deskthority

I just started this hobby of collecting and modifying vintage keyboards in November of 2019. It has been immensely rewarding so far. I enjoy most aspects of it for a many different reasons. I enjoy trying to get reuse out of something old especially in this case because the construction was of such higher quality. I enjoy trying to find my own recycler streams like Wodan has. I enjoy the problem solving that comes with deciding what to do with each acquisition. I imagine like many that I start way too many projects and honestly have yet to finish any.

I decided to put forward this project to share how it has evolved as I've pursued it and, if anyone finds it interesting enough, to get help as I continue.

I have a video playlist that I add to: https://www.youtube.com/playlist?list=P ... ylPHigCb8y
and an imgur post with some images: https://imgur.com/a/KqFJNDP

Hope you don't mind if I ramble a bit before starting in on my LK201.

Since I'm so new to the hobby, I've spent most of my first few months exploring some of the cleaning aspects. I'm a full poser in the sense that I'll read and watch what others have done, then buy a set of tools that I decide I want to use and then, hopefully, get around to actually using the tools that I've purchased. I'm at a point in my life where I want a hobby that gives me joy and if a small tool is under $20 and I think I'll use it I get it, and if a bigger tool is around a couple of hundred then I may invest in that too.

Wodan videos were my early inspiration. One of the early things that I bought was a decent sized ultrasonic cleaner. I've ran many sets of keycaps through it. Another set up I invested in was for retrobrighting. I probably went too far on this one. I bought a grow lamp, a sous vide, a giant aluminum roasting pan, and a roll of tinfoil all for the heat and light environment. I bought a vacuum sealer that conveniently is wide enough to put keyboard case halves in along with hydrogen peroxide. I told myself there would likely be some reuse. And there has been a little, sous vide for cooking and retrobrighting setup gets reuse for post curing resin parts. I've never been a tool guy so I think that is half the fun with this hobby, thinking of an approach to a problem and getting some tools and just trying it out to see if it works or not. Very rewarding to me.

My thoughts on Retrobrighting are currently that yeah it can restore nicely in some cases but it can also come out with even results. Even in the successful case, beige is kind of boring. It got me thinking that doing something else with the old cases might be better. There are several options. One I was introduced to at the Seattle winter keyboard meetup was an early teenager that had hydro dipped his case. It was cool and artful. Then I end up watching Adam Savage's channel and seeing lots of crossover to our keyboard hobby in people that make props and build all sorts of things. The idea that I became most attracted to was cerakoting. So in full poser passion I've bought a DIY setup including sand blaster, air compressor, smoker, cerakote starter kit. I'm still waiting to use it because I'm trying to get some outside my house space to do it. I won't have a large proper setup but we'll see if will work, fingers crossed.

Another idea I'm trying to pursue is to reuse a beautiful case by trying to put cherry mx or alps guts into it. This is especially true for the membrane and rubber dome vintage boards that look beautiful but the guts are not. I have a Delphi logo keyboard with maxi switch dome with slider guts. I didn't know how to approach doing this so I thought about it and assumed I could start by making a metal plate that I can mount to the keyboard top and then go from there. I used keyboard-layout-editor to swillkb to lasergist and got a $35 plate in the mail a month later. It was too wide so I decided to get a dremel setup for that and other reasons. I have yet to cut the width down but I just got a rockwell bladerunner2 saw that I think I'll be using to do the job. The other project at the top of the list here is a silicon graphics nmb rubber dome. It's curved and everyone says that brings a lot more challenges but I really like the look and hope to change the guts to cherry mx. The yellowing due to age caused it to go from a gray to a pea green soup color. Retrobrighting really did a nice job of bringing it back to gray. The case gray now looks great with to 2 shades of gray alphas and modifiers. Everything is great except the innards!

The main reason I want to share all of my ideas up front is that if anyone else has similar pursuits or want to discuss, please PM me. I love collaboration and am looking for more of it. Thanks.

Now on to the DEC LK201BA. What caused me to fall in love with it was the different colored double shot keycaps. Most of them are words on the num pad with actual numbers as second class citizens on the side!



I don't even use a num pad but that doesn't bother me at all. I'm attracted to cool colors and think it will be fun to resurrect none the less. Also, there's a cool 2u 'Do' key that would be great to reuse.

I knew that I was never going to restore this board in original condition because it's a membrane board with giant cruciform mounts in the keycaps.



Now is a good time to say that this keyboard is in my top 2 of the most over engineered boards ever. So many pieces and so many chads, i.e. broken tiny pieces during the teardown. The plastic assemblies had feet that were "heat stake" mounted to the plate which seems like a plastic equivalent to metal rivets. I broke all those off.



Each keycap had 4 plastic stems protruding from the normal bottom of a keycap, many broke when removing the keycaps.



I had a giant pile of metal springs and plastic assemblies as well when I was done.




So another idea of reuse is to reclaim the keycaps and use them on another keyboard. Recently, I decided to start on on getting my reuse out of these lovely keycaps. I didn't know the name for it at the time but finally stumbled on the term "stem mod". Still didn't find much info but found a little. I decided that the best way to remove the existing keycap stems was to use my dremel with the tiniest wheel for cutting plastic. I sat down to do it and I did one keycap but it sure made a mess and wasn't too fast.



I currently still have no idea if that technique would work well enough or if I'd have enough patience to grind through a full set of keycaps. The reason I don't know is because I recalled that the teardown of the entire board proved how brittle many of the parts were and I realized that the most efficient way to remove the keycap stems was to simply break them off. I think this technique worked because of how particularly long all of the stems were, unlike most keycaps. This epiphany was one of those rewarding moments about the hobby that I've mentioned before. Coming up with your own solution that you didn't read in a manual has its own rewards! Might be trivial to most of you but it makes me happy!

I was able to snap certain colored double shots bare handed. For the rest, I put them in a vise and used a screwdriver as leverage and covered it with a cloth so that the parts weren't flying off into my face.



I just realized that I'm probably not even half way of getting my words out there but it seems like a good time to post this start and then continue writing the remainder.

So for the next stage of the journey I want to discuss the idea I had for doing the stem mod. I thought of it ahead of or at the same time I had decided to dremel and break the old stems off. What led to my idea was the irregular surface that would be left at the base where the stem attached to the keycap. I'd have to try to dremel or sandpaper it smooth in such a confined space. It looked to me that Ripster had maybe done something like that. Anyways, I thought of the problem from a different angle. I could ignore the irregular surface at the base if I suspended the new keycap stem over the middle of the upside down keycap and then fill it with enough glue to hold it in that exact spot.

So that led me to figure out if there was an easy way to 3D model the idea I had in my head. I stumbled upon Meshmixer somehow and watched the videos and it was easy enough so I started that exciting part of my journey. I didn't know or realize that the piece I was designing to hold the new keycap stem suspended over the middle was called a jig. Now I do, haha. my concept is that the if I could overlay a jig on the upside down keycap with a cherry mx switch stem protruding from it, then I could attach my new cherry mx stem to that and then pour in glue, let it harden, and simply remove my jig and the new keycap stem would remain, glued in place. I still don't know if this is going to work but that was and still is the premise of my idea.

Here's a couple pics of my Meshmixer jig:





And here's a pic of my replacement cherry mx keycap stem:



I ended up putting key dimension numbers on the outside of the parts so that I could tell one result from another as I was dialing in the dimensions.

Around this time I was trying to figure out how/where to print my models. I didn't want to have to do cloud printing because of the long turnaround time. I asked my son to see if he had access at his university but to no avail. A few years had passed since I had tried 3D printing with a printbot jr. I wanted to see how much the technology had advanced and prices had dropped. Somehow I stumbled on resin printing and instantly fell in love. The downside of working with resin is not great, I'll admit, but the upside of getting high resolution prints of small objects is wonderful. I bought a Mars Elegoo MSLA printer and started cranking out prototypes and iterating on them.

Here's a pic of my printer and post cure (using my retrobright setup).



Here are some pics of my resin prints hopefully conveying my idea a little more. Sorry they aren't great quality pics. From my usb microscope.

Step 1: Hold the jig



Step 2: Put the new keycap stem on the jig



Step 3: Place the keycap upside down



Step 4: Place the jig and new stem over the upside down keycap



Step 5: Squirt glue inside, enough to surround the base of the new keycap stem, and let it harden

No picture, haven't glued any for real yet.

Step 6: Once glue has hardened, remove the jig leaving the new keycap stem centered in place in the keycap



I have researched and bought the glue that I'm going to use. I chose 3M Scotch-Weld Acrylic Adhesive DP8410NS after watching a video showing how it bonds tremendously even to non-roughed up plastic surfaces with oil on them. Didn't know what I was getting into because I also bought the 3M gun for 10:1 delivery and a set of nozzles for that glue. I don't really know chemistry and glues but I trust that 3M does so sticking with that for these keycaps.

I was actually about to start doing the stem replacement on my DEC LK201BA keycaps but then I realized that there's a lot more to resin printing. I'm actually using prototyping resin from Elegoo and I've read that there are different resins depending on what you are going for such as pliability, strength, etc. I decided that I want to use Siriya Blu as a strong engineering resin to print the keycap stems that I'm going to glue into my keycaps. That decision has a cascading effect that resin is temperamental when it comes to temperature. Siriya Blu does best above 25C/80F so I have to make a DIY heater/thermostat to put inside my elegoo mars hood before proceeding. Parts have been ordered.

This is about the end of part 2 of my journey. There are 2 things I will mention to give a sneak peak to what's coming next:

One is that in making my jig and getting dimensions, etc. I was shocked to find out that the DEC LK201BA keycaps are non-standard dimensions. I had thought that all keycaps were 18mm x 18mm. The DEC keycaps are 18mm wide but 19.6ish long in the vertical dimension. This kind of threw a wrench into my plans to use these caps on another cherry mx board.

Two is that I made 11 revisions in Meshmixer to dial in dimensions and to do that I had to start with the 2D dimensions then extrude out my 3D solid because you can't edit the solid like that. Super tedious so I started looking for a better way.

To start off this part I want to share one of the greatest epiphanies I've had in recent memory. I realized after searching that what I was missing in Meshmixer was parametric 3D modeling. I did my research and chose Autodesk Fusion 360. It's free for hobbyists, has great cloud collaboration, and miles of youtube videos! It was scary at first because I remember being intimidated by it 30 years ago. I wanted to make sure it wasn't my imagination so an internet search showed me that Autocad originated in 1982. First of all I recall thinking it would be hard to learn. Second of all, I wouldn't be able to do anything with the output back then because it would all have been large machinery and I wasn't anywhere near any. Fast forward to today. I was blown away with how easy it was to use and the combination of that with the ability of my $220 MSLA to make high quality prints of my 3D modeling designs had an immediate effect of changing how I see and think about the world. Mind blowing. I can look at anything small in the environment around me now and instead of wondering how they made that in some plant in China, I'm pretty sure I can make a 3D model of many things I see.

Here is my quick overview to 3D modeling. There are 2 sides to this modern stuff. There's the free flowing stuff like taking 360 degree images of an object then getting it converted into mesh and touching it up and printing. I'm not particularly attracted to that part. Then there is the traditional CAD/CAM side which relies on the more traditional way of laying out rectangles and circles in 2D and then extruding them to 3D. This is the part I've been using. In Fusion 360, you make a sketch that defines some 2D elements that you are going to use and then you extrude them into 3D.

Now back to my LK201 journey. I realized that to used my extra vertical length keycaps I'd need to start making a keyboard layout to accomodate that. One realization I had with my resin printer is that I don't need to send away for a $35 metal plate until after I've prototyped it in resin first to get the dimensions and shape I want. I would have to slice the plate into small enough pieces that each fit in the build volume of my printer and then glue the pieces together later. I have yet to get that far along. I did print one chunk 5mm thick before changing ideas. The idea was my old one which was to design a cherry mx plate with the dec layout to screw to the top part of the original case. So I made my layout in KLE.



And then I brought that data into swillkb and exported a DXF file from there that Fusion 360 could open. I then extruded the 2d plate to 5mm and chopped a piece based on the build volume of my 3d printer.





I then sat down to eyeball the results and start to calculate new dimensions for the plate to stretch vertically to accomodate the extra vertical length of the keycaps.



As many of you know, the DEC LK201 has a curve plate setup. I believe I could have made a flat plate by shaving off the rounded parts of the case top. But as I was looking staring at things, along with my new found of looking at the world in which I can 3D model and print anything I want, I came upon the idea of simply making new plastic assemblies with cherry mx switch holes in them. I thought of it because I was wondering about reusing the metal plate and noticing it had all the cutout holes and that the plastic assemblies fit in those holes and are what aligns that extra vertical space that those keycaps take up.

So it blew my mind again that 1 hour of design and 1 hour to print and 1 hour for secondary cure and I was staring at my equivalent of a plastic assembly. And I tried it and it fit because all the measurements are on even boundaries. Easy math.







I then got better at how to factor things in Fusion 360. The plastic assemblies are made up of 1 of 4 objects. A normal switch piece, a left switch piece, a right switch piece, or a blank piece. I have a normal switch sketch that defines everything I need. I then copied that component (with linking) several times to add to the sketch to cut or extrude as necessary to make each of the others. I then made the components that copy in the right combinations of those to make the final result pieces.

Here's what my normal switch sketch looks like:



and what the bodies extruded from that look like:



The other slight variations I won't show individually because I'm already showing too many pics.

Here are 2 of the plastic assemblies that use those 4 individual switch parts.





So that's pretty much where I'm at so far.



By choosing the plastic assembly route I'm staring at an upcoming hardwire rather than a custom PCB. I've never done either so that'll be fun. I ordered a bunch of Amoeba single switch PCBs to see if I'll be able to use those. I've liked what I've seen about them over the old school hand wire.

I'm sure I'll run into a bunch more problems. Again, if you're interested at all, please chime in with advice or PM me if you want to collaborate on anything.

Thanks,

Kralcifer

You are doing an amazing job. Thanks for sharing.

This is great. The quality of those prints is amazing. I think programs like Fusion360 are great. I've been trying to move to a program like that for modelling and once you get beyond the initial learning curve they are very powerful. This is very impressive and I can't wait to see how it turns out

I try to redo my fusion sketches as I learn more. I redid the cherry mx plate hole sketch with the minimal amount of constraining that I'm aware of.



User params for outermost width and height. Then 5 other lengths based on their relative amount to the outermost. I mostly used the equal constraint on all the others along with perpendicular constraints.

Should never need to change sizes but for example doing something like sneakyrobb is doing, making an 40x larger Big Alps switch should scale everthing appropriately.

Fusion has many places to get lost. One is that I wanted to link that clean sketch of cherry mx plate hole into other designs. What happened to me is that I was organizing my projects by folder but that prevents the easy insert method. It only works across designs if they are in the same folder. So I moved stuff out of the folders into just a single folder and I renamed to more descriptive design names.



So if you don't see an 'Insert into Current Design' in the right context menu of a design in the left data panel, that's likely why.

Finally did my first cerakote last night. It was a fiasco but the result is promising. I think it could have great results in spite of my DIY setup.

I used a Dell crap modern rubber dome as as my sacrifice
Here's the top, normal color



Here's the bottom cerakote H-Series Smith & Wesson Red, unevenly painted but it's a start



Cerakote is a wonderful technology. ceramic paint layer .001" thick. super durable and resistent.

haha, thanks!

I tried sandblasting for the first time earlier this week and that went fine. Last night though, man... Here's the fiasco from last night in detail:

I shook up my color and added it with the bonding agent and shook that and then when I went to poor it through the cheap paper cup 150mm strainer into the HPLV gun, I poured too fast and it shot went through the strainer above the HPLV cup top and put a 3 inch path down my knee to my ankle!

I set up a super budget paint booth by cordoning off a third of my shed with some sort of plastic painter drop cloth. I had some vent filters covering holes in the side of the shed and I had a shop vac hooked up outside to draw the air through the air filter. Well, when I went to start painting I had forgotten to turn on the shop vac. Needless to say, there was virtually no ventilation so the air was filled with the paint particles. I had a 3M organic vapor respirator on but man, the filters are both pink due to the red paint in the air. guess I'll need a new one. I think I'll spray outside in the open next time to see how that goes. Felt pretty dangerous.

All the cerakote videos show pristine clean environments and tools. I could never find how the heck you are supposed to clean up the mess so that was a big joke as well. I tried pouring Acetone over my HPLV gun to get the paint off. I did it over a bucket that had a bottle container sitting in it with a plastic funnel in the opening at the top. Problem is it was wobbly and I needed 3 hands and you have to scrub, etc. What a mess. I couldn't get all of it out of anything. The glass beaker used to measure amounts of cerakote and bonding is streaked / smeared red, the HPLV cup and gun are similar. I don't know if it was all supposed to clean up nicely but it sure didn't so I guess that's the norm at least for me.

kralcifer wrote: ↑

05 Aug 2020, 20:54

I used a Dell crap modern rubber dome as as my sacrifice

The original Dell Quiet Keys (before they started making them in China) were actually some of the best rubber domes.

fohat wrote: ↑

05 Aug 2020, 21:38

kralcifer wrote: ↑

05 Aug 2020, 20:54

I used a Dell crap modern rubber dome as as my sacrifice

The original Dell Quiet Keys (before they started making them in China) were actually some of the best rubber domes.

Oops! I think I did sacrifice a decent one. It had the individual rubber domes like my SGI keyboard has.

kralcifer wrote: ↑

06 Aug 2020, 00:38

fohat wrote: ↑

05 Aug 2020, 21:38

kralcifer wrote: ↑

05 Aug 2020, 20:54

I used a Dell crap modern rubber dome as as my sacrifice

The original Dell Quiet Keys (before they started making them in China) were actually some of the best rubber domes.

Oops! I think I did sacrifice a decent one. It had the individual rubber domes like my SGI keyboard has.

says Made in Thailand on the back

I have had them with blue (light and dark blue) and also yellow individual domes on a translucent flat sheet. My experience was that they were particularly nice for rubber domes. The later ones made in China were not nearly so smooth.

This dell had light blue and my SGI had yellow.