axial flux .... coil questions etc....

[basicbill]

Hi everyone,

I am finally getting around to building the 24 volt version from Hugh's plans. On page 22 (May 2004) the coil connections are shown with a common connection on the start and the finish of each coil goes to a bridge rectifier. Do adjacent coils get connected to the same rectifier or are 'opposite' coils connected to the same rectifier?

Also, has anyone used purchased blades to go with an axial flux alternator and if so, which ones. How effective are they etc?

Thanks for your consideration,

Bill W.

 

[Flux]

Each  coil connection goes to an ac connection on the rectifier, it doesn't matter which coil end goes to which rectifier.

 Effectively the rectifier consists of 10 diode pairs commoned on the positive and negative side, with the junctions of the diodes being the ac connection. If you used individual diodes the whole thing would be symmetrical. The fact that pairs of these are in individual single phase rectifier blocks for convenience makes no difference.

Flux

[Ungrounded Lightning Rod]

Sounds like a delta, Flux.  I think he's asking how the more-than-three coils in the device are interconnected, in addition to how they are connected to the recitfiers.

I don't have the plans handy or I'd check.

[Ungrounded Lightning Rod]

How many coils in the stator and how many magnets per rotor?

[Flux]

No , I think it is the 12 magnet 10 coil parallel 5 phase, where the 10 coil ends go to the rectifier. That is why I think there may have been a worry about keeping each phase group on the same rectifier, but it doesn't matter.

Flux

[basicbill]

Hi,

This is Hugh's 10 coil / 12 magnet design. I wound the coils for 24 volts.

Thanks for the time to consider this.

Bill W.

[Trivo]

Hi Bill

Here is the process

get a multimeter and measure a pairs of leads spinning at a constant speed by hand, some will have a high voltage and some low just keep paring them till you have 5 sets of high voltages

if you get 4 sets and one low split up one of the high voltage ones match them with the low voltage ones eventualy they will match, connect the pairs to one of the rectifiers

Trivo

[Ungrounded Lightning Rod]

OK.

You have a five-phase machine, with two coils for each phase.

Calling your five phases "V, W, X, Y, and Z" your coils (picking an arbitrary start point and working around the stator) are V, X, Z, W, Y, V, X, Z, W, Y.  Thus the two coils opposite each other are of the same phase.

The senses of all the coils are the same, i.e. the start always represents the same polarty of its phase, the end always the other.  This means several things:

 1) You can hook the two coils of the same phase in series by hooking start on one to end on the other and using the unhooked start and end as the start end end of the double-voltage composite winding.

 2) You can hook the two coils of the same pase in parallel by hooking the starts together and the ends together, and using the tied starts as the start of the double-current composite winding, the tied ends as the end ditto.

 3) You have no problem identifying the start and end of the various phase windingswhen you're combining multiple phases.

Once you've tied the two coils of each phase together there are several ways to connect them.  (Output voltage of 1 is what you'd get rectifying one coil alone.)

 - Separate rectification (10 AC terminals on your bridge rectifier):  Tie each start and each end to a bridge rectifier AC input.  Output voltage is 1 unit if each phase's coils are paralleled, 2 units if seriesed.

Two variations on Y:

 - 5-phase star (a five-pointed asterisk, the N-phase analog of the "Y"):  Tie all the starts.  The ends are your five phases.  Output voltage:  2 * cos(36 degrees) = 1.62 units if paralleled, 3.24 units if seriesed.

 - 10-phase star:  Same as the seriesed version of separate rectification above, but with the jumpers between the two coils for each phase tied together.  Output 2 units.  (No particular advantage vs. separate rectification except that you get a good center-tap.)

Two variations on Delta:

 - Pentagon:  Tie: Vend-Wstart, Wend-Xstart, Xend-Ystart, Yend-Zstart, Zend-Astart.  Outputs are these 5 ties.  Output voltage:  2 * cos(36 degrees) = 1.62 units if paralleled, 3.24 units if seriesed.

 - Pentagram (christmas star):  Tie:  Vend-Xstart, Xend-Zstart, Zend-Wstart, Wend-Ystart, Yend-Vstart.  Outputs are these three ties.  Output votlage:  1 unit if paralleled, 2 units if seriesed.

The pentagram gives you the same voltage as separate rectification, but only needs five AC terminals (10 diodes) rather than 10 AC terminals (20 diodes) in your bridge, and five rather than ten wires from the genny to the bridge.  It may also give you somewhat more current (but if so will dissipate disproportionatly more heat in the stator).

The pentagon gives you the same voltage as the 5-phase star but may give you significantly substantially more current.

Both the pentagram and the pentagon may also produce some circulating currents and resulting losses and drag due to harmonics in the generation waveform.