The Alumaplop, a compact keyboard/trackball combo in process

[Zekromtor]

PCB blanks aren't too expensive, I wouldn't sweat the loss. They look pretty damn professional when completed too, at least the resolution we can see them at. I'm interested to hear if you can solve the hole alignment issue and whether you can determine if it was stretching of the toner/paper or not. Let us know when you get more done on it. Still my favorite project on here

[tufty]

For your casting issues, and depending on what resin you're using, you might want to try a 15-30% by weight addition of glass fibres to your resin before pouring.  30% can take you from "brittle" to "unbreakable", and you can apparently get them down to ~0.5mm length (I've been using 3mm, but I'm casting much bigger pieces).

[kurplop]

PCB blanks aren't too expensive, I wouldn't sweat the loss.

I'm glad to hear you say that. I'll send you the bill for them.

They look pretty damn professional when completed too, at least the resolution we can see them at.

Thanks. With each attempt, they seem to improve in quality. I must be learning something but I'm honestly not sure what I'm doing to make them better. I think adjusting the trace separation has made the etching crisper. Have you considered the possibility that I'm not that bad a photographer and the poor resolution is a smokescreen to hide my limited PCB skills?

I'm interested to hear if you can solve the hole alignment issue and whether you can determine if it was stretching of the toner/paper or not. Let us know when you get more done on it. Still my favorite project on here

I printed new toner copies the other day but haven't checked them for dimensional accuracy yet. I'll let you know when I do. I'm guessing that it's happening during the ironing process though.
I'd like to solve the problem but it's not concerning me too much because the the hole placement is determined by the mill's DRO, so as long as the pads fall around the holes, it should still work.

For your casting issues, and depending on what resin you're using, you might want to try a 15-30% by weight addition of glass fibres to your resin before pouring.  30% can take you from "brittle" to "unbreakable", and you can apparently get them down to ~0.5mm length (I've been using 3mm, but I'm casting much bigger pieces).

Thanks for the suggestion. Can you tell me where I can source the glass fibers?  I guess I could just pull a few batts out of the attic. ) Also, does the fiber affect the viscosity of the resin or are there any other possible negative issues using it? I've read that the curing time is retarded with thin castings because of less heat generated during the chemical reaction. Do you think that could also affect the ultimate tensile strength of the fully cured product?

[tufty]

Thanks for the suggestion. Can you tell me where I can source the glass fibers?

Mine came from the local builder's merchant - at 3-5mm they're sold as strengthener for cement.  Those are probably too long for you, though, especially in the stems.

You probably want something like this : http://www.fibertecinc.com/microglass.htm or http://www.fibreglast.com/product/132_inch_Milled_Glass_Fibers_38/Fillers

1/32" is 0.8mm, that should do you fine.

It does affect the viscosity somewhat.  It shouldn't affect surface appearance as long as it's well mixed, but you might want to consider slipping some gelcoat or pure resin on the cap areas before pouring "just in case".  You will *absolutely* need to vaccum your resin before pouring, mixing the fibres in will bring in a lot of air.

My process is this:

- My material is a 2:1 resin - hardener mix.  30% fibre is about the same volume as half the resin...
- Split resin into 2 equal volumes.  Mis fibres with one volume of resin, go slowly like mixing eggs into flour.  Once it's mixed, hit it up with a vacuum to get the air out, then add the rest of the resin, remix, split again if necessary, and hit up with vacuum (again).
- Mix hardener into resin, vacuum (again, wasn't that fridge pump a good find?), pour.
- Bung the mould into pressure chamber, hit the whole thing with 2 bar (my pressure chamber is a DIY bodge, I'd pump it up to 10 bar but I'd probably get debris in my eye or something) and let it sit.

[jacobolus]

http://lcamtuf.coredump.cx/gcnc/ch4/

[kurplop]

Not much to share but I did get the blank PCB's in the mail and they are ready to prep.

It looks like the printer is making dimensionally accurate copies. It looks like the ironing process might have been the problem. Any suggestions?
 
I'm hoping to have time this weekend to etch the new PCB's and solder up the components. I'll leave the keycap issues for another day.

[Oobly]

Sorry, haven't had time to take pics of my process for using baking paper and laser printer for making PCBs, but it's essentially the same as this: http://www.instructables.com/id/Toner-transfer-no-soak-high-quality-double-sided/?ALLSTEPS

I just use an iron instead of the laminator, since I don't have one. It works quite well if you have a dark enough print and you can see the toner colour change from the back side of the backing paper as it adheres to the PCB and loses hold on the baking paper.

I also use 3M Magic tape at all four corners of the baking paper rather than just the top edge to prevent it folding and the tape is very thin and seems to handle the process well.

Have you lubed your ML switches? I have found this to make a massive difference to how they feel and they can handle off-axis presses like a champ, too.

By the way, this is looking amazeballs!

[CPTBadAss]

@oobly, my happiness level this morning > 9000. Thank you for posting that link on how you etch custom PCBs! I've been wondering how GHers have been doing that for a little while .

When you say iron, do you mean a iron for your hair? Just trying to understand the details.

[Oobly]

@oobly, my happiness level this morning > 9000. Thank you for posting that link on how you etch custom PCBs! I've been wondering how GHers have been doing that for a little while .

When you say iron, do you mean a iron for your hair? Just trying to understand the details.

glad to help.

Clothes iron, nice flat surface to heat the toner with so it melts and adheres to the PCB copper and "let's go" of the baking paper.

[kurplop]

Oobly, thanks for the link. It contains a lot of good ideas.

I ordered some baking paper in the event that my paper stretches again. I refuse to fail a fourth time!

tufty, thanks for the info on the fibers. I'm going to try a few more times without glass with fresher resin first because I still have some very fast setting stock that probably would kick off before I could thoroughly mix the glass into it.     If it still too weak, it's good to know there are other options.

[kurplop]

It was a fairly productive weekend. I got the PCB blanks in the mail and once again used the toner transfer method to prep the board for etching. Unfortunately, the ironing process is the problem. I had almost the same expansion as the last time. It will work fine and once again, the holes I drilled out are all within the pads so it shouldn't affect the operation. I did try the silicone baking sheet method and consequently learned how to remove the toner roll to remove the jammed up paper in the printer. I was warned not to try it with the Brother laser printer. I decided to stick with the brochure paper.

As you can see, I got plenty of PCB stock.


The company I got the PCB's from had a very fast delivery and their prices were very reasonable.


The pre-etched PCB. Notice the areas that didn't transfer well and I used a Sharpie on. Also the areas I scratched the toner from where the traces were a bit close together. Other than that, ready for etching.

The etching went well and I'm currently soldering things up. Pictures to follow soon.

I got a great deal on some Cherry ML circuit boards with switches and keycaps.
It came out to about 7cents per switch/cap shipped!
 

I'll share my keycap stem solution later this evening.

[Zekromtor]

Can't beat that price for switches. Looking forward to more pics.

[berserkfan]

This is one of my favorite threads.

Also Kurplop's hands are some of the least geeky looking. They look like construction worker hands. Quite interesting.

That said, using ML switches... very sad.

[Zustiur]

This might be a silly idea, but irons are designed to stretch fabric as part of the crease removal process. Perhaps a stream press would be the solution?

[VoteForDavid]

Regular iron with the steam turned up high, with a sheet or two of (t-shirt?) material between iron and paper to reduce friction?

[kurplop]

This is one of my favorite threads.

You just made my day.

Also Kurplop's hands are some of the least geeky looking. They look like construction worker hands. Quite interesting.

My spouse thinks that my rough, burly hands are a turn-on. By the way, I'm a woman. (Just kidding)

Actually, I am a construction worker. I hope that my GH membership isn't revoked, now that this info is out.

That said, using ML switches... very sad.

I know. I seem to be always bucking the trend. I also like caps lock and will make sure it is in a prominent spot on my keyboard.

I chose ML switches for their compact size. I wanted to remove as much bulk from the keyboard as possible. I wasn't willing to part with the big trackball however,  in spite of the compact size goal. I tried using a smaller one in my last keyboard and it just wasn't working well for me.

I don't mind the ML experience. In fact I prefer their shorter throw. The only real reservation I have with them is that they seem fragile. I have found that on several occasions, I have accidentally removed the stem and switch top with the cap I'm trying to remove.

This might be a silly idea, but irons are designed to stretch fabric as part of the crease removal process. Perhaps a stream press would be the solution?

I think your summary is right on. I wonder what is available in a compact size?

Regular iron with the steam turned up high, with a sheet or two of (t-shirt?) material between iron and paper to reduce friction?

That might help but I'm already having trouble getting enough heat to transfer the toner with an iron directly applied to the paper. It seems to take a long time to complete the transfer.
By the way. Welcome to GeekHack!


Now for the new business.

I mentioned earlier that I was having trouble with broken stems. Some of you offered good solutions but I came up with something that was free and readily available to strengthen the stems.

Notice the trimmed paper clip in the corner of the mold.


It fits nicely in the stem recesses in the mold.


Here it is in the keycap. These are really tough stems! I tried to pulling one apart with pliers and couldn't break it without completely destroying the entire cap.


Here they are nestled in the keyboard. They are pretty rough looking compared to the caps I made for the PortaPlop. I was more concerned with getting something in place so I could move on . I'm going to remake the mold soon but first things first.


I have pictures of the caps I glued up to make the master for the new thumb keys but whenever I upload more than 4 pictures I tend to lose everything. I'll post them soon.

[neverused]

Awesome job! I always get excited to see a new post from you here.

[CPTBadAss]

Nice looking caps Kurplop!!

[Zekromtor]

My spouse thinks that my rough, burly hands are a turn-on. By the way, I'm a woman. (Just kidding)

lol

I hope the MLs work out for you in the long run. I haven't been completely satisfied with the leaf mechanism in either of the two alps brands I've tried. The smallest bits of dust/debris tend to wreak havoc with connectivity, causing multiple presses (not from bouncing), and I have to keep a can of air around to blow them out frequently.

[metalliqaz]

Seems like so much unnecessary work when you could just send the design off to a PCB fab and get it back perfectly the first time.

Love this project.  Keep it up!

[Zekromtor]

Seems like so much unnecessary work when you could just send the design off to a PCB fab and get it back perfectly the first time.

*projectile vomits*

[kurplop]

Thanks everyone for your words of encouragement.

Seems like so much unnecessary work when you could just send the design off to a PCB fab and get it back perfectly the first time

It probably does seem like a big waste of time and effort but I didn't see sending it to a PCB fabricator as an option since I can't seem to figure out the PCB design programs. I downloaded Kicad and it didn't take long to realize that I'm not familiar enough with electronics to figure it out.

My skills are more of the hands on type where I can physically manipulate a medium into submission. I find the trial and error and occasional successes very satisfying, albeit more time consuming. Don't misunderstand, I wish my PCB's were as clean as the pro's are. Every time a blob of solder flows beyond the pads because my traces are bare, remind me that my boards could be better. Maybe a competent PCB designer wants to collaborate with me?

I hope the MLs work out for you in the long run.

Thanks. I hope so too.  It is all about size. This picture pretty much sums up why I decided to gamble on the ML's


As promised, here are a couple of pic's of the caps.

The first shows the height of the lower (original) caps in relation to the height and profile of the upper (modified) caps.


The second is of the 2 caps I glued together then filled the side gaps to make the master for the upper thumb caps. Conveniently, the 2 caps were different colors. You can see how I over sanded the top profile to try to get the best feel. Unfortunately, where I burned through left enough of a line to be noticeable in the casts. Today I applied resin over the cap and will resand, trying to avoid burning completely through the resin. I will then remake the molds which had other problems I needed to address as well.


Sorry about the out of focus pictures and the dusty subjects. My shop isn't exactly a cleanroom.

I hope to finish wiring up the keyboard components this weekend. Then it's just installing the trackball, hooking them up to an internal hub and a few hundred other loose ends.



 

[Zekromtor]

I'm really curious about those switches now. I didn't realize how different they were from other cherries. I'm wondering 2 things:
1. Can you remove the leaf that provides the tactile bump to make it linear and still have it work properly?
2. As I understand it's 2mm of travel and activation is at 1mm.... does it look like it might be easy to mod to make the activation be closer to the bottom of the press?

[kurplop]

I'll take one apart and experiment on it but they are small switches and are difficult for me to work with. Cherry's spec sheet says they have 3mm travel with activation at 1.5mm. It may be possible to limit the bottom travel by inserting something into the center barrel.

[berserkfan]

Kurplop, you are awesome and I would like to see you succeed with the lowest possible difficulty.

Check with Fohat.digs, he's like you in that he doesn't know electrical engineering but is great with mechanical/ physical figuring things out. The two of you might be able to trade works, thereby increasing efficiency.

Bipiphany is fond of PCB design and sometimes helps people too.

I gave away a gigantic bag of ML blanks last year and have none left. But you get ML blanks in every Crap Bag from Signature plastics. Hundreds of geekhackers have Crap Bags. If you need keycaps, just post, and you will get more ML blanks than you can use.

I also can recommend the Cherry MX jailhouse mod if you are interested in trying out Cherry MX switches, due to the shorter throw. But alas, the MX switch is taller and heavier so there's no getting around the ML switch's advantages in this regard.

Did a bit of construction work when younger; always respected the pros. Every profession has its exponents and I am ready to swear that one pro is equal to ten untrained peasants. I wish my own country has real professional lifetime career construction workers with high competence and efficiency rather than importing peasants from abroad for cheap.

[Zekromtor]

I'll take one apart and experiment on it but they are small switches and are difficult for me to work with. Cherry's spec sheet says they have 3mm travel with activation at 1.5mm. It may be possible to limit the bottom travel by inserting something into the center barrel.

I was actually wondering if it would be possible to delay activation until it hit nearer the bottom, possibly by trimming a piece of contact metal, but I honestly can't tell how it works based on the few pics i can find of it taken apart. 3mm of travel actually sounds pretty nice. If you can fix the scratchiness and they function well, might have a winner here in spite of what people tend to say about them (just as your other thread suggested).

[jacobolus]

There are two big problems with ML:

(1) They wobble a great deal, and bind hard if you hit them a little off center. Even a standard full-size keycap is bigger than you want for ML: you can’t safely press its corners. There’s a reason all those >1u ML keycaps are hat shaped, otherwise they wouldn’t work at all. Keys with stabilizer wires (like the left shift / enter on the Cherry board pictured a few posts upthread) work pretty well though. I want to try some of these keycaps someday, which I suspect would partially solve the problem:

More

(2) They’re quite scratchy unless you lube them, and they’re a pain in the ass to take apart, lube, and put back together.

Otherwise, I think they’re great, considering their size. People should use them for mouse buttons IMO.

[kurplop]

Berserkfan– Thanks for all the ideas. I'll follow up on some of them. I know there is a lot of talent out there that would be willing to help. GHer's are the greatest. It has always been harder for me to ask for help than to give it. I guess I need to start exploiting everyone's generosity.

I consider myself more of a jack of all trades and also admire the truly proficient specialty tradesmen that I associate with. Here in So. California, the immigrant population has taken over many of the construction trades, which I do not begrudge. I'm far more bothered by the excessive regulations which make it next to impossible to be in full compliance and still conduct a profitable business without driving up costs to the consumer.

Zekromtor– Check your personal messages regarding ML switches.

jacobolus– I agree on all of your points, especially working on them. They are really small. None of those are deal breakers for me however. I was aware of the drawbacks going into the project.

I have found that most of the ways that benefit by lubing can be reached externally and while it isn't as good as a complete disassembly, it does help considerably. I've also observed that broken in ML's operate more smoothly than new ones.
 
One benefit of a compact keyboard is that the keys are right there. Not having to overreach may make accurate key presses easier to achieve.

To solve the >1u issue, I simply eliminated them. This may be impractical on a commercial keyboard that has to be usable for many hand sizes, but if the key sits exactly where you would naturally strike it, larger keycaps aren't really necessary. I was surprised recently when a former employee of mine tried out the Alumaplop. Although he is about 6 foot 6 and has fingers that are as long as Dodger Dogs, he found the keyboard to be very comfortable to use.

I appreciate everybody's interest and hope to have a progress report by the end of the weekend.

[Zekromtor]

If you hadn't mentioned it, I probably wouldn't have seen that message for a while. Responded.

And I do wonder if .7 spacing, even horizontally (I'm already sold on <.75 vertical spacing), might actually be ideal. From the looks of it, my finger tips are skinnier than yours, so if you have no issues with the spacing, I may have to try that on my next build.


Edit: I respect your desire to keep capslock functionality, but what I'd recommend is that you possibly combine your capslock and shift keys into one in the following manner: retain normal shift function, however, a press and release of shift without any other keys being hit while it was down would serve as a capslock toggle. I can't think of any downsides to this since caps is never used in key combos and the shift key is only ever used _with_ key combos. Of course, this is only if you can actually make use of the extra key it would free up.

[kurplop]

And I do wonder if .7 spacing, even horizontally (I'm already sold on <.75 vertical spacing), might actually be ideal. From the looks of it, my finger tips are skinnier than yours, so if you have no issues with the spacing, I may have to try that on my next build.

The nice thing about the ML switch is that it's small enough that the limiting factor will be anatomical not because the switch is too big. People seem to be able to accurately use tiny cell phone keypads so I think the main hesitation to go smaller is convention. The only reason I kept the ML's .7" vertical pitch is the convenience of using available keycaps, otherwise I would have reduced it to .65" or even smaller.  The sub cap size would have the added benefit of making off center presses even less likely. 

Edit: I respect your desire to keep capslock functionality, but what I'd recommend is that you possibly combine your capslock and shift keys into one in the following manner: retain normal shift function, however, a press and release of shift without any other keys being hit while it was down would serve as a capslock toggle. I can't think of any downsides to this since caps is never used in key combos and the shift key is only ever used _with_ key combos. Of course, this is only if you can actually make use of the extra key it would free up.

You make a good point. Do you think that I could maintain the caps lock indicator light if I made such a change? If so, the only downside of me removing the caps lock is that it may threaten my reputation as a recalcitrant old man.

[Zekromtor]

jacobolus,

those keycaps you linked do look like a good potential solution to the problem you described. I'm a fan of rounded keytops, too, since you really shouldnt be relying on corners for presses anyways.

Thanks to kurplop I'm going to be doing some hardcore ML testing soon. 3mm of travel seems ideal to me, especially if I can mod them to activate at the bottom, which I've come to realize is my preference. They just have a lot of things going for them that work for me from the smaller footprint to the keycaps that will be easier to machine than standard cherries.

kurplop, there's no physical reason why you couldn't still have your caps lock indicator light work with that config, other than if you are relying on firmware you don't have the code for. still, it may be more work than it's worth, but being a fellow recalcitrant old man, i have a tendency to push everything to the limit to try and eek out every potential bit of input efficiency -- for better or for worse.

[jacobolus]

And I do wonder if .7 spacing, even horizontally (I'm already sold on <.75 vertical spacing), might actually be ideal. From the looks of it, my finger tips are skinnier than yours, so if you have no issues with the spacing, I may have to try that on my next build.

The nice thing about the ML switch is that it's small enough that the limiting factor will be anatomical not because the switch is too big.

Personally I think the ideal keeps the horizontal spacing about as it is, but compresses the vertical spacing dramtically (down to maybe .6"), but combine that with quite a bit of height step between columns, and place different fingers’ keys at different heights. The key tops can be reduced to a fairly small size (.6" x .4" ?) and they should probably not sit in totally parallel columns, but rather include a bit of the natural splay you get when extending the fingers straight.

[kurplop]

Personally I think the ideal keeps the horizontal spacing about as it is, but compresses the vertical spacing dramtically (down to maybe .6")

With the right shaped cap I think .6" would not only be practical but would reduce the vertical distance in 5 rows by 3/4". The possibilities with those scratchy, sticky, hard to open, problematic keys are staggering.

On another front, I got sidetracked because of limited time to work on another detail of the Alumaplop's construction. I knew from the start that the arcade buttons would be an issue because, to put it bluntly, that don't fit in the shallow profile of the case. I had some ideas but tonight I stumbled on to a great solution.

The standard 24mm Sanwa arcade switch is a full 1" from the face of the keyboard case to the back of the switch; Actually, 1 1/4" including the terminals. I've disassembled the switch from the housing in this photo.


I cut a donor board's PCB around a ML switch. I left a little extra PCB on the sides for mounting into the Sanwa housing.


Inside the housing there's a rectangular hole that looks like it was made for the ML.


 By trimming a little notch in the bottom of the housing where the PCB would fit, the PCB section snaps right into the Sanwa housing! Depth is reduced to 5/8". The switch feels great and no hot glue. Total time of conversion– 3 minutes. I still need to check the range of the plunger to make sure a hard strike doesn't push the ML out of its home and if necessary, put some kind of limiter on the plunger.
 

The white switches are just for experimenting, I will use black ones in the Plop.

[kurplop]

While on the subject of arcade switches, I thought I'd share something I tried but decided against.

I turned a switch housing out of aluminum. When I fitted it into the Plop I realized something just didn't look right but I couldn't put my finger on what. It came to me this evening. It reminds me of the cigarette lighter in the Old Man's '62 Chrysler.
 

[Zekromtor]

Haha. What if you black anodized it though?

[tufty]

It reminds me of the cigarette lighter in the Old Man's '62 Chrysler.

...and you can't easily get 12V over USB.

[kurplop]

Haha. What if you black anodized it though?

It reminds me of the cigarette lighter in the Old Man's '62 Chrysler.

...and you can't easily get 12V over USB.

It's comments like these that push the state of the art of keyboarding to the next level. Thanks for the thoughts.

[jacobolus]

It reminds me of the cigarette lighter in the Old Man's '62 Chrysler.

...and you can't easily get 12V over USB.

You can with USB 3.1 type C connectors. 20 volts!

[kurplop]

But I don't smoke.

[tufty]

But I don't smoke.

Pump 12v down your USB line and let your controller do the smoking for you.

Those modded caps look dead good, by the way.

[Zekromtor]

Yeah, the custom black keycaps have a textured look that matches the anodized finish very well. I hope your plan is to have it all blacked out, no labels on any keys, just that big deep red trackball in the middle to set it off. That oughta get you a feature on kotaku.

[BlueNalgene]

This is some handsome workmanship.  You've got some great stuff there.  I've got nothing worthwhile to contribute to the thread other than that tiny ego boost of a faceless person on the internet thinking your crafts are pretty neato.  Keep it up.

[ECA]

This is the beginning of what is hopefully about a 2 month thread, detailing the completion of my latest project.
(Attachment Link) (Attachment Link) (Attachment Link)
I decided to test my patience and improving skill on a mill by taking a 10 lb. chunk of 6061 aluminum and making about 9.25 lb. of shavings leaving me with a sleek housing for a keyboard loosely based on the ErgoDox and once again integrating a Slimblade trackball. Repeating this fundamental combination attests to my overall satisfaction with having a large trackball set between a split keyboard.

My objective is to make a semi portable keyboard that shares the basic layout as my desktop setup. An earlier iteration aka 'The ErgoPlop', unfortunately had to be renamed the ErgoFlop. While there were many things that worked well on it, its execution seemed to fall between two stools; most notably, the 20+ degrees of tenting simply made it too tall to be portable and the smaller trackball just didn't feel right after being spoiled by a larger one.

Some of the design features include, using Cherry ML switches because of their smaller profile, of course the Slimblade, centered between the keyboard halves, repeating the use of modified arcade buttons for mouse functions and reducing the keyboard to 62 keys, actually 64 because the two arcade buttons above the trackball are keyboard commands.

The keyboard measures 7.5" high by 13.25" wide by 1.5" high to the highest keys (less trackball). The one piece aluminum chassis is covered beneath with a 3/8" walnut baseplate rabbeted into it. The ML switches are PCB mounted only, so I designed and home etched a flawed prototype PCB that can be partly seen through the thumb pockets. I made some mistakes in the design but drilled it out to confirm various dimensions and key location. I have since corrected the design (hopefully) and will be printing the new ones soon. I used the laser printer toner method to mask the copper I wanted to remain, then burned off the exposed copper with ferric chloride. I think the process worked pretty well but it is obviously not the quality of a professionally produced PCB.

More details to follow...

I would like one..

HOW MUCH??

[kurplop]

 

This is some handsome workmanship.  You've got some great stuff there.  I've got nothing worthwhile to contribute to the thread other than that tiny ego boost of a faceless person on the internet thinking your crafts are pretty neato.  Keep it up.

I'll take faceless ego boosts wherever I can get them.    Actually, the best part about hearing encouraging words like yours is knowing there is a connection. That others understand and share the same passion that I do. We see it everywhere, whether cheering for the same team, enjoying the same music, movie or art, sometimes even in a conversation with a stranger. I think it's one of the greatest joys in life. Thank you for commenting.

I would like one..

HOW MUCH??

 
I was thinking about trying to come up with a clever answer. A few came to mind.

1. I could tell you, but then I'd have to kill you.

2. More than you could possibly imagine. (Read with Alec Guinness's Obi-wan-kenobi accent)

3. If you have to ask, you can't afford it.


Now for a serious response.

If a manufacturer mass produced them, probably over $500.

If I was commissioned to make one it would be $3000+ (I wouldn't be interested however)

What I would pay for one? $1000–$1500

How much will I sell it on Craig's List for when I make another one? $50?


Value, like beauty is in the eye of the beholder it's  hard to know what something's really worth.

By the way, welcome to the geekhack forum. There are a lot of great people here.

[Zekromtor]

Good answer.

I've got a friend that says about my keyboard "come on, just make me one. how much will it cost?"

It's a labor of love... and I wouldn't love making it for anybody else. It's tedious, tedious work. Especially the soldering, which I refuse to do indoors, making it even worse because of the florida heat and the lack of a distraction like some bad TV. My keyboard is far simpler to manufacture, and I used a CNC, not a manual mill, and it's still easily 40 man hours including the machining of all the keycaps. I can't even imagine how many hours go into the Alumaplop, machined from a solid chunk of aluminum.

Speaking of that... do you have a way of doing more than 1 keycap at a time? Using a single key mold to make all the ones you need sounds pretty painful.

[kurplop]

I plan on only making the caps that aren't available from my donor boards.  After all, I have 19 ML boards to pull from.  I'll make a mold for the standard cap for the keys I don't have in the single unit size and another mold for the thumb keys. If I experiment with different spacing I will have to make custom ones, but I'll cross that bridge when I come to it.

[Zekromtor]

19 ML boards? Holy crap.

[jacobolus]

There were a bunch of replacement laptop boards with ML switches and dyesubs on ebay a couple years back for like $4 each + shipping. They’re still available for $20 each + shipping http://www.ebay.com/itm/260217472925 or $16 shipped http://www.ebay.com/itm/350632420547 or $50 OBO for a lot of 5 http://www.ebay.com/itm/281640926821 (that looks like lasered caps), etc.

[kurplop]

I got a similar deal. The seller wanted about $20 shipped. On the first round I offered $36 shipped for 3 boards which they accepted. A month later the seller still had about 15 left so I offered $7 per board for all 15 shipped which they accepted as well.

Lesson-  Buy in volume and make a low offer (they can always say no). I think I had a different seller and they just wanted to move the product.

I'm not sure about the legends. The ones I tried to sand off took some plastic with the legends.

[jacobolus]

I'm not sure about the legends.

The ones you have pictured above are dyesubs.