Although I've built a few wind generators in the past, I'm a newcomer to the axial flux alternator and have spent the last couple of months looking at the design problem. This board has supplied a wealth of information and a lot of opinions, which has helped considerably. Flux has helped me (and others, I'm sure) a lot to understand the electrical side. I have some of the components for a proposed 16' machine, which is to the stage now of needing the alternator and it's mechanical design remain to be done.
I notice that newcomers that want to build a machine are often advised to follow the examples put forth by others and not to think outside the box. I'm afraid I can't fit into that mold. If the traditional, commonly accepted ways don't make good sense to me I'm compelled to look further. Such is the case for the cast-resin stator assembly.
The basic requirements are:
- Provide a way to hold the stator coils, subjected to strong magneto-mechanical forces, in a rigid, fixed position.
- Allow effective heat dissipation from the coils, which can be considerable, as burned-up stators have testified.
- Minimal distortion under higher temperatures.
- Impervious to the elements.
Other issues are:
- Good electrical insulation.
- No metallic elements in the magnetic path.
- Reparability.
- Reasonable construction and weight.
The cast stator does pretty well for 1, 4, 5, and 6 but is less than satisfactory for the others, and in particular, number 2. Numerous bandaids have been proposed and tried and the success of some has yet to be seen. I think there's a better way.
The problem of coil heat dissipation is one of convection - movement of heat from the coils to the surrounding air. Embedding the coils completely is just the opposite of what needs to be done. As much coil surface as possible needs to be exposed to the air but this is somewhat in conflict with requirement 1.
What we need is a relatively thin (compared to coil thickness) support structure that has sufficient strength and rigidity with the coils inserted centrally and bonded into cutouts in this structure. One material that I believe fits all the requirements is G-10, FR4 fiberglass sheet that is available in a large range of thickness. FR4 is a fire resistant form of G-10. G-10 is amazingly strong and rigid and I suspect that in ¼" thickness, would be sufficient for this purpose even in large machines. It is formed by many layers of glass cloth impregnated with epoxy, that are compressed and heated, forming a dense composite. It's often used in the electrical industry. The down side is that it's fairly expensive, roughly $20/sq ft in ¼". However, when you compare the overall costs to a cast stator with it's resin, fillers, glass cloth, mold materials, etc, it doesn't do badly. Another factor is that it's difficult to machine being hard on cutting edges. A water jet job would be ideal.
Coil construction would be somewhat more demanding, requiring uniformity to fit well in the cutouts. I would make the coils before specifying the cutouts. The coils would also have to be impregnated with a high quality, high temperature insulating varnish and baked to provide good weathering and a rigid structure.
I believe this approach would yield a superior stator with advantages of better heat dissipation, reparability and construction. I intend to use it on my 16' machine.