Group Build prototyping phase

[Muirium]

My wallet could use a breather too!

Let's spend a month or two devising ever more cunning schemes instead of actually making them, yet.

[matt3o]

Not 100% sure but it is likely that I will have a very small test run in just few days. I want to test wooden case + alu or steel plate with top frame (not open frame). I'm not a huge fan of wooden case, but it's very cheap to laser cut and I have some very nice ideas to make it look very nice (the layers will be glued together and they will look like a solid piece of wood. it will also be polished and treated with special sealers).

The wood sheets are big enough to host 2 keyboards, so if anyone is interested I could have two keyboards done complete with super cool double sided PCB.

If the experiment turns out well, it could be the default case.

[Muirium]

You know I'm always interested. (If not as reliably ready to fund it!) Wood's more to my taste, already anyway.

About that PCB, reckon you could lay out something for this:



I've got the switches, and a 60% of these little guys (in the same layout as my last one) would be something to behold!

[matt3o]

those are not plate mounted right?

[Muirium]

Correct. They look like this:



Half the travel and compatible with MX caps. But PCB mount only.

Don't worry, there's no rush on my stuff. You've plenty on your plate at the moment!

[matt3o]

those would require some engineering. the plate holds the PCB. Without plate you have to screw the PCB to the base (using spacers or something I guess). Being wood it's pretty easy to work with, but still it would require a completely different template. Actually you wouldn't even need the top frame, since you don't have to conceal the plate... hum....

[Muirium]

Indeed, it's not a great candidate for your trial run.

The other project I'm lining up is a Matias 60% or TKL. But that one's complex thanks to stabs, as usual.

I do have loads of MX switches lying around spare though, and lots of caps on the way. Even 4u space bars, and thanks to HzFaq even a matching 4u plate mount Cherry stab for it!

[matt3o]

I could work on the Alps one, I have stabs and everything... maybe time has come... but honestly with all the MX caps that I have lining up, maybe I just should work on my ghetto green

[Muirium]

Love that term. Ghetto Green was what we called the local park where I grew up!

Do whatever's better for you. I'm not kidding about money right now! And I haven't bought those Matias switches yet. At least they're better value than MX…

[scottc]

At least they're better value than MX…

That's not always true! I've noticed that people sometimes sell MX black for as little as 10c per switch. If you're worried about wear, combine that with replacement springs and you've got 100 good-as-new switches for 20 euro. Though that's not a great solution if you're not into linear MX. They just happen to be my favourite, though - I don't like tactile MX at all, and while the clickies are okay, they don't measure up to their buckling spring or ALPS brethren in that regard. So it works out fine for me!

However, results may vary if you're thinking of making a keyboard of 100% MXLOCKs.

[Vierax]

Not 100% sure but it is likely that I will have a very small test run in just few days. I want to test wooden case + alu or steel plate with top frame (not open frame). I'm not a huge fan of wooden case, but it's very cheap to laser cut and I have some very nice ideas to make it look very nice (the layers will be glued together and they will look like a solid piece of wood. it will also be polished and treated with special sealers).

Alu plate + wooden case is a thing I was working on : the case is made but the alu plate isn't (and won't be since it my layout has changed). I made it with glued pieces of MDF because it's easier to work than a real living matter like wood but plywood should be great on lasercut.
For protection, I still have a selfmade Dammar varnish to test before using it on my oil paintings.
Wood is a bit light, maybe a steel plate should be at the bottom to add weight and keep it more stable.

The wood sheets are big enough to host 2 keyboards, so if anyone is interested I could have two keyboards done complete with super cool double sided PCB.

If the experiment turns out well, it could be the default case.

I could be interested, depends on what layout you plan to do.

[matt3o]

layout can be anything. Mine will be a 65%.

it would be great if the PCB wasn't wider than 30cm though. Otherwise I'd have to split it into two pieces.

[Madhias]

As i've looked on my key-pressing-stats (home and work), i will go without F-row. Just 4 keys i use (F5 - of course, F8, F9, F10 and F12). 65%! Definitely with arrows.

[matt3o]

This would be my wooden experiment:



It's pretty to tears

[Broadmonkey]

Perfect, matt! That bottom row is pretty much what I have come to think would be the best way to build a 16col 65% keyboard (if we have to use a 6.25 spacebar).
Having used my keyboard from the last gb a fair deal now, the space between the arrow keys and the bottom modifier to the left of it is actually preferable, as is should help a great deal in quickly locating the arrow cluster.

[matt3o]

thanks!
another option is to place 3 1u to the left of the arrow cluster, but I think I prefer this solution best.

now I'm trying to understand if it is better to have a top frame that "closes" the keyboard, or if the top frame should be glued to the sides instead. The goal of course is to hide the alu plate from side view, so the keyboard looks like a big block of wood.

[Vierax]

layout can be anything. Mine will be a 65%.

it would be great if the PCB wasn't wider than 30cm though. Otherwise I'd have to split it into two pieces.

You sure about the "anything" ? because you know my taste for weird splitted layouts like ergodox
Actually, I'm not a fan of PCB + mounting plate design (my desoldering skill is a bit low) but I can handle this for a testing purpose and if I mount the MX I want, I don't really need to change them except for repairing (witch won't happen before a while with brand new switches) so it's all good to me. I know, my talk is pointless.

[matt3o]

oh well the PCB is totally optional, I thought it was nice to have.

[Muirium]

I'm interested in PCBs this year because of:

• PCB mount stabs
• PCB mount switches (specifically little M84s)
• Tighter case design with PCB mounted Teensy or naked chip
• Saves a lot of effort!

I think I'll even convert my beloved Shiny 60% to PCB to clean it up in there. My wiring skills are proper hairy!

[matt3o]

tried cherry stabs. sucking confirmed for custom keyboards. back to costar.

[woody]

Where do you get Costar stabs from?

[matt3o]

Where do you get Costar stabs from?

wasdkeyboards

[rindorbrot]

tried cherry stabs. sucking confirmed for custom keyboards. back to costar.

What?
Please make them work for the final version, I hate Costars...

[matt3o]

costar stabs are easier to deal with in custom keyboard simply because the error margin is pretty high. with cherry stabs you can't really afford a fraction of a mm error.

[Muirium]

Plus you plain hate Cherry stabs. It's a personal preference thing, though. I'm with Rindorbrot!

[matt3o]

Plus you plain hate Cherry stabs. It's a personal preference thing, though. I'm with Rindorbrot!

it is not a personal preference. if you build custom keyboards costar is better because:

- building custom spacebars is easier. To make space for a teensy I placed a stab in an unusual location and then cut/paste the stems on the spacebar. That would be very very difficult with cherry stabs.
- plate holes errors are easily fixable (by filing the hole or the stab itself). Almost impossible on a cherry stab

I have no preferences between cherry or costar stabs on a keyboard (even though costar still makes a better sound imho), but when it comes to custom keyboards there's no reason on Earth to prefer cherry.

Edit: unless of course you go plate-less, in which case you have little alternatives.

[Broadmonkey]

I think in a board with both plate and PCB I would always go for Cherry PCB mount stabs, they both feel and sound much better, plus they are easier to change keycaps on. They're also not very picky about the cutout in the plate so long it's not too small.

But the benefits of Costar stabilizers in a plate-only custom keyboard cannot be trumped. They allow for a very compact design as the controller (ie. Teensy) can be mounted more or less under the stabilizer, something Cherry plate-mount stabs will not allow. If you have enough space to mount the controller elsewhere, I would go for the Cherry variant, but that is rarely the case.
I disagree, however, with matts statement about Costar stab hole errors being easily fixable, mine still gives me hell especially since it works fine with my OEM spacebars and shitty with my GMK.

[pasph]

This would be my wooden experiment:



It's pretty to tears

Can this be made also for Alps (Dell style stabs)?
But i'll buy a wood case regardless...

[Vierax]

Adding holes for Alps has been already done before with success on Community MX PCBs. It's just extra work.
Comparatively speaking with MX, the only problem are the differences of plate holes and the distance between plate and PCB. But if it's an Alps exclusive keyboard, my concern is irrelevant.

oh well the PCB is totally optional, I thought it was nice to have.

I don't need one but I'll be glad to have a PCB only if it's to improve your skill of the homemade PCB craft (insulation+perchlo bath or milling?). I still could use it to make a second board for my best friend, but still not sure about some details in column staggering and rotations. And designing a PCB will be a first to me.

Now, I have to end my SVG file, test it in print to adjust the layout and including the case frames design : a bit of work
Is it safe to have a wired matrix and no Faraday cage to prevent ECM? Should I add alu/copper tape inside the wood case ?

[guilleguillaume]

This would be my wooden experiment:



It's pretty to tears

I like that layout. Kind of Kmac Mini without the F keys.