[COMPLETED] Laser Cut Prototyping mini-GB

[mtl]

Awesome! I need to get on the ball and finish ordering parts!

[Broadmonkey]

Awesome indeed! I can't wait to try out the layout! Now I just need to wait for my Teensies to be posted and I should be more or less set... well, I still have to nick some parts from school

[Zifle]

My plates just arrived, and everything seems to be perfect! I can fit my stabs without any issues, and there's space between the stab insert on the keycap, and the one on the plate.

I probably should have made the screw holes slightly bigger, as the screws have a hard time fitting in there, I'll look into drilling just a micron off to the sides, or simply going with 2.5mm screws if it becomes too inconvenient.

Thanks for the gb, my order ended up exactly as I requested it, and shipping hasn't seemed to done any damage whatsoever, either. Excellent!

(I'll probably take some pics at a later time, when I actually start building the board, I still need switches for it!

[Broadmonkey]

That means that mine is not far away!
What switches do you plan to use?

[Zifle]

Jailhouse blues, so I have a lot of work ahead of me!

[Muirium]

Jailhouse blues? Is that for banjo or harmonica?

[pasph]

Mine is still to be delivered
5 days for 280 km...

[pasph]

Jailhouse blues? Is that for banjo or harmonica?

http://geekhack.org/index.php?topic=45992.msg1040217

[Zifle]

Jailhouse blues? Is that for banjo or harmonica?

http://geekhack.org/index.php?topic=45992.0

edit: Grrr pasph !

[Muirium]

Hmm. I like the idea of raising the activation point, but clicky switches gotta click for me. Are there similar travel reducing tricks for linears? Not just typical O-rings but a way to push the activation point up?

[Zifle]

That wouldn't be possible, unless you physically change the stem itself (eg., 3d print / mold new ones), as the reduction is done by keeping the clicker (white part) in place, and not allow it to move. With that said, jailhouses are highly variable, and can be done with or without click, depending on the way it's modded. This also makes it a somewhat unfeasable switch, as you can easily end up with some keys clicking, while others do not.
And then there's the spring force issue, where you get an extra 20ish gram of weight needed to press it down, without modding the spring, or swapping in a lighter one. It's certainly not for everyone!

[matt3o]

My plates just arrived, and everything seems to be perfect! I can fit my stabs without any issues, and there's space between the stab insert on the keycap, and the one on the plate.

Awesome! Don't forget to post some images!

Jailhouse blues, so I have a lot of work ahead of me!

now that would be something!

Mine is still to be delivered
5 days for 280 km...

well, Rome postal hubs are notoriously... lazy

[Muirium]

No surprise that shipping outside of Italy is quicker than sending down it. I've been to the south, speed and efficiency is not top priority down there! (They know the important things instead. It's one place I might retire if I ever afford to…) It took until Monday for my parcel to leave the country. I reckon I might receive it today.

Re: Jailhouse Blues. Heavier feel is fine by me, as I've a bag of greens to put into my plate once it's here. But I daren't do these switch mods myself. Too fiddly and error prone for me. I'd snap things while learning and then have to do it all over again once I realised the final switches deviated at all.

I must make my own capacitative switch sometime so I can experiment with activation point. My fingers say they want it higher than it is in MX. But only experience can tell.

[matt3o]

I must make my own capacitative switch sometime so I can experiment with activation point. My fingers say they want it higher than it is in MX. But only experience can tell.

if you make the activation point of Topres higher... you would need no pressure at all

[Muirium]

The high activation point is one of my favourite things about Topre. I'd like to try something similar over a changeable spring, like MX.

So my idea is to try to rig up something with a light conical spring and capacitative sensor (elements from Topre's design) with with an MX mount for caps and the main spring. Then I could dial in different activation points (by the sensor's logic) and put in different springs. Having control over both would give me a great linear test rig.

Clicky and tactile I don't know. But it's feasible for a linear experiment.

[matt3o]

So my idea is to try to rig up something with a light conical spring and capacitative sensor (elements from Topre's design) with with an MX mount for caps and the main spring. Then I could dial in different activation points (by the sensor's logic) and put in different springs. Having control over both would give me a great linear test rig.

that would be a very interesting experiment... mmmh... (mumbling)

[Broadmonkey]

I tried jailhouse blues once and liked them, the higher actuation point is welcome, but they are also kinda weird in their own way... and very tedious to make! I made mine with small strips from an old social security card and got a very stable stem! but it was also harder to make than with a wire.
Just got home and found that the mailman had tried to deliver my package earlier today :/

[matt3o]

Stabs filing tutorial.

It seems that the specs for the stabilizers are few microns off. The costar stabilizers fit the hole but the stems hit the stabilizer making the key-press scratchy.

This is the side view:

Code: Select all

             |              |  <---- Keycap
             +---+-+--+-+---+
                 | |  | |  <---- Stem
                 | |  | |
The stab ----> /|| +--+ | |\
hits here     / |+------+ | \  <---- Stabs
             |  |         |  |
             +--+---------+--+  <---- Plate
             +---------------+

Fixing this is extremely simple. The stab holes are just a little too tight. First of all clear all laser cut left overs with a file along the stab hole perimeter. Then all you need to do is to file the upper part of the hole a fraction of a millimeter. Proceed at small steps trying the key each time.

Code: Select all

                     +---------+
File here ---> +--+  |         |  +--+ <--- and here
               |  |  |         |  |  |
               |  |  |         |  |  |
               |  |  |         |  |  |
               |  |  +---------+  |  |
               +--+               +--+

On aluminum this will take you just few minutes, on steel of course you'll need some more time, but still feasible.

If you end up filing too much, you can simply glue the stab with epoxy.

Also consider that YMMV, I noticed that some of the plates are better than others, so maybe you may not need this filing session. But in case you do, don't despair

[Muirium]

Noted. The tutorial I'm most looking forward to is the one about programming your Teensy. I know the overview — TMK firmware collection, use the GH60 code, then modify it to match your chosen physical matrix — but not the details. You've done it a few times now, fortunately.

[Muirium]

The moment I start painting, the postman turns up at last. Thanks Matt. Now I'm waiting on my Teensies… and my desk.

[matt3o]

The moment I start painting, the postman turns up at last. Thanks Matt. Now I'm waiting on my Teensies… and my desk.

painting wall or canvas

anyway! glad the package reached you safe and sound! take pictures of course

[Muirium]

Painting walls. The tricky kind to work around. Then it's hammer and file time!

[pasph]

got it

[Broadmonkey]

Picked mine up at the post office earlier today!

[matt3o]

I think only MTL misses it.

[mtl]

I think only MTL misses it.

Inviato all'estero (US) dal centro postale di MILANO CSI WINDOWS in data 24-SET-2013
Accettato dal centro postale di FIRENZE VR in data 20-SET-2013

Sent abroad (U.S.) from the center post of MILAN CSI WINDOWS on 24-SEP-2013
Accepted by the postal center of FLORENCE VR on 20-SET-2013

I'm guessing this means it is currently somewhere over the Atlantic?

[matt3o]

I think only MTL misses it.

Inviato all'estero (US) dal centro postale di MILANO CSI WINDOWS in data 24-SET-2013
Accettato dal centro postale di FIRENZE VR in data 20-SET-2013

Sent abroad (U.S.) from the center post of MILAN CSI WINDOWS on 24-SEP-2013
Accepted by the postal center of FLORENCE VR on 20-SET-2013

I'm guessing this means it is currently somewhere over the Atlantic?

It is in US already, you can track it on USPS website

Processed Through Sort Facility
September 24, 2013, 5:33 pm
ISC NEW YORK NY(USPS)
Registered Mail™

[mtl]

Oh cool.. Should be here any day then. Thanks for the tip!

[mtl]

The shipment arrived today. Thanks, matt3o! Everything looks good and I am impressed at how some of the detailed features came out, like the small screw holes cut into 3mm aluminum. I think the keyboard will be difficult to assemble since the layers will require some coercion to line up and at the same time I'll be trying to get wires to lay in there right. But that challenge is a long way away.

Here's a pic of the TrackPoint. I'm happy it lines up very well, since it was a difficult piece to measure with just a ruler. I'll have to shave off parts of the MX switches, but that was unavoidable without increasing the key spacing. I turned the G, H, B, V, and N switches 90 degrees to give a little more room for the screw holes and avoid having to cut out necessary parts of the switches.

tp-top.jpeg (19.39 KiB) Viewed 4866 times

So far I've found one measurement mistake I made, and it's a pretty big one. I think it is fixable though. The notches for the left and right mouse buttons are way off. They should be lined up with the large circles of the black plastic frame in the pic below. Don't know what I was thinking. I will have to cut or drill through 3mm aluminum to fix it.

problem.jpeg (56.41 KiB) Viewed 4866 times

Here's a bonus picture of the bottom layer with the components laid out.

Spoiler:

components.jpeg (47.58 KiB) Viewed 4866 times

Overall I'm happy with the parts and excited to get started putting this thing together. Still waiting on some fasteners but everything else is ready to go.

[matt3o]

So far I've found one measurement mistake I made, and it's a pretty big one. I think it is fixable though. The notches for the left and right mouse buttons are way off. They should be lined up with the large circles of the black plastic frame in the pic below. Don't know what I was thinking. I will have to cut or drill through 3mm aluminum to fix it.

problem.jpeg

If the layer you need to drill is aluminum, despair not! You can drill it like butter. So I wouldn't consider that a big mistake at all!

Keep us posted with the progress of you great project.

PS: remember to insulate the base!