OK... I'm an idiot. Well I know you generally want 3 coils for every 4 magnets and run 3 phases. Well now I know why. I was going to run 4 phases on my 12 coils (16 magnets per plate) as I was running thinner wire than I lreally wanted and was worried about the voltage drop for the longer distance caused by the 4th coil in each phase of the 3 phase set-up. Well it did not work so well and I started thinking about it. With a 3 phase set-up I would have had all 4 coils in a phase sitting over the same pole magnets at the same time. In a 4 phase setup I have 3 coils, but one would be on a north pole, one on a south pole and one inbetween poles. They would be fighting each other... and that is what it did. The voltage was very low and wandered around a bit. I thought a bridge rectifier had shorted out, but even checking the AC output on a single phase did not show too well.
I also found the aluminum struts are a bit too springy, so I will switch to steel instead. It will make it a little heavier, but will be much stronger.
I also put the blades on the wrong way around and it did not spin much at all unless the wind gusted. I switched them around and flipped them over and now it actually starts spinning up slowly in as little as about 2MPH of wind... admittedly it spins VERY slowly. I also found that my house fan idea was not the best as the actualy wind speed after just a few feet is actually pretty low. It will spin it, but not much.
I can fix all these issues, but it will take more time. I have it ripped apart again and have the stator ring out. I made the connections on the inside of the plate and will try to chip some of the epoxy away so I can rewire the stator to 3 phases. I will also added 4 more blades for testing, but will stick with the aluminum struts for now.
One thing though... those bearing work really good and this thing spins with almost no pressure. There is mass, but almost no resistance. That part makes me happy. My temp tower for testing worked well too and actually made it much easier to work on the generater with it mounted. It took about 5 minutes to take off the 4 blades, seperate the magnets plates (not a fun task), and remove the stator ring. Hopefully Saturnday I will have it all back together with the stator ring fixed and the extra 4 blades added. Then I can do some real testing.