CAUTION: This is an experimental design. Keep in mind that no magnetic analysis has been done and the design is not tested so the performance is unknown. The only thing I can guarantee is that it will turn. That's it.
Non-printable parts needed
- 2 8x4x3 mm ball bearings. Min 30.000 RPM rating and preferrably with metal caps on both sides to keep dust off. Metal cap designation is ZZ.
- 12 neodymium block magnets 8x4x2 or 7x4x2 mm. Any heat resistance class is OK since the printed parts are the bottlenecks. Go with as powerful as you can find so anything in between N48-N52 is a good choice. If you want high temperature resistance look for designation M or H at the end like N52H etc...
- A half-thread M4 bolt as a shaft. You can also use precision DIN pins and thread them yourself. You can probably be creative and make it work with any standard M4 part.
- 0.6 mm insulated winding wire like enameled copper etc...
- 2 M2 set screws.
There are 2 different manufacturing methods that can be used
- 3D print the whole thing.
- 3D print parts as a master for epoxy/resin casting.
For a quick but highly inefficient result you can 3D print everything. If you do so, some things to consider are:
- Stator core should be printed with a magnetic compound that can withstand high temperatures. I would guess thinner the layers better the results would be.
- Bell should be sliced in a way to maximize radial strength as the centrifugal forces at high RPM will try to take it apart.
My 3D-printing-inexperianced recommendation would be to use:
- Ferromagnetic-PLA for the stator core. Looks like this is the only magnetic filament that exist. It's not resistant to high temperatures.
- Nylon or Polycarbonate for the housing, bell and the spacer.
If you want a stronger and possibly more efficient motor you can use the 3D printed parts as a master to cast the production part in epoxy/resin. Things to consider are:
- There is actually a core production method called SMC (Soft-magnetic composite) so the stator core can be made from very fine iron powder and high temperature epoxy/resin. I've never tried it but theoretically it's possible and will probably yield better efficiency than magnetic filament.
- Bell should be cast from a strong material. Analysis (link below) show that at least ABS level tensile strength is needed. There are epoxies that can provide it.
- Wind the stator in dLRK combination with 7 turns each tooth (See this for more details)
- Glue the spacer to inside of the housing and put the bearings on each side.
- Glue the stator to the housing.
- Glue the magnets to the bell.
- Put the shaft and the bell to the housing.
- Use M2 set screws to constrain the bell to the shaft.
For the CAD designs click here
For the bell analysis click here