Thanks for your interest, Chris and David!
About the lever material: Plastic would be a natural choice; however, each one would need to be designed individually to fit, which is certainly a risky thing to do as long as one doesn't know whether the whole thing works in the first place... the material I've got scheduled to test are 0.5m brass rods with a square profile which I found in a supply store for model airplanes. I intend to cut those to size and see where I get from there. I'm absolutley sure that some of the things I thought up aren't going to work at all, so in a "real life" project, there'll be several evolutionary iterations.
David, you are absolutely right about the outer [ ]. The first material I laid eyes on for the levers were metal rods with a round profile, and I was sort of worried that these might bend over time and therefore lose their exact positioning. SInce a design cycle is fairly long and expensive, I followed the philosophy that I'd rather design unnecessary things in to later remove than find I didn't have something I later on find out to need. Oh well.
The outer single [ is a crutch. You'll notice that it is the only one that doesn't match with a ], the only reason being that I didn't have space because that room was taken by the next Button. The positions of the buttons were given, and for some reason I didn't think of the possibility to move the corresponding reed towards the outside (for which there would be ample space). So the button post itself must serve as the other ]. It's one of the known weak spots of the design.
What you don't see in the picture is that all of the [ ] shaped blocks have three holes in their side walls. The idea was that these blocks serve as pivot points for the levers. Before I had the pan printed, I wasn't sure whether one could drill holes in the pan to fix the lever fulcrums at arbitrary positions, so I "hardcoded" the pivot points. Looking back, it's probably not a good idea to align the pivot points of the long levers flush, but that's another thing to take the test of reality.
The last thing that's worth mentioning is that I plan to put the springs in the bottom of the button holes. Dana Johnson (who was extremly generous, valuable and patient in private message brainstorms when I discussed the idea with him - he's responsible for everything in the design that might work, and I'm in charge of the rest - thanks Dana!) suggested that I look for conical springs for that purpose. Even though the area I live in appears to be a natural habitat for spring makers (there are about a dozen or so companies within driving distance that specialize in spring making), I haven't been able to find somebody to make prototypes according to my specs at reasonable prices; maybe I'll need to find a creative way to coil those myself.
But as I wrote before - It's one of those projects that have initial momentum but then always end up in the bottom of the stack of things to do. This discussion actually got me to thinking about moving the project up the stack again.
In the meantime, I'd be more than happy to share the design files and the STMs in case anybody is interested. I used FreeCAD which is not the best CAD Software for more complex shapes but works fair enough for simple projects.
Edited by RAc, 11 September 2017 - 02:49 PM.