Hi willib.
First off I must say great work. Your putting forth an awsome effert here. Bravo, I like it.
You are one of the few to question the staus quo (spl(. Thinking outside the box is only going to make things better.
A few questions and you may have allredy answered these so forgive me if you have. What is the plastic stator material?
Is your wire 17 gage? As near as I can tell by my wire conversion chart the closest gage to .045" is 17 gage.
I'm impressed with your small diameter coil center hole and still having good results.
Reason being I'm building several dual rotor test alts and the present unit I'm building has coils slightly smaller then magnet size. I thought at first this might be a problem- although DanB has sugjested it works fine.
The first alt I'm building is the 8 magnet per disc with 6 coils 3 phase. I will build other units latter with the 12-9 configuration.
With the standard 8 magnet per disc and 6 coils wired star the configuration is wired with the 2 coils directly accross from each other in sires. This is phase one. Then do the same with phase 2 and then phase 3. Then the accepted thing to do is wire them star.
I'm doing just that as mentioned above. In this curent small dual rotor I've wound the 6 coils with 19 gage wire.
In star the coils from one phase are basicly in sires with the coils of either of the other 2 depnding on rotor postion.
This then is equal to 4 coils conected in sires.
Do to this "4" # in sires I've also wound 6 coils of 25 gage wire. 25 gage wire is 1/4th the circular mills of 19 gage. 25 gage = 320.4 circualr mills. 19 gage = 1288 circular mills.
In order for 1 coil of the smaller wire to equal 4 coils of the larger wire in voltage it must have the turns count of all 4 of the larger coils.
Therefor I've wound 150 turns per coil of the small wire and 37.5 turns of the big wire. I actuly wound the large wire with 40 turns since its been sugjested here that its unfair to compair star wireing against "Jerry Rigged" since 4 of the 5 volt star coils conected in sires won't equal 20 volts it will only be 17 volt (out of phase coils not working well with each other). Therefor star needs a bit of a handycap to catch up with "Jerry Rigged".
Itresting thing about these 2 coils types. They ocupy the same space and they weigh exactly the same. 1 once each on the postal meter.
Keep this in mind. If your going to do "Jerry Rigged" you must do some math and make each the coils an equal to the total turns count for 2 phases of normal star.
For example. If you are doing coils for the 12- 9 config. Each "Jerry Rigged" coil would have the turns count of 6 of the star coils and be 1/6th the gage. Simple connecting the same coils either "Jerry Rigged" or star won't produce the desired results.
You could also do "Jerry Rigged" by making each phase have the turns count of 2 phase with wire half the gage of the star phase. Then wire the 2 (for 8-6) or 3 (for 9-12) in sires then off to there own rectifier. This reduce requiered bridge count to 3.
Many posabilities. Keep us posted. I'm following this with great intrest as I'm involved with simular testing.
JK TAS Jerry