Michael is dead right. I was looking at making some and the fact that the spoke follows the inner rim shape means CNC mill-- two operations--and with the curved spoke the program would have to be mirrored for the second side, and a nest made to hold the part in relation to the first side--then carefully cut the concave in the rim with a lathe (again with a nest).
I think that Jose is correct--lost wax would be the most practical.
____________________ Fix it again, Mr Gates--it still works!"
Thanks for the comments everyone. I did some research on the net about lost wax castings. I will go anyway for 3D printing. I have to change my design a bit but that method is easier and cheaper, at least for me, I don't got any of the tools and materials needed for lost wax castings.
I assume I might be too late from your last post.
I would suggest you locate a 3D printing service bureau that will print these on a 3D systems Viper Si2 in high resolution mode (.003" diameter beam and 0.002" layer thicknesses) should be able to yield 0.020" thick walls & thinner edges using Accura Clearvue or Somos watershed resins..or similar)
Another option may be to find a source for Envisiontec's Micro printer. I believe the resolution may be even higher. https://www.youtube.com/watch?v=3KO6XMLvgJQ
For a static model, either printer;s materials should suffice but for a powered model you will definitely need to go to investment casting. Jewelery foundries may be a good source location.
____________________ In theory there is no difference between practice & theory; in practice, there is.
If you haven't got your wheels yet can I suggest taking a look in the ship modelling area for capstan drums? These have a concave radiused outer edge that could be mated to dowels with a "close enough" matching diameter as rails. They could be available in a range of diameters that could be sutable for what you want.