XL keyboard technical drawings

[+scorpio_ny]

10 hours ago, flemingt said:

No problem, do you want to give his version a try instead?

cherry to stackpole replacement stem v2.stl 156.28 kB · 2 downloads

Sure! I will try to see if I can print it tonight.

[+scorpio_ny]

Hi @flemingt

 

I managed to print the latest version tonight. I wanted to get you some photos so you can see the latest version in action. Sorry if the photos are a little blurry. Sorry again for the wall of photos.

 

                       

 

      

 

My observations on v2 stem:

1. Fit is much better with the key cap! Very slight up and down wobble movement of the keycap.

2. Keycap comes off easily off stem. But, it has enough friction so it does not fall off if the keycap is upside down or shaken.

3. When the switch fully pressed, the bottom part of the keycap does not hit the top of switch case. The gap is very small too! See last picture (Good job on measurements!)

 

This was tested with a Gateron Blue switch.

Any questions?

 

 

Edited February 2, 2022 by scorpio_ny
Correct grammar

[flemingt]

Oh wow, thanks for checking that out and documenting so thoroughly.

I figure there should be a 0.2mm gap between the base of the cap and the body of the switch. I can probably minimise that further.

Do you think there's sufficient friction to hold the cap on? I'm not thrilled that there is some wobble.

I can always make the stackpole part bigger in section, from 4.76mm square to something like 4.78mm square. If it's too big it can be sanded I suppose.

[+scorpio_ny]

3 hours ago, flemingt said:

Oh wow, thanks for checking that out and documenting so thoroughly.

I figure there should be a 0.2mm gap between the base of the cap and the body of the switch. I can probably minimise that further.

Do you think there's sufficient friction to hold the cap on? I'm not thrilled that there is some wobble.

I can always make the stackpole part bigger in section, from 4.76mm square to something like 4.78mm square. If it's too big it can be sanded I suppose.

The wobble is due to the fact that the hole on the bottom of the keycap is not a true square hole. If we think about it, it is really two extended "fingers" that are on top of a hole that was designed to sit into another recessed housing on the Atari keyboard. Maybe to add some small protrusions so that part the keycap has something to "sit" on at those spots? Another thing to keep in mind is that I am printing with a FDM printer to do some initial prototyping. If this model is printed on a resin printer where it has more resolution, the fit may be a lot loose due to the finer detail that the resin can produce.

 

[flemingt]

If I'm understanding you correctly I have some thoughts. However it'll be difficult to implement without getting some accurate measurements for those wings. I'll give it a shot.

[flemingt]

rev 3, hopefully this is better. I recreated the CAD from scratch in Fusion 360 because kit-bashing the stl file I found was more trouble than it was worth. I've added some stabiliser tabs, I eyeballed the dimensions!

cherry to stackpole replacement stem v3.stl

[+scorpio_ny]

Ok great! I got home from work late yesterday, so I did not get an opportunity to do another test print. I will see if I can print the newest update sometime later today.

 

[+scorpio_ny]

Hi @flemingt,

 

I am sorry that I took longer with the new updates. Work was busy for me. Here are the latest photos:

 

                

 

               

 

In this revision there are a couple of issues in relation to the addition of  the tabs to support the keycap.

1. The tabs need to be a little higher. In the photos with the top of the stem hits the top of the keycap before the lower part of the keycap can rest of the taps.

2. The tabs prevent the stem from passing completely through the upper housing of the switch. The only way possible be to do so is by turning the stem 90 degrees.  I turned it and it was still too wide. So I tried to tilt it to one side. This way I could get one tab through, but the lower part of the stem prevents any further movement.

 

A couple of suggestions:

a. The tabs need to be moved a little higher on the stem so the cap can actually sit on it.

b. The tabs need to be a little smaller to allow the stem to be more easily assembled in the upper switch housing and large enough to allow the keycap to sit on it.

 

I know measurements would help but my caliper is not working and I do not have a ruler on hand. I need  a working caliper anyway, so I ordered a new one. I should receive it sometime tomorrow. 

[flemingt]

Thanks for all your support on this. Your pictures have been very helpful. 

Rookie mistake by me ignoring the top hole of the switch enclosure. I've established the maximum width from the base of the switch body and made sure the wings are narrower than that. The wings have also been raised by 0.75mm. It'll probably still require a 90 degree rotation once through the cover or a bit of fiddling to get through. I'm hoping the fillet edges on the stackpole part allow that, otherwise I'll have break the stackpole part to put a cylindrical section underneath the wings. If that doesn't work I'll just remove the wings and marginally increase the section of the stackpole for a better friction fit.

cherry to stackpole replacement stem v4.stl

[+scorpio_ny]

6 hours ago, flemingt said:

Thanks for all your support on this. Your pictures have been very helpful.

It is appreciated that @ivop , yourself and others are taking the time on working on various projects that may benefit the Atari community. I am glad to be able offer some small assistance to that may lead to the completion of a project. ?

Downloaded the file and running the print as we speak. I will keep you posted!

 

[+scorpio_ny]

HI @flemingt! Here is are the latest updates for v4:

                    

 

     

 

1. In this version, I was able to get into the upper housing of the switch! To do so, I had to insert at 90 degrees and "twist" in like a screw.

2. The keycap sits very nicely with this version! Very little wobble up and down as before 

3. The keycap fit is loose. There is some friction, but not much. It did not fall off when I turned it upside down and shoot, but I was able to launch the keycap from the stem.

4. I did not mention this in the previous items is that the pieces of the the stem (the tiny points) are very brittle and break off. I am not sure how critical they are for the function.

5. Finally, in testing the stem in my hotswap keyboard, it registers a key press on  the upward movement of the switch. In other words, it registers two key presses. I am not sure where the issue lies, but I am going to take a closer look and report back

 

The motion of stem is very good in the switch housing. Even as an FDM print. I want to give you an idea on how it is like, here is a link to a video of the stem in action: V4 Stem in Swtich Housing

 

 

[flemingt]

Thanks for the feedback.

Excellent that it now fits through the housing.

It seems from the images that I can pus the wings even further to the top to improve seating. What layer size are you using for the printer? I can estimate how much from that.

When your calliper does arrive would you mind taking a measurement of the side length of the stackpole piece? Maybe your FDM is undersized. We can figure out if a scale factor is needed.

I replicated the tiny points from the actual switch drawings, I'd guess that they'll be more robust when printed in resin.

Is it possible that a surface finish issue is causing the upward motion to register a keypress? Let me know if you find something when taking a closer look.

 

Thanks for the video too, the motion is quite good in the housing. I'm very happy at how this is shaping up. I figure a few more revisions once I have your feedback and it'll be done.

 

[+scorpio_ny]

Hi @flemingt,

 

Yes! I did get my new calipers! So if you need more measurements, please let me know. Here are the measurements that you requested:

 

           

 

I have been printing using a layer height of 0.16mm.

 

Edited February 8, 2022 by scorpio_ny
Spelling, grammar corrections

[flemingt]

Sorry I've been quiet, I've been waiting on parts. The first of which just came...

[+scorpio_ny]

1 hour ago, flemingt said:

Sorry I've been quiet, I've been waiting on parts. The first of which just came...

Looks great!!

[ivop]

Looks great indeed! And the serial number box is white this time (you can draw on it with a sharpie).

[NuClear235]

17 hours ago, scorpio_ny said:

Looks great!!

I miss 5pin switch holes in PCB...

[ivop]

8 hours ago, NuClear235 said:

I miss 5pin switch holes in PCB...

Huh? I clearly see three holes for the plastic stabilizer pins, and two for the switch pins. Contrary to Rev A, which only had one hole for a plastic pin.

[flemingt]

Started soldering today. I should have gone for 5 pin switches instead of 3 to prevent some of the rotational misalignment, but it's what stock that was available!

Alignment of the holes with the metal backplane is pretty good. I'll tweak them a bit in kicad and uprev the github next weekend or the weekend after, I can already foresee some other routing changes I'll need to bake into rev 3.

Once I get new switch internals and test fit in the chassis I'll work on the spacers to get the keys to the correct relative height. Hopefully spacers will allow for proper routing of ribbon cables.

[ivop]

14 minutes ago, flemingt said:

Started soldering today. I should have gone for 5 pin switches instead of 3 to prevent some of the rotational misalignment, but it's what stock that was available!

You now experienced exactly why I advocated for 5-pin switches It's a real pain in the rear to get the 3-pin switches aligned ?

 

So cool to see this project being furthered! Thanks.

 

[+scorpio_ny]

11 minutes ago, flemingt said:

Started soldering today. I should have gone for 5 pin switches instead of 3 to prevent some of the rotational misalignment, but it's what stock that was available!

Alignment of the holes with the metal backplane is pretty good. I'll tweak them a bit in kicad and uprev the github next weekend or the weekend after, I can already foresee some other routing changes I'll need to bake into rev 3.

Once I get new switch internals and test fit in the chassis I'll work on the spacers to get the keys to the correct relative height. Hopefully spacers will allow for proper routing of ribbon cables.

Hey @flemingt,

 

It looks great so far! it would be good to see how it fits into the case.

One thing to keep in mind: If you solder the switches in, it will make it a lot more difficult to pop open the switches if you intend to replace the stems.

 

Also, I picked up a 0.2 mm nozzle for my Ender 3. I am hoping to get better results for printing the stems. I will be printing some tonight to see how it goes.

 

 

[flemingt]

Sorry for dropping off the face of the earth. I knew work was going to get busy so I was eager to solder the switches. 

I just spent some time this morning testing out the keyboard pcb, and it works!

The connection is a bit ghetto at the moment but I'm happy. 

Next step is to do about more CAD for the space bar stabilizer. Move some of the holes relative to the metal backplane a fraction (though I may not use them for mounting).

I think I'll try recommission my FDM printer and make some of the adaptors so I can get a feel for the keycap height relative to the chassis.

Last thing to do, when it's all buttoned up will be to desolder the keyboard ribbon connector and put some headers in its place.

 

[ivop]

8 hours ago, flemingt said:

I just spent some time this morning testing out the keyboard pcb, and it works!

Very nice! And the LED glows

 

The project was named after my grandmother, Helena Maria, who died in 1986. I think she'd be proud. Two life changing gifts were given to me by her. A Rubik's Cube, and a Nintendo Game & Watch Donkey Kong II. Soon after that my dad bought us a VIC-20, and two years later an 800XL. The rest is history

 

Edit: are this 5-pin or 3-pin switches?

Edited March 18, 2022 by ivop

[flemingt]

Thanks for sharing that story, the name will definitely stay as tribute.

 

I've made some more changes to the PCB that I need to upload to GitHub. I moved the connector further north on the board to stop an overlap with the cable connecting the daughter board. I also tweaked the position of some of the backplane holes.

 

Unfortunately my printer was fighting me today, so no progress on the switches.

 

They're 3 pin but the holes are there for 5.

 

[+mytek]

2 hours ago, ivop said:

The project was named after my grandmother, Helena Maria, who died in 1986.

My mom's name was also Helena. She was from what used to be called Czechoslovakia, was deported to Germany following WWII, and met my dad when he was stationed there with the Air Force. She passed away in 1994.

 

Never got any Atari stuff from either her or my dad, but that's because I'm a tad older than you. Atari wasn't a thing back when I was still living with them.

 

Keyboard project is looking really nice ?