One more three phase question...

[ezee]

Seems like most of the dual rotor axial three phase designs I see online

http://otherpower.com/turbineplans.html

http://www.windstuffnow.com/turbine%20kit.pdf

are using nine coils and 12 magnets.

Is it safe to assume one can design a three phase dual rotor axial using any iteration of 3:4?

e.g. 3 coils 4 mags, 6 coils 8 mags, etc?

Thanks!

[Tritium]

Yes.

Thurmond

[windstuffnow]

  The single layer windings using that formula ( 4 pole 3 coil ) is the easiest and quickest to fabricate. ( Thus the popularity ).  You can use 2 pole 3 coil and 1 pole 3 coil layouts as well but it gets more complicated and becomes more labor intensive.  

.

[JW]

Hi All,

 I thought the displacement of the 3 too 4 ratio allowed for 120* out on sync with the waveform, like prime numbers work. Think about it, the rotional geometry does not match up with 4 magnets to 4 coils. Why is it logical that 6 coils to 8 magnets generate the same type of waveform? How does this devide into 120* evenly.

 Am I making any sense.

I relize the rectifier(bridge) does not care about the integrety of the waveform, and for the most part cross-canceling of waveforms is elminated with an un/balenced multi-phase rectifier, wired with star connection.

 But, if you wernt running series delta it wouldnt matter? Is there some shunt, im over looking here, or am I completly wrong?

 Doesnt magnet spacing catch up to you, at some point, or are the coils overlapped?

JW

[ghurd]

The number of magnets or coils doesn't matter.

The ratio needs to be the same.

The 120 is between waves and is still 120 if the 3 phases are balanced, regardless if it is 6/8, 9/12, 18/24...

It could be enlightening to sketch out a 9/12 with magnets and coils, then erase 2 phases of coils.  The interaction may be easier to understand.

Magnet spacing is important, but I don't see it as relevant to the concept you are missing.

Overlapping coils, etc, is another level.  Best avoid that until later.

G-

[JW]

ghurd,

 Thanks for the reply.

"It could be enlightening to sketch out a 9/12 with magnets and coils, then erase 2 phases of coils.  The interaction may be easier to understand."

Yes,yes

 I see what your saying, If a star wound stator has 6 branches, with the neutral unused, the rectifier would be balenced. I think this is what I was eluding to with the series delta type of stator. If the stator is not wired in delta, phase canceling is [or could be] irrelivent.

Question-

 If one used star wiring in the stator, would that make an overlapping coil design more forgiving?

JW

[ghurd]

Yes.  Star or Jerry Rigged will be more forgiving than Delta.  Regardless of overlapping coils or not.

Ed's numbers for overlapping coils show some advantage to overlapping the coils, but I think it's better suited to a single rotor with laminations.  The space for wire gets small- Fast!

It feels like there is again some confusion about "Jerry Rigged".  It is each phase rectified separately, then the positive outputs are combined, and the negative outputs are combined. The 3 rectified phases are paralleled.

G-

[JW]

Hmmmm,

 Well then, the only [possible]advantage to wiring in delta would be to minimize theoretical rectifier losses. But if you wired in delta the stator would be inter-connected and magnet-spacing would matter in relation to coil ratio. But if you wired in star or wye with the neutral un-used, youd have twice as many rectifier junctions, for example 3/with delta and 6/with series star. But each equalling the same output in watts per circuit division to main DC output buss.

  With low volt DC output applications, Im not even sure if theres even that much of an margin, both need properly designed heat sinks. Im not even considering laminations working with windings. Since most dual rotor applications are air-core.

I have not seen{noticed) a direct comparison between 'jerry rigged' and star or wye with unused neutral, but I welcome the idea...

 I thought jerry rigged used capacitors(and diodes)... to boost voltage to a usable range, with series delta.

JW

[JW]

oohps,

 Thats-

'and 9/with series star.'

not

"and 6/with series star."

For the 24 mag dual rotor with 9 coils.

JW

[ghurd]

Jerry did extensive comparisons on dual rotors a couple years ago.

It is similar to Delta, but has no parasitic current flow.  There are no caps.

I personally use the configuration with motor conversions to get a lower resistance and raise the cut-in speed.  High resistance and low cut-in are problems with the little things I make.

Delta tends to have lower resistance and more power output at higher RPM, while Star tends to be more efficient at the lower RPMs most windmills see.

Ed's "Alt from Scratch" may show some interesting things.

Like how there is no room at the inside and outside for overlapping coils, and a comparison chart for the PMA in both star and delta.

http://www.windstuffnow.com/main/alt_from_scratch.htm

[JW]

ghurd,

"e.g. 3 coils 4 mags, 6 coils 8 mags, etc?"

[from the top of ezee's posting here]

The 6 coils with 8 mags is where im getting lost.

 3 to 4 relates to 9 to 12 in a way that is not divisable by 6 to 8.[pie r squared 3.14] Since 6 and 8 are both even numbers. Wheras 9 is prime and 12 is even, as 3 is prime to 4 as even. Generally speaking using this offset(ratio) 3 coil to 4 magnet spacing is inherently correct? But with 6 to 8- the coil number(6) is not prime. I do relize that 2 is considered a prime number, but does this give the whole lot, a reason to say that coils with even spacing(not odd numbers) will work with even spaced magnet # scenerio's...

Yes, a single phase stator with 2 coils (wired in series but a single phase), using 4 magnets(single rotor) on the rotor, would work fine with 2 AC inputs to a single diode bridge. But you only need 2 diodes to rectify a single phase by itself, when 2 or more phases are connected to the same DC output buss, since the neutral is not tied down. one can rectify 3 hot phases with only 6 diodes to a main DC buss.

 With 6 phase star, there is a descrete phase loss when used in relation to an single 8 magnet rotor, let alone 16 magnet dual rotor, even with a good magnetic circuit.(for air core)

THAN WHEN compared to a 9 phase star using 12mag single rotor or 24mag dual rotor. The loss from the rectifiers, is not as extreme as waveform losses, with 6 coil to 8 mags for single rotor or 16mags with dual rotor). But both arrangements will produce power, without getting hot(the coils) under nominal loads.

right?

With delta, the coils should get hotter, 9 of them in parallel, equaling 3 phases needing 6 big diodes and heatsink.

by comparison, thered be 9 phases with the common neutral that is not used, with star, needing 18 diodes on the buss, instead of the only 6 diodes needed for delta.

There is no question, star with 9 phases is equvelent to delta with 3 phases, only with the 9/12 coil magnet ratio per single rotor, but is the output with a 6 coil to 8 magnet 2/3rds as much as 9/12.

CHEERS

JW

[finnsawyer]

With three phases the number of coils must be divisible by three, and 6 is certainly divisible by three.  The 8 comes about because of the 3:4 ratio between coils and magnets that is required for proper operation.  Whether the numbers are prime or not is irrelevant.

A normal single phase alternator would have the same number of coils and magnets, a 1:1 ratio.  In my diary I have proposed a single phase alternator design that would have a 3:2 ratio of coils to magnets.  Such an alternator could have 10 magnets and 15 coils, for instance.