Good stuff Chris, I think I follow what you are saying.
I guess I should ask it this way....
If a set of blades is known to be able to produce x amount of power, how do you determine the cubic inches of mags needed and the corresponding amount of copper?
The only good way I know of to do that is by experimentation with a totally new design. Theoretically you can use a gauss meter to measure the flux of a particular set of magnets, then do a bunch of calculations to figure out how much wire you need. But I've never been able to get that method to work out. The old test coil will tell you more than anything.
The speed of the magnet does matter. If you have one coil and one magnet and move the magnet over the coil really slow nothing will happen. But if you move the magnet over it fast it'll cause current to flow. If you move it over the coil faster yet i'll make more current flow. I don't know the relationship there either - I just know thru experience that bigger rotors with the same mags take less turns of wire than smaller ones.
In theory, a 10 foot machine that's 30% overall efficient should make about 1,800 watts @ 25 mph wind speed. A 13 foot machine at the same 30% overall efficiency should make 3,150 watts at the same wind speed. My 10 foot will make about 1,400 watts in a good 30 mph wind. My 13 foot only makes about 100 watts more than that - about 1,500 watts. But the 13 foot has proven it puts out more power than both 10's combined at 12-14 mph wind.
I suppose there's some way to apply some sort of formula to all this, but I don't know how accurate it would be. I made a spreadsheet awhile back that used complicated formulas to predict power output of a design, but it's only accurate within a certain range of like +/- 20%, which is quite a bit of error. The reason it's hard to predict, unless you're working with a known design, is because the volts you get per rpm do not stay constant. When I built my 13 I tested it at three different speeds and got like 6 rpm/volt at 60 rpm, 5.2 rpm/volt @ 150 and 4.8 rpm/volt @ 450.
Again, I suppose there's a way to apply scientific formulas to this. But how accurate it would be is questionable. A bench test will always tell you more than 16 pages of mathematical formulas. So to sum up, when you're working from a known, proven design you have something to go by. When you're designing something new you have to be able to accept failure if it doesn't work, say, "Well, yeah, that didn't work worth a darn", redesign it, learning from your mistakes, and hopefully the next one will work better.
I have a couple or three stators laying around here that are labeled "Top Secret - Experimental" because they didn't work. If you're not willing to start a Top Secret Experimental stator collection like I have, then you need to work with a known or proven design. And that includes furling systems too - I got another collection of stators that don't have any smoke left in them. Those have accumulated in my High-Output Stator Display.
--
Chris