Sanity check on movable stator coils

[Masked]

Hi All,

I've been learning a lot about wind-powered alternators and 3 phase power, and two basic instructions got me thinking about a solution that I haven't seen discussed anywhere.

First, there's the common mistake of placing a pair of coils in opposite rotation (one CW and one CCW) that causes them to cancel out each other's voltage.

Second, when setting up a 3-phase system, it is noted that the common ground will ideally always have zero voltage, or potential difference, because the sum of 3 sine waves 120° out of phase is always zero. (when they have equal freq. and amp.)

It occured to me that if a pair of stator coils had one coil rotate 90° out of phase then, regardless of rotor speed, a linear decrease in output voltage would be seen while maintaining the same frequency.

It seems to me that for the purposes of voltage regulation, dump loads, furling, etc. it would be better to use this method to simply reduce the load rather than divert the load to heating elements or mechanical braking methods.

I've seen discussion about dynamically increasing the air gap, but that provides an exponential decrease rather than linear - which is hard to make use of.  Am I missing a basic principle here, or should this method work?

...and does anyone know if it's been tried?

I've designed one for a three phase alternator, and the simplified picture below shows one way to do it for a set of 3 coils wired in series. (all in phase 1 of 3). The rotor here has 8 alternating magnets producing a full sine wave every 90°.  The left position would render full output voltage, and the position on the right shows the fully moved coils in blue, where each of the three coils contributes a sine wave 120° out of phase with the others to produce a net of 0 from this set.

Thanks for your input,

-Masked

[ghurd]

If the output is reduced, the blades overspeed.

G-

[Masked]

Thanks for the reply, and good point.  

In my case, the current (test) turbine is only 3 feet wide so I hadn't worried about that yet.

I probably should have been more specifc thatI wasn't worried about tip speed or structural integrity - I was more trying to verify that the wiring works as a linear voltage regulator amd have things not melt in 80mph gusts of wind.

-Masked

[ghurd]

I guess the first thing I wondered was about where the other 2 phases coils move to.

The only way I can see 3 phases fitting and rotating around is with overlapping coils, and I have a feeling the resistance will be less with heavier wire and non-overlapping coils.

The 3-ph 8 slice pie has me wondering about what I am missing, but for what I think what are asking the answer is "sort of"?  It would reduce the voltage.  I don't think it will be linear with the angle of rotation.

G-

[Masked]

The placement is a little tough, especially with a "set" of 3 coils in series.

To build a 3 phase, you need 3 wedges for your stator, two of which move.

And to get get nice round integers for measuring it out you need weird configurations like 9 coils/20 magnets.   Configs with 9c/24m, 18c/30m, 27/40 or 36/60 work too... but using 3 coils in series and moving two of them out of phase isn't necessary - just more efficient for spacing.

Much easier to make your stator as two big pie slices.  Picture a normal axial setup with 24 magnets & 24 coils. The coils are in sets of 2 wired in series.

Here each magnet occupies 15° and a full sine wave occurs in 30° of rotation.

That means you can rotate half the coils (one wedge) 15° to bring all pairs of coils 90° out of phase and drop your voltage linearly to 0.

So if you imagine popping out 3 adjacent pairs of coils, you free up 90° of space around the circumference of your stator and you have room for one half, or wedge, to rotate the required 15°.   Less efficient when you have removed so much magnet wire, but easier to visualize.

I did the graphs on a calculator, and may have gotten it wrong, but you can check at:


http://my.hrw.com/math06_07/nsmedia/tools/Graph_Calculator/graphCalc.html

Equations are:

sin(x) + sin(x+(ð/3)*2*0.4 ) + sin(x-(ð/3)*2*0.4 )

...as you vary the 0.4 from 0 up to 1 they gradually slide from being in phase to being 120° out.

Thanks again, you were spot on about the blade speed  That's exactly why I hadn't seen this; normal windmills are big enough that it matters quickly.

-Masked

[Masked]

Those weird symbols are supposed to be PI.

My calc was in Radians mode - not degrees.