Author Topic: [IC] Omega 60 - By Windeh  (Read 6779 times)

ptiede, Ephemeral, juaninamilli, tentboy, Furikurichemy, Jecxz, Kasterborous and 8 Guests are viewing this topic.

Online Windeh

  • Thread Starter
  • Posts: 270
  • a no name
[IC] Omega 60 - By Windeh
« on: Sun, 27 January 2019, 09:39:24 »
______________________________________________________________
  • 27th Jan 2019
Interest Check Form - Please fill in
______________________________________________________________


211476-0


This is the Interest Check thread for the Omega 60, a Floating Bagel Sandwich Mounted 60% Custom Keyboard kit.

The mount for this board will be identical to my last 2 designs, Reflex & Paradox. For more info on the mounting system, please check out the link below:

https://geekhack.org/index.php?topic=97859.msg2672186#msg2672186

This IC will focus on updates and changes from my last design, while addressing some skepticism on the smol brane spacebar cutout for the plate.

Having experiences from designing the Reflex & Paradox, and with several prototypes made, I focused on maintaining the essence of the original designs, while streamlining and refocusing the overall package.


Design Highlights:

  • Weight - The internal weight have been maximized within the allowable space, with a fully built weight of 2.2kg (4.85lbs), in a relatively compact and low profile footprint.
  • Mount & Plate - Mounting and plate designs are unchanged, as the design works wonderfully.
  • Chassis rigidity - With additional material in key areas, without exposed weight or cutouts, and tighter tolerance between the case top and bottom, together with the increased weight and small size, the board feels incredibly substantial and rigid in hand and whilst typing.
  • Ease of assembly - Putting together and taking apart the Omega 60 is effortless, with only 10 screws to assemble, and only 6 are structural. All 10 screws are of the same standard size and length, so no mistaken screw lengths, and just 1 bit size.
  • Reduced bezel gap


Objective - To hide the screw holes while keeping the rear angle

Feel free to skip this section if you don't care for the design of this board.

More
Looking at how past keyboard designers dealt with screw holes, some chose to hide them, and are limited by the location of the screw hole in terms of how much rear angle they can play with, while others choose the opposite path, and proudly design around the exposed screws for a more industrial looking board.

I wanted to keep the angled look of my first 2 designs, while hiding the screw holes. The Reflex had exposed screw holes, and the style suited the "louder" design. The Paradox used weights to hide the screw holes, however the design was expensive to manufacture and to fit correctly, and it requires you to remove the weight before being able to remove the case top screws, increasing assembly time & complexity.

I thought to use a nameplate to cover the screw holes, however was uncertain on how to keep the plate in place, while allowing easy removal for quick access. Screws was of course the obvious choice, however this defeats the purpose of hiding the screw holes in the first place.

Magnets comes to mind, however the nameplate material is not magnetic, and to stick a magnet onto the plate using epoxy, then stick another piece of magnetic material onto the case would not be very elegant. How to then remove the plate is another concern.

I ultimately came up with the "clamp" design. The plate is simply dropped into the recess, and the 3 clamping screws are tightened to secure the name plate. The tolerances of the nameplate must be very accurate to not cause excessive deformation of the clamping mechanism, so each name plate will be hand fitted to their respective cases.

211480-1

To remove the nameplate, simply slightly loosen the 3 clamping screws, and the plate will be free to drop out.

While the tolerances between the nameplate and the plate recess will be kept to less then 0.2mm, the clamping mechanism can accommodate up to 0.5mm of deformation without damage, and will return back their original position upon loosening.





The "Smol Brane" spacebar plate cut

There are some discussions within various Discord channel on the questionable design decision to place a mounting point directly below the spacebar, then to try and fix the stiffness by disconnecting the spacebar switch from that mounting point with a plate cut, potentially increasing complexity of the design, and ultimately the cost to manufacture.

I will break down how this design came to be, with some Displacement (Flex) Simulation, uniform 10N on surface / Rigid mounting points:

Goal for plate design -
  • Reduce rigidity of the spacebar switch plate position
  • Maintain overall uniform switch sound and feel (flex) across the Alphas and the Spacebar area of the plate.
  • Have the least amount of complexity to achieve the above goals.

Simulation 1 - Full mounting position, no disconnect cut
More
211490-3

While the alphas are "somewhat" uniform in their displacement, the spacebar switch position are super rigid, as is expected.


Simulation 2 - Full Mounting Position, WITH disconnect cut
More
211492-4

The disconnect cut successfully reduced the rigidity of the spacebar switch position, while also reliefs the center portion of the plate for more uniform displacement across almost the entire Alphas area of the plate.


Simulation 3 - Remove the bottom center mounting position entirely
More
211494-5

Without the center bottom mounting position, the plate experiences vastly inconsistent displacement, with much more flex closer to the spacebar cutout, and stiffens out across the whole plate towards the mounting points.

I hope with the above, somewhat simplistic simulations, that we can agree that a simple cut which adds about 0.5 dollar to the costs of the plate, is the most cost effective solution to the rigid spacebar problem, without negatively affecting the consistency of the typing experience.


** Whilst I'm not trying to simulate every detail from the switches, keycaps, stabilizers and every component on the PCB etc, however given that those are pretty evenly spread out across the plate and are consistent, with the only variable being the plate design, the simulation serves its purpose to demonstrate and gives a visualization of the areas of relative flexibility between different plate designs. **


Spec:
  • Case material: 6061 / 5052 Aluminium (TBC)
  • Weight material: Sandblasted & Clearcoated Brass
  • Plate material: Anodized Aluminium in Deep Red or Royal Blue / Sandblasted Brass (Optional)
  • Screws: YFS Grade 12.9 Alloy Steel Hex Black Oxide Screw AND THE SS 316 A4 Stainless Steel Screws - Only the best fasteners
  • Color Options: Standard anodization Black / Dark Grey / Deep Red / Navy Blue
  • Nameplate Material: Brushed 304 Stainless Steel - NAMEPLATE LOGO LASER MARKING DESIGN TO BE CONFIRMED
  • Fully built weight with Brass Plate: ~2.2kg (4.85lbs)
  • Layout: R1 will be unblocked top only - Tsangan layout with optional split backspace & stepped caps lock
  • Angle: ~7 Degrees
  • PCB: Not included. Satan GH60 PCB can be purchased for $38 with GB, or $30 from KBDfans
  • Base kit price: $350 for Paradox/Reflex participants, otherwise $380 (4% PayPal Fee & Delivery cost not included)
  • R1 Slots: 18 Sets
  • GB Date: Sometime after CNY


Feel free to comment below, let me know if I've missed anything, and do discuss your concerns with my design here, as it can be hard to catch up with your comments in the Discord channel.  ;)


Extra photos and renders:

More
211482-6

211484-7



211488-9


211474-10
« Last Edit: Tue, 29 January 2019, 23:45:00 by Windeh »

Online ArchDill

  • * Esteemed Elder
  • Posts: 1065
  • Location: OK
Re: Omega 60 - By Windeh
« Reply #1 on: Sun, 27 January 2019, 09:42:30 »
This thing looks pretty rad

Offline airlangga07

  • Posts: 68
Re: Omega 60 - By Windeh
« Reply #2 on: Sun, 27 January 2019, 09:46:52 »
This time I definitely join! Good Luck with the IC!

Offline Rensuya

  • Posts: 54
  • Location: NC
  • If there's brass on the field, bottom out!
    • Where I post all things nerdy and keeb related if anyone is interested. Profile image credit to Owler.
Re: Omega 60 - By Windeh
« Reply #3 on: Sun, 27 January 2019, 09:47:08 »
Nice

Sent from my Pixel 2 XL using Tapatalk


Offline garbo

  • Posts: 88
Re: Omega 60 - By Windeh
« Reply #4 on: Sun, 27 January 2019, 10:10:30 »
Good IC post, super thorough as always.

The computational approach to looking at plate flex is a great idea.
What's your process for running those? If it's something that other designers could start to adopt, it could really take some of the trial and error out of plate design.
Do your models also account for the effect of the o-ring burgers(/bagels)?

Offline kawasaki161

  • Posts: 1044
  • Location: Germany
Re: Omega 60 - By Windeh
« Reply #5 on: Sun, 27 January 2019, 10:29:12 »
Do your models also account for the effect of the o-ring burgers(/bagels)?
He said the simulations ran with fixed mounting points, but those aren't even the main problem.
 
You'll never use the plate on it's own, you always have switches in it and a PCB underneath. Unless those are part of the simulation it's worthless.

Online Windeh

  • Thread Starter
  • Posts: 270
  • a no name
Re: Omega 60 - By Windeh
« Reply #6 on: Sun, 27 January 2019, 10:37:55 »
Do your models also account for the effect of the o-ring burgers(/bagels)?
He said the simulations ran with fixed mounting points, but those aren't even the main problem.
 
You'll never use the plate on it's own, you always have switches in it and a PCB underneath. Unless those are part of the simulation it's worthless.

No data is worthless. This gives a general idea for the plate in a vacuum, and it is helpful to visualize the affects of relief cuts and mounting positions for the plate.

Or you can hand wire without a PCB... so yes, you can use the plate on its own, should you choose to, no need to be so negative.  ;)

Online juaninamilli

  • Posts: 77
Re: Omega 60 - By Windeh
« Reply #7 on: Sun, 27 January 2019, 10:43:51 »
This one is a no brainer.  The only thing that will stop me from getting this board is the limited quantities.  Great job again!

Offline garbo

  • Posts: 88
Re: Omega 60 - By Windeh
« Reply #8 on: Sun, 27 January 2019, 10:52:41 »
Do your models also account for the effect of the o-ring burgers(/bagels)?
He said the simulations ran with fixed mounting points, but those aren't even the main problem.
 
You'll never use the plate on it's own, you always have switches in it and a PCB underneath. Unless those are part of the simulation it's worthless.

Oh yeah, missed that.

You have a point about the limits to the way it's modelled, but I still think that the deformation of an unpopulated plate is indicative enough of the way it would perform with the other parts (at least as a general suggestion of evenness, not saying there's anything quantitative to learn from it).

Online tex_live_utility

  • Posts: 269
Re: Omega 60 - By Windeh
« Reply #9 on: Sun, 27 January 2019, 11:09:54 »
How steep is the typing angle?

Offline Vigrith

  • Posts: 1395
Re: Omega 60 - By Windeh
« Reply #10 on: Sun, 27 January 2019, 11:15:15 »
Classic 60% and the actual aesthetics of the case are not for me but the thought you put into your designs is really interesting and definitely appreciated - it's very impressive to see and I look forward to the possibility of supporting you in one of your future projects.

Online Windeh

  • Thread Starter
  • Posts: 270
  • a no name
Re: Omega 60 - By Windeh
« Reply #11 on: Sun, 27 January 2019, 11:19:02 »

Offline Coolbeanz

  • Posts: 14
Re: Omega 60 - By Windeh
« Reply #12 on: Sun, 27 January 2019, 11:30:18 »
Would love to see more colour options especially white/e-white!

Online beekey

  • Posts: 173
Re: Omega 60 - By Windeh
« Reply #13 on: Sun, 27 January 2019, 11:52:53 »
Plate option with split space please. Like the design a lot.

Offline pixelpusher

  • Formerly known as reececonrad
  • Posts: 2530
  • Location: Tennessee - USA
Re: Omega 60 - By Windeh
« Reply #14 on: Sun, 27 January 2019, 13:12:55 »
love the side view.  The bezels don't look right to me though.  Maybe in time it will grow on me.
:)

Offline pr0ximity

  • Posts: 2310
  • Location: Vacationland
Re: Omega 60 - By Windeh
« Reply #15 on: Sun, 27 January 2019, 13:32:46 »
Did you write a simulation yourself or is there something one could plug a plate design into and run themselves?

Online Fabi

  • Posts: 14
Re: Omega 60 - By Windeh
« Reply #16 on: Sun, 27 January 2019, 14:29:27 »
Did you write a simulation yourself or is there something one could plug a plate design into and run themselves?
I think he (or she) is using Fusion 360 which has a module for simulations like these.
Please correct me if I'm wrong.

Offline Remsky

  • Posts: 221
  • Location: Portland, OR
Re: Omega 60 - By Windeh
« Reply #17 on: Sun, 27 January 2019, 17:22:32 »
I see why you have screw holes in the corners (I presume for alignment), but since this is a sandwich style mount, it will contribute towards the pressure put onto the plate. IMO remove them (saves  money spent on threading as well and makes 4 less screws that could strip) and also remove the bottom center mounting post. Even with the cut, having a mounting point directly beneath the spacebar switch will probably make the spacebar sound a bit stiffer instead of letting it flex a little bit more for that thock. You could also do a near row wide relief cut in between the numrow and QWERTY row to even out the flex on the alphas (with the current 5 point mount plate design anyway). to make the flex more consistent.

Online Windeh

  • Thread Starter
  • Posts: 270
  • a no name
Re: Omega 60 - By Windeh
« Reply #18 on: Sun, 27 January 2019, 18:16:17 »
I see why you have screw holes in the corners (I presume for alignment), but since this is a sandwich style mount, it will contribute towards the pressure put onto the plate. IMO remove them (saves  money spent on threading as well and makes 4 less screws that could strip) and also remove the bottom center mounting post. Even with the cut, having a mounting point directly beneath the spacebar switch will probably make the spacebar sound a bit stiffer instead of letting it flex a little bit more for that thock. You could also do a near row wide relief cut in between the numrow and QWERTY row to even out the flex on the alphas (with the current 5 point mount plate design anyway). to make the flex more consistent.

Thanks for the detailed suggestion! However I think there is some misunderstanding on how my mounting system actually holds the plate. I have a video or two for you later today when I'm back from work.  :thumb:

Online Windeh

  • Thread Starter
  • Posts: 270
  • a no name
Re: Omega 60 - By Windeh
« Reply #19 on: Mon, 28 January 2019, 08:37:51 »
I see why you have screw holes in the corners (I presume for alignment), but since this is a sandwich style mount, it will contribute towards the pressure put onto the plate. IMO remove them (saves  money spent on threading as well and makes 4 less screws that could strip) and also remove the bottom center mounting post. Even with the cut, having a mounting point directly beneath the spacebar switch will probably make the spacebar sound a bit stiffer instead of letting it flex a little bit more for that thock. You could also do a near row wide relief cut in between the numrow and QWERTY row to even out the flex on the alphas (with the current 5 point mount plate design anyway). to make the flex more consistent.

Corner screws

The corner screws are not actually used to align the top & bottom case, as the top self-aligns during installation;


Those screws are there to ensure that the corners of the case are tightly clamped, so there are no visible gaps. This "Feature" can be seen in most custom keyboards. My first Reflex/Paradox protos omitted the corner screws, which were added in subsequent updated design, for no other reason then to ensure a good tight gap-less fit. I would have loved to simplify the design with less screws, but felt that these are necessary for a premium product finish.

Screw stripping is an issue I have occasionally seen in the community, stemming from the use of tiny fasteners, sub-par drivers, sub-par screw grade, soft materials, shallow thread depth etc.. That's why I chose to use the largest screw size I can fit into the design (M3 / M4), YFS/THE branded fasteners, and Grade 12.9 alloy steel & A4 SS316 Stainless screw material. These fasteners are the most trusted in the industry, with accurate Hex Bit size and material strength, to minimize chances of stripping the screw socket. Both stainless & alloy steel screws will come with the kit.

211534-0

211530-1

Rule of thumb for thread depth are that a full 2.5-3 thread engagement will be sufficient for the application, as those first few threads takes most of the load. All my designs are around 10 thread engagement.

Lastly regarding pressure on the plate from the corner screws, the following section analysis and video will hopefully clarify how the mounting system functions:

211532-2

From the corner screw section analysis, you can see that the plate does not touch the case at all.

Below video shows what happens if you do not install ANY O-Ring, with all screws tightened:


As the video hopefully demonstrates, the plate in my design wholly relies on the O-Rings to function where it is kept "floating" 0.5mm away from any hard case surface, and without them, the plate is free to move about.


Bottom Center Mount Post

Since you took the time to offer your suggestion, I will also offer my time to simulate your plate design suggestion.  :D

Let's call it the Remsky Special:

211536-3

Hopefully this plate design is what you described.

At first glance, it does seem that the "Row Wide Relief Cut" created a very large displacement area for the alphas. However, it is still inconsistent displacement, where the Alphas near the center of the board have more flex, and gradually become stiffer towards the two sides. Also, as this is not a traditional top mount with rigid mounting points, too much flex may cause the PCB to bottom out within the case.

In reality, using a PCB that have no relief cut which mirrors the plate (specifically a relief cut that separates portions of keys, i.e. a cut NOT on the edges of the plate), the effect of the cut significantly diminishes, as the PCB is now the bridge linking the 2 sides of the cut, preventing most of the flex. The effect of a PCB relief cut with a correspondingly "Reliefed" (half) plate, results in a hilarious amount of flex, as can been seen in Nathan Kim's GSKT.


With my current design, there are no PCB to bridge the cut, and the cut works wonderfully to disconnect the space bar switch from the mounting post, creating that "Thock" without having to sacrifice the alpha zones with inconsistent sounding and feeling keys.



Hopefully with the above explanation, you can follow my thought process for these design choices, and we can at least agree that some thought have been given to create this keyboard.  :thumb:

Please do let me know if you have any questions or suggestions, or would like to offer any counterpoint to what I have written above. I do thoroughly enjoy these technical discussions!  :D
« Last Edit: Mon, 28 January 2019, 08:54:56 by Windeh »

Online Windeh

  • Thread Starter
  • Posts: 270
  • a no name
Re: Omega 60 - By Windeh
« Reply #20 on: Mon, 28 January 2019, 09:15:00 »
By the way, If any participants would like to have the Remsky Special plate made themselves, I will send them the plate file.  :cool:

Not sure if Remsky would want royalties though...

Offline Unforgivable

  • Posts: 635
Re: Omega 60 - By Windeh
« Reply #21 on: Mon, 28 January 2019, 09:46:37 »
Windeh,

The amount of work you place in all your creations is amazing. Id totally be in if R1 had a HHKB top, but sadly I may have to wait until R2. Regardless, Im in whenever that passes by (Hopefully shortly after R1?).

Offline lac29

  • Posts: 108
Re: Omega 60 - By Windeh
« Reply #22 on: Mon, 28 January 2019, 11:00:21 »
Some really thoughtful design choices that you rarely see in these ICs. I'm not interested in a 60% but if you do a 65% or or 75% I'd be in I think.

Offline hineybush

  • * Maker
  • Posts: 655
  • Location: Dayton, OH, USA
  • too many ideas, not enough money
    • hineybush keyboards
Re: Omega 60 - By Windeh
« Reply #23 on: Mon, 28 January 2019, 17:53:29 »
pretty good analysis on stuff. I've been meaning to do some basic displacement sims too, just haven't had a chance :( good info, it's definitely of use to others.

Offline jizzaztartist

  • Posts: 15
Re: Omega 60 - By Windeh
« Reply #24 on: Mon, 28 January 2019, 23:45:44 »
Quote
NAMEPLATE LOGO LASER MARKING DESIGN TO BE CONFIRMED


Or you could just not have one.

Online fishinaspacesuit

  • Posts: 48
  • Location: Singapore
Re: Omega 60 - By Windeh
« Reply #25 on: Tue, 29 January 2019, 03:11:48 »
There's really quite a bit of thought that went into these, thank you so much for showing your design motivations through the videos :)

Online Windeh

  • Thread Starter
  • Posts: 270
  • a no name
Re: Omega 60 - By Windeh
« Reply #26 on: Tue, 29 January 2019, 07:31:49 »
Windeh,

The amount of work you place in all your creations is amazing. I’d totally be in if R1 had a HHKB top, but sadly I may have to wait until R2. Regardless, I’m in whenever that passes by (Hopefully shortly after R1?).

Thanks!!

Don't really have a timeline yet, but the demand for HHKB/WKL blocked tops are quite high, so I'll definitely try to make it happen ASAP.


Some really thoughtful design choices that you rarely see in these ICs. I'm not interested in a 60% but if you do a 65% or or 75% I'd be in I think.

I'm definitely interest in designing a 75% or TKL.  :thumb:


pretty good analysis on stuff. I've been meaning to do some basic displacement sims too, just haven't had a chance :( good info, it's definitely of use to others.

If you zoom in, you can see the maximum displacement values. However those values are quite meaningless as the force of 10N is arbitrary, and a PCB would stiffen the plate right up, further diminishing their usefulness. The sims are for visual reference, to give an idea on relative stiffness of areas within the plate.

Oh and I don't lurk nearly enough.  ;)


Quote
NAMEPLATE LOGO LASER MARKING DESIGN TO BE CONFIRMED



Or you could just not have one.

I could! In fact I will offer optional extra blank name plates with the kits.  :thumb:


There's really quite a bit of thought that went into these, thank you so much for showing your design motivations through the videos :)

I have the luxury of coming after all the great designs that came before mine, knowing what works and what does not helps immensely.

Still much to learn, and many bridges to build.  :)

« Last Edit: Tue, 29 January 2019, 07:57:21 by Windeh »

Online vikipedeer

  • Posts: 70
  • Location: Thailand
Re: Omega 60 - By Windeh
« Reply #27 on: Tue, 29 January 2019, 08:03:04 »
sadly missed out on Reflex/Paradox, finally recoup from the financial problem and I can hopefully join this GB
Current board:KBD75V2, Whitefox, Time TKL

say yes to bacon overlord

Offline aznreaper

  • Posts: 243
  • Location: AB, Canada
Re: Omega 60 - By Windeh
« Reply #28 on: Tue, 29 January 2019, 09:31:07 »
Yeah I would be interested in a TKL as well with a similar clean look and design

Online beekey

  • Posts: 173
Re: Omega 60 - By Windeh
« Reply #29 on: Tue, 29 January 2019, 10:14:37 »
Plate option with split space please. Like the design a lot.
Will there be other plate layouts?

Online juaninamilli

  • Posts: 77
Re: Omega 60 - By Windeh
« Reply #30 on: Tue, 29 January 2019, 12:06:49 »
One plate layout to rule rule them all. I like it

Offline Lbibass

  • Posts: 29
Re: Omega 60 - By Windeh
« Reply #31 on: Tue, 29 January 2019, 13:54:23 »
Damn. This is incredibly well designed, with a good explanation of why you made the design decisions. That's incredible.
Plus: BEAUTIFUL design.


Offline pixelpusher

  • Formerly known as reececonrad
  • Posts: 2530
  • Location: Tennessee - USA
Re: Omega 60 - By Windeh
« Reply #32 on: Tue, 29 January 2019, 14:20:16 »
I get a single layout plate design, b/c I'm not a fan of swiss cheese plates.  But why has everyone moved away from switch top opening cuts?  It's very reassuring to know that you can go back and fix problems with switches down the road.  When a switch gets a roughness to it, or a spring starts to squeak... I'd much rather open it up and fix it than have to take apart the entire board and desolder it.
:)

Online beekey

  • Posts: 173
Re: Omega 60 - By Windeh
« Reply #33 on: Tue, 29 January 2019, 14:51:11 »
One plate layout to rule rule them all. I like it
Why? Other GBs offer custom plates or multiple base layouts. This is an IC, so I ask for options. Does not mean to have the one universal-do-all plate. For me the typical default layout sucks. Would not buy.
 

Online juaninamilli

  • Posts: 77
Re: Omega 60 - By Windeh
« Reply #34 on: Tue, 29 January 2019, 15:01:20 »
Haha! Easy man. Just being silly. Tsangan and wkl are best, IMO.  They're the same plate so . This isn't standard layout, so not sure what you're referring to. What layout would you be looking for? (Don't say arrows)
One plate layout to rule rule them all. I like it
Why? Other GBs offer custom plates or multiple base layouts. This is an IC, so I ask for options. Does not mean to have the one universal-do-all plate. For me the typical default layout sucks. Would not buy.

Offline Lbibass

  • Posts: 29
Re: Omega 60 - By Windeh
« Reply #35 on: Tue, 29 January 2019, 15:27:44 »
I get a single layout plate design, b/c I'm not a fan of swiss cheese plates.  But why has everyone moved away from switch top opening cuts?  It's very reassuring to know that you can go back and fix problems with switches down the road.  When a switch gets a roughness to it, or a spring starts to squeak... I'd much rather open it up and fix it than have to take apart the entire board and desolder it.


I agree. Switch top opening is super important to me, and I think that most would rather have it than not.

Offline mdlt97

  • Posts: 74
  • Location: Toronto
  • This **** is still better than Keebtalk
Re: Omega 60 - By Windeh
« Reply #36 on: Tue, 29 January 2019, 15:34:20 »
what other layouts could ever you want? unless you want more bottom row spots there's nothing else to be added

also for having holes for switch top takes away from the board, the whole point of this board is to be a great typing feel and sound and switch top can affect both of those

also id assume if you really wanted switch top you could ask windeh to make plate files for that type of plate
TX60 | Duck TC-V3 | Duck Eagle V2 1/1 Bottom | Zephyr r2 | Duck Sidewinder | Mech27v2 | TX75 Purple | LZ CLS | Duck Viper v2 | Orion 3

Offline Oblotzky

  • Posts: 1304
  • Location: Cologne, Germany
  • Backspace is for people that make mistakes.
Re: Omega 60 - By Windeh
« Reply #37 on: Tue, 29 January 2019, 15:56:23 »
Lovely design  :thumb:

Online fishinaspacesuit

  • Posts: 48
  • Location: Singapore
Re: Omega 60 - By Windeh
« Reply #38 on: Tue, 29 January 2019, 19:19:25 »
One plate layout to rule rule them all. I like it
Why? Other GBs offer custom plates or multiple base layouts. This is an IC, so I ask for options. Does not mean to have the one universal-do-all plate. For me the typical default layout sucks. Would not buy.

Not all GBs offer that option, some examples of GB that offer just one layout are the RAMA M60-A and KOYU.

Offline Remsky

  • Posts: 221
  • Location: Portland, OR
Re: Omega 60 - By Windeh
« Reply #39 on: Tue, 29 January 2019, 19:30:12 »
I see why you have screw holes in the corners (I presume for alignment), but since this is a sandwich style mount, it will contribute towards the pressure put onto the plate. IMO remove them (saves  money spent on threading as well and makes 4 less screws that could strip) and also remove the bottom center mounting post. Even with the cut, having a mounting point directly beneath the spacebar switch will probably make the spacebar sound a bit stiffer instead of letting it flex a little bit more for that thock. You could also do a near row wide relief cut in between the numrow and QWERTY row to even out the flex on the alphas (with the current 5 point mount plate design anyway). to make the flex more consistent.

Corner screws

The corner screws are not actually used to align the top & bottom case, as the top self-aligns during installation;


Those screws are there to ensure that the corners of the case are tightly clamped, so there are no visible gaps. This "Feature" can be seen in most custom keyboards. My first Reflex/Paradox protos omitted the corner screws, which were added in subsequent updated design, for no other reason then to ensure a good tight gap-less fit. I would have loved to simplify the design with less screws, but felt that these are necessary for a premium product finish.



Since you took the time to offer your suggestion, I will also offer my time to simulate your plate design suggestion.  :D

Let's call it the Remsky Special:

(Attachment Link)

Hopefully with the above explanation, you can follow my thought process for these design choices, and we can at least agree that some thought have been given to create this keyboard.  :thumb:

Please do let me know if you have any questions or suggestions, or would like to offer any counterpoint to what I have written above. I do thoroughly enjoy these technical discussions!  :D

Your mounting is a modified method of sandwich mounting, with grooves for o-rings to be fit for sound and/or feel. Any additional pressure used to seal the two case halves will directly affect the plate, adding unnecessary pressure to the plate. That's what the four screw holes on the corners are doing. I think you could have standoffs/insert system instead to seal the corners, there will still be pressure, but less so than with the torque from screws.

Also I think the pressure tests would be more accurate if you were able to do them while the plate was mounted, although that maybe trickier because then you would have to simulate the amount of force the case puts onto the plate and torque from screws, etc.

Thank you for considering my suggestion with the plate.
« Last Edit: Wed, 30 January 2019, 12:41:13 by Remsky »

Offline PikaJoyce

  • Posts: 31
Re: Omega 60 - By Windeh
« Reply #40 on: Tue, 29 January 2019, 19:36:13 »
I appreciate the thought you've put into your design, Windeh. This seems like an evolution from the Paradox/Reflex. Will be keeping my eye on this :)

Offline fireworm

  • Posts: 215
Re: [IC] Omega 60 - By Windeh
« Reply #41 on: Tue, 29 January 2019, 23:53:44 »
... Perhaps having screw in blockers for hhkb and WKL are a compromise?

But maybe they look too low quality...?

Online adb7

  • Posts: 15
  • Location: US-IL
Re: [IC] Omega 60 - By Windeh
« Reply #42 on: Sat, 02 February 2019, 18:15:10 »
Will definitely get one if native WKL is offered. Probably not if they are screw in blockers though.

Online juaninamilli

  • Posts: 77
Re: [IC] Omega 60 - By Windeh
« Reply #43 on: Thu, 07 February 2019, 13:20:48 »
I know you haven't selected colors yet, but with all of these new cerakote opportunities, do you think you might offer a raw option?

Offline reidd

  • Posts: 26
Re: [IC] Omega 60 - By Windeh
« Reply #44 on: Thu, 07 February 2019, 14:18:42 »
This is hands down the best looking 60% I've ever seen.

Online tex_live_utility

  • Posts: 269
Re: [IC] Omega 60 - By Windeh
« Reply #45 on: Fri, 08 February 2019, 17:06:40 »
What are the pros and cons of 6061 and 5052 for a keyboard?

Online Windeh

  • Thread Starter
  • Posts: 270
  • a no name
Re: [IC] Omega 60 - By Windeh
« Reply #46 on: Mon, 11 February 2019, 03:34:27 »
I get a single layout plate design, b/c I'm not a fan of swiss cheese plates.  But why has everyone moved away from switch top opening cuts?  It's very reassuring to know that you can go back and fix problems with switches down the road.  When a switch gets a roughness to it, or a spring starts to squeak... I'd much rather open it up and fix it than have to take apart the entire board and desolder it.

I get a single layout plate design, b/c I'm not a fan of swiss cheese plates.  But why has everyone moved away from switch top opening cuts?  It's very reassuring to know that you can go back and fix problems with switches down the road.  When a switch gets a roughness to it, or a spring starts to squeak... I'd much rather open it up and fix it than have to take apart the entire board and desolder it.

I agree. Switch top opening is super important to me, and I think that most would rather have it than not.

I may look into switch top opening in future plate designs. My concern is that the additional removed material weakens the plate somewhat, which increases the likelihood of the plate dimensions distorting during laser cutting.



Plate option with split space please. Like the design a lot.
Will there be other plate layouts?

Not for this GB to keep the costs down. I can send participants the plate file should you wish to create and cut your own plate layouts.



Lovely design  :thumb:

Thanks! I'm sure you are not at all biased by those lovely keycaps.  ;)



Online dantambok

  • Posts: 388
  • Location: Philippines
Re: [IC] Omega 60 - By Windeh
« Reply #47 on: Mon, 11 February 2019, 03:51:36 »
i'm sad that there wont be WKL for now  :mad:

Online Windeh

  • Thread Starter
  • Posts: 270
  • a no name
Re: Omega 60 - By Windeh
« Reply #48 on: Mon, 11 February 2019, 06:23:53 »
Your mounting is a modified method of sandwich mounting, with grooves for o-rings to be fit for sound and/or feel. Any additional pressure used to seal the two case halves will directly affect the plate, adding unnecessary pressure to the plate. That's what the four screw holes on the corners are doing. I think you could have standoffs/insert system instead to seal the corners, there will still be pressure, but less so than with the torque from screws.

Quote from: Remsky on 01/30/2019
I literally told windeh that the 4 screw holes in the side would contribute towards the pressure on the plate, therefore making it stiffer
he said no
then he said how they align the case by keeping the plate down in those corners (WUT? where did I say that??:-\)

I think there is still confusion on the mounting design, where you think the plate is sandwiched between the case as per a traditional sandwich mount (e.g. TX60), with O-Rings thrown in. This is NOT how my mount is set up.

Firstly, the Aluminium part of the case (i.e case top & bottom) does NOT touch the plate AT ALL when the case is assembled. Although this was demonstrated in my previous post on the subject, let me try another method to convey the design.

If the case touches the plate, there will be electrical continuity.


Since the plate ONLY touches, and is held in place by the O-Rings, it is completely isolated from the case.

212683-0


Now that we have established that the case does not touch the plate directly, with the only thing putting pressure on the plate being the O-Rings.


Now all that's left is to determine HOW the pressure is applied to the plate through the O-Rings, and whether the corner screws applies any additional pressure the O-Rings.

For my design, the screws is only responsible for joining the top & bottom case. From the section analysis below, you can see that once the top and bottom part touches, you cannot apply any additional pressure to the plate via the O-Rings, no matter how much torque you apply to the screws, or even the number of screws you use.

212685-1

The O-Ring compression ratio is set by me, via the distance between the top and bottom O-Ring recess shown below, as the distance marked "X".



Therefore, for a higher preset pressure, X can be reduced for greater O-Ring compression. Conversely, increase X reduced the pressure placed on the plate through the O-Rings.

With a predetermined O-Ring compression, user can easily adjust the pressure on the plate by the last and only variable remaining; the O-Ring Hardness, or its durometer value.

Hopefully the above is easy to follow. Not sure how much better I can explain my design other then to show you in person, but please rest assured that the corner screws have no effect on the typing experience.


Also I think the pressure tests would be more accurate if you were able to do them while the plate was mounted, although that maybe trickier because then you would have to simulate the amount of force the case puts onto the plate and torque from screws, etc.

Since there are minimal deflection of the O-Rings even under large amount of pressure applied to the plate, modelling them as rigid is sufficiently accurate for my purpose.

Computer aided simulation of anything is always a compromise between accuracy and ease of simulation. As complexity of the simulation increases, the benefit to increased accuracy decreases exponentially. Therefore we always seek the sweet spot of the least complex simulation, while obtaining the most amount of useful data.

I have seen people criticizing the simulation for its lack of PCB simulation, which I agree will have the most direct affect to the plate deflection, to simulation the switch, soldering, mounting, O-Ring Hardness, screw pressure etc, which substantially increases the simulation complexity while provides disproportionally small benefit to the simulation.

Some gems from Discord regarding the simulations: "Retarded", "useless shinfo" "it's ****in pointless". Why so much hate?  ;D

I will always take 60% of data over 100% of nothing.  :thumb:
« Last Edit: Mon, 11 February 2019, 07:30:26 by Windeh »

Online Windeh

  • Thread Starter
  • Posts: 270
  • a no name
Re: [IC] Omega 60 - By Windeh
« Reply #49 on: Mon, 11 February 2019, 06:33:20 »
There are so many gems from Discord talking about my design, I must find time to do a BEST OF DISCORD:))

I wish one day I'll be so experienced in the hobby that I will know more about the design then the designer, from the sound, how it will feel terrible, to how "wobbly it will feel" (WUT?), just by browsing through the Interest Check.  :cool: