I could be way off base here, but my first thought is you’ll have very little room for error with the tolerances for your proposed Alps style peg/hole system. If the hole’s too big, it’s going to rattle around on the peg, if the hole’s too small, the peg’s not going to fit. The last thing an artesian cap maker needs is another potential element that can go wrong and render that cap defective.
I wonder if something like the two-piece IBM buckling spring keycap design might allow for looser tolerances and be a little less error prone?
Show Image
Reference: http://deskthority.net/wiki/Keycap_mount
That's a very good question. To answer your question, there's a few reasons I didn't.
First, Clack has done that and I wanted to make something that was different.
Second, you run into the potential that you won't have walls thick enough to print or to comfortably cast without issue or you may have walls too thick to fit.
Third, long term durability may not be there depending on the properties of the resin and thickness of the cast. Playing with the stem may be less of a problem than popping the cap off. Or it might not be a problem at all.
Fourth, casting difficulty. The bases, especially the buckling spring, become much more difficult to cast.
Fifth, cost effectiveness. If you cannot easily cast the bases, you'll have to print them. Making them easily and cheaper to print helps control costs.
Sixth, it makes it very easy to make a single piece mold if you don't want swappable stems.
Seventh, and last, I'm not sure my CAD skills are good enough to get that close of a tolerance.
I know I'll probably have to figure out a better modular stem system in the long run, but it's a stepping stone.