ScreamingAtTheRadio

OK, quick update. I added a 26 1mm FFC connector that should be compatible with @ZuluGula's adapter. Please do check that I got the orientation right since there is some ambiguity about the orientation of the ribbon on each side. On the keyboard side, contacts 6 and 7, and contacts 8 and 9 are merged, which should translate on the Atari motherboard side in the reverse, 18-19 and 20-21. That's if I understood correctly. Please correct me if I'm wrong. This should be clear from the new schematic. I'll also build an adapter for the Atari side that will have FFC and the same 2 row connector as on the keeb side. That should make it easy to use clean cables or header cables as one prefers. I've also looked at PCB plates, and that's made difficult by the fact that square holes are not that easy, I'll have to design holes for the keys that are not exactly rectangular but take into account the rounded corners that come with routed inside edges.

Absolutely, that's a great idea, I love it. I'll keep both options in, so you can go to any option you prefer, from header cables to the nice small ribbon. I like the idea of the adapter on the Atari side, and may build a similar adapter for the style of ribbon cable I had originally selected.

I ended up doing things in reverse order from my plans... Here's a draft of the Rev. C PCB. Main changes are the space for an optional Pi PICO + headers for use as a PC keyboard for emulation, and a 2-row connector for the connection to the Atari board. The reasoning behind that is that one end of the ribbon cable will be in a more reliable and easy to make connector. On the Atari side, we don't have as much choice, and we have to go for a single row, so the individual lines in the ribbon have to be separated at that end. The cheapest cable you can buy for this new version can still be a bunch of header cables, but you'll be able to build a sturdier cable if you want to. I also changed the outline for a better fit inside the XE case, and improved the silkscreen so pin numbers are on the back, and other help text is duplicated on both sides. Otherwise, same board. Let me know if you have feedback, this is not going into fabrication before next week.

I can add the stl, it's just that it's a generated file and as such, one more thing to push every time I make a change, but now that the design is stable, no problem doing that. Eventually, I'll also push that stuff to Thingiverse. Stem: 3d-junkyard/keeb-stem.stl at main · bleroy/3d-junkyard (github.com) Space stabilizer: 3d-junkyard/keeb_costar_stabilizer.stl at main · bleroy/3d-junkyard (github.com) For fabrication best results of those on a resin printer, I recommend adding support and turning the model 30% from the vertical.

Some updates on next steps of this project: Blog post detailing reproducing and assembling the current state of the project. Will aim for writing it this week. Emit fabrication files for a PCB version of the plate and order a batch of that. Also this week for the design, will go for a minimal batch of 5 together with the PCB rev 3 to minimize shipping, happy to do more if more than 2 people want one. Will sell for fab price + shipping and packaging. Update the 130MX PCB to better fit the Atari XE cases, fix a couple of silkscreen issues and allow for an optional Pi Pico controller to use the keyboard with a PC with less wiring. Probably next week if other projects allow. Will also have 5 or more done, let me know if you want one. For either the plate PCb or the main PCB, let me know in PM if you want one.

Sorry, no, I'm happy to help with models, information and tips, but I unfortunately don't have enough free time for that. I'd advise the following: 1. find a local maker space with a resin printer. 2. buy one of the cheap resin printers out there (I use an Elegoo Mars Pro 2 myself -the Mars 3 is coming out soon- and it's ridiculously cheap for the quality of the prints; I do realize it's still a significant amount so you may want to try one of the other approaches if that's the case). I also understand resin printers have some drawbacks that make them unsuitable for some homes. 3. organize a group buy of sorts. Option I would not advise: use an online 3D printing service: those are prohibitively expensive and would cost you more than a printer. This is pretty niche so I doubt enough interest could be built to justify an injection molded batch (just mentioning this to be thorough).

I've experimented with 3D printing for smaller keypad footprints and the results even at smaller scales are not great in my experience: 3D printing is not very good for large flat surfaces because even with support the material tends to sag. I also think any plastic is going to be too flexible for that size. Finally, you'd have to use a printer with an exceptionally large bed, or print in multiple parts. Alignment would probably be fine, but the result in terms of rigidity and structure will likely be inferior. Happy to be proven wrong. The best material is definitely a metal that's not too flexible, so steel is perfect in that regard, it's just that cutting it is expensive. I don't know how expensive other options such as water jets, plasma or machining would be but I'd be surprised if it wasn't in the same ballpark. As for corrosion, that sure is a valid concern with steel. What I've done is sand the surface and give it a rub with WD-40, which should preserve it for a couple of years at least. Then again, it's structural and even some superficial rust on it wouldn't be a huge deal. The last option, definitely the cheapest that would give good enough results, is to have a PCB cut to the same dimensions. I haven't tried it myself on this particular keyboard but it's a common technique for custom keebs so it should be no problem. You should be able to use the same file I used for the steel plate and make a manufacturing file that any PCB shop will take. Drilling more holes always brings the price up but it should still be a lot cheaper than laser-cut steel. Another option could be laser-cutting plastic, but I'm not sure what the right material would be. I wouldn't use acrylic since that's too brittle and would almost certainly break when fitting the switches. Any plastic less brittle that can still be laser-cut would likely be too flexible, especially at 1.5mm. I'd love to hear if folks think they know a material that would work well. I'd still go with steel as first option, and PCB second.

Yes, I'll make another video at some point with just the sound of the keys, without the Atari clicks. The switches are quite clicky. I've been in the business of selling low volume kits before and in no hurry to do it again Tell you what though, I'll write a full-length blog post going into the details of making and sourcing the parts as well as assembly steps, so anyone can reproduce what I've done with no difficulty and under a reasonable budget.

That is an excellent question. I haven't counted what it cost me since I had to do many iterations of most parts, but if you started from the files in their current state, it would go something like this: * Resin: a few dollars at most, but you need access to a resin printer. Online printing options are prohibitive. * PCB: $31 for 5 pieces with taxes and shipping from JLCPCB (I have a few spares BTW). * Metal plate: this is entirely optional but adds a lot of value to the project. It's expensive, $65 from Ponoko for 1.5mm laser-cut steel. * 62 Kailh Box Pink switches: you can find bundles of 90 switches for around $33. * Costar stabilizer: you have to print your own for the space key but the ones for the switch key can also be bought off-the-shelf for less than $10. * Some wires and pin headers you probably have around your workshop or can get for a few bucks. * A donor Atari keyboard for the keycaps. So in the end, probably around $110 if you can group-buy the PCBs, which is not bad for a custom mechanical keyboard.

My Atari 130XE keyboard rebuild project is finally done and I'm excited to share it with you! There's been a number of previous projects to put a modern keyboard on Atari 8-bit computers. I don't think anyone's done it while preserving the original keycaps though (would love to be proven wrong). This is exactly what I did though, by 3D-printing modified stems for Kailh Box Pink switches that take the Atari XE key footprint instead of the Cherry MX cross. Cheap resin printers can print at an amazing 50 micron resolution, which is enough to print very small parts with very fine details. Switch stems are such parts, and must be printed very precisely to reproduce the touch and feel from the original. It took a lot of tinkering to get there, and trying a lot of different resins (in the end, Elegoo black and translucent red were the most reliable), but the keyboard feels absolutely great. What's more, you can't tell the difference from an aesthetic point of view... Check this out: This is built on a new original PCB design, a custom laser-cut steel plate, Kailh Box Pinks and Costar stabilizers. The stabilizer metal parts are also preserved, but the one for the space bar has a diameter that's considerably larger than off-the-shelf Costar stabilizers can handle, so I also 3D printed new ones: I have a video showing the end result in action: All files for this project can be found here: https://github.com/bleroy/3d-junkyard/tree/main/Atari130MX Let me know what you think!

The board has been claimed.

Hello everyone. I wanted to show you all what the board looks like in black, as I just received a new batch made in this color. I'll start building this week-end, and will post updates here as it progresses. This is a batch of 5, 3 of which I'm keeping for myself and a friend. Another is already sold, which means there's one remaining, that I'll send to the first who direct messages me (don't respond/spam here).