I think the bottom line, is how much copper can we get in a given space - having most of the windings in about the right place.  In that respect, it'd be the same - it would probably run more smoothly, but any 3 phase configuration is going to be pretty smooth with regard to vibration.  When the magnets are "working" the hardest - the torque on them (or on the coils) is the most, and when they pass over the holes in the coils I think it's the least.  In a single phase machine this creates lots of vibration... in a 3 phase machine its much smoother.  The advantage of whats being proposed here... is it might be a bit smoother yet... the magnets would be passing over thinner coils (because you have them stacked) - so they'd not be working as hard.. but there'd be no holes - so there would be less vibration and the magnets would always be working.  In the end though, I'm not sure there'd be much advantage unless you think you can fit more copper in this way.

Ah.

Actually, what you were referring to a "regular three phase" was just the non-overlapping coil axial-flux alternator design described often on this board.  (Who did come up with that one, anyhow?)

The one youre referring to as "6 phase" is the overlapping coil three-phase wind that we occasionally see discussed for axial-flux designs.  But it's actually the common wind for radil flux designs (such as the induction motor to magneto conversions).

Yes, that should "just work".  You can put the winds in parallel for more current at a given RPM, or in series for more voltage.  Or you can wind them with half as many turns of thicker wire and put them in series for the same voltage and more current, and so on.

As I understand it, the single layer of coils is simply a construction advantage, getting better cooling and easier construction at the cost of using twice as many magnets to induce power in  given number of coils.  If you do the overlapping coil version you need to bend the wire to keep the magnet gap small.  With an air gap rather than an iron core through the coil even a slight increase in the gap would make for a major cut in flux - losing your gains in magnet-use efficiency.  Bending the coils (and having a thicker section of the stator just outside the pole-swept area) makes the stator (and perhaps the whole assembly) harder to construct.  Also, the bend required to miss the next coil over means a slightly longer wire path and thus more resistive losses.