basing all your fittings and pictures on the poker/phantom plate design. But you can’t really make the assumption that these new hole design is exactly the same as the poker/phantom plate, which they’re not. The corner notches on my design is actually a tad bigger, combined with the whole top and bottom edges, that side flap was not needed. Refer to THIS post.




The switch holes here are the holes from Figure A. Those corners are as long as possible. If they’re any longer, you won’t be able to open the switch. I do know what you mean by that extra tiny gap next to the corner though, but that gap doesn’t affect anything. The corner alone holds the switches firmly, but the two top and bottom sides also add support. The switch can’t be moved at all. This obviously won’t work if you rotate the switch 90degrees, because that plastic lock spring has nothing to clamp on, and the switch tabs can’t be opened. This is why Fig. B is necessary for a rotated switch, which is basically Fig. A, just rotated 90degrees.



Now if you look at Fig. C, it's just an overlap of A and B, and that's why I know the final design will work. The corners as well as the solid blue line at the center of each side hold the switch in place. That solid blue line is actually touching the switch to clamp it down, and it's just long enough to still allow you to open up the switch tabs. If you darkened all the important lines and got rid of all the reference lines from Fig C, then you get Figure D, which is the final design of the switch hole that will allow: 1) any rotation orientation of the switch, 2) opening of the switch, and 3) clamp the switch in place and for the top/bottom plastic clips to lock.


Man this is pretty hard to explain and to point out which parts I'm referring to, but hopefully it's coherent. lol does anyone actually understand me?

Offline alaricljs

  • I be WOT'ing all day...
  • ** Moderator Emeritus
  • Posts: 3715
  • Location: NE US
Re: GH60 prototype betatesting
« Reply #142 on: Mon, 21 January 2013, 22:50:20 »
I do, but that doesn't count... :)
Filco w/ Imsto thick PBT
Ducky 1087XM PCB+Plate, w/ Matias "Quiet Click" spring-swapped w/ XM Greens

Offline hasu

  • Posts: 2964
  • Location: Tokyo, Japan
  • @tmk
    • tmk keyboard firmware project
Re: GH60 prototype betatesting
« Reply #143 on: Tue, 22 January 2013, 00:37:31 »
I didn't say your design won't work and I think I understand your design intention.

In design D four corner notches can hold right/left side of switch but not top/bottom side,
so tabs are important in this design, right?
BTW, Poker plate seems to has same size notches as yours in the photo.

My concern is only width of four middle tabs. I found shorter width tab won't hold switch firmly, because it touch only at plastic clip part.

I think middle tab should be long enough to reach solid part of switch case. maybe 6.00mm or so? as far as it doesn't interfere opening switch.
TMK products:HHKB Alt  ⌨ConvertersAlps64FC660C AltFC980C Alt

Offline WhiteFireDragon

  • Posts: 2276
    • youtube
Re: GH60 prototype betatesting
« Reply #144 on: Tue, 22 January 2013, 01:46:11 »
Yeah, those tabs in the middle of every side is absolutely necessary in the design so that the two plastic locking clips have something to clamp on. I specified the width of those tabs to be 5.10mm. You really only need 4.25mm for them to properly work. I took tolerance into account, so 6.00mm is probably too close to the switch clips that allow opening of the switches, 5.10mm is a good intermediate.

Anyways, I'll stop ranting about this. I'll move this discussion to the main thread if there is anything else to say since this thread is specifically for actual testing of prototypes.

Offline hasu

  • Posts: 2964
  • Location: Tokyo, Japan
  • @tmk
    • tmk keyboard firmware project
Re: GH60 prototype betatesting
« Reply #145 on: Tue, 22 January 2013, 02:43:18 »
I'll also stop posting about this :) Sorry for derail, guys.

Hmm, looks like I couldn't convey what I want to say. Or I'm completely confusing about this.
5.10mm is almost same as Poker's size.
I still think switch will wiggle along top/bottom direction with that width tab.
TMK products:HHKB Alt  ⌨ConvertersAlps64FC660C AltFC980C Alt

Offline The_Beast

  • * Maker
  • Posts: 3964
  • Location: Wisconsin
  • I like wood ಠ_๏
Re: GH60 prototype betatesting
« Reply #146 on: Tue, 22 January 2013, 14:16:27 »
Can someone give me the exact size of the PCB (L x W x H) and where the center of the USB port is?
Vendor Status: Sadly, not taking any orders/pre-orders at this time

Vendor Quick Links: | Vendor Forum | Hardwood Wrist Rests | Hardwood 60% Cases | Customer Gallery | Giveaway |

Offline komar007

  • Thread Starter
  • Posts: 712
  • Location: Poland
    • komar's blog
Re: GH60 prototype betatesting
« Reply #147 on: Tue, 22 January 2013, 14:20:22 »
285mm x 94mm, USB port's center is 17.57mm from the left.

The vertical dimension may change slightly, because the stabilizer holes are too close to the edges now.
GH60 rev. B w/ ali's case|Cherry G80-3000 HFU/05|IBM Model M (51G8572)
Check out the GH60 project! | How to make a keyboard

Offline The_Beast

  • * Maker
  • Posts: 3964
  • Location: Wisconsin
  • I like wood ಠ_๏
Re: GH60 prototype betatesting
« Reply #148 on: Tue, 22 January 2013, 14:21:59 »
285mm x 94mm, USB port's center is 17.57mm from the left.

Thanks, how thick is the PCB and how thick is the PCB w/ the USB?
Vendor Status: Sadly, not taking any orders/pre-orders at this time

Vendor Quick Links: | Vendor Forum | Hardwood Wrist Rests | Hardwood 60% Cases | Customer Gallery | Giveaway |

Offline Acetrak

  • ssk.ℜ
  • Posts: 1079
  • Location: East
  • [Ctrl][Alt][Fn`]
Re: GH60 prototype betatesting
« Reply #149 on: Tue, 22 January 2013, 14:22:09 »
^^Quick and precise, I like it :D
(ノ◕ヮ◕)ノ*:・゚✧ HHKB 55g Type-S | LZGH 65g Blues | IBM SSKawa