What determines the geometry of coil windings for a slow rpm (home made for VAWT) 3 phase alternator? i.e.the actual shape of the wire form. I have seen many different examples, but is there a rule?.
When I saw your question originally, I thought you were referring to the section through the wire. It's clearer now, it seems to me that the form would be determined by the physical location of the winding. So the shape of the coil in plan could be designed (rounded) trapezoidal to sit together on a circle at a radius. The flux density would be harder to calculate than for a circular coil, as you have an effective diameter, but empirically people have probably worked out what is best.