Rotor design question

[Aelric]

I was reading a post recently

http://www.fieldlines.com/story/2005/9/6/52841/15548

with references to the design by Windstuffnow:

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

and I was wondering, what would happen if for the steel laminatation you used a coil of steel, insulated it, then wound the coils around it, then resin the whole thing and sandwich it between its magnetic rotors.  I know this is probably a silly question, and have heard some mention it is better to have the lamination behind the coils but I still am not understanding why this is better.  I know it has to do with the flux path.  Anyway hopefully someone can clear this up for me :-)

Thanks,

Aelric

[kitno455]

as i understand it-

we are trying to make a magnetic circuit. in a dual magnet rotor PMA, the flux 'leaves' the N end of a mag, jumps across the air gap to the opposite S mag, then splits in two in the steel backing plate, makes a right angle turn, and heads over to the next mag, where it joins up with flux coming the opposite direction, makes another right angle turn, passes thru that mag, and so on across the air gap.

placing iron lams in the middle of such a two rotor on the back of the coils would only sheild half the mags from being able to 'see' the coils.

now, what i think you are saying is why not wrap the coils around the lams, so that both legs are in the same flux path. that wont work if the mags are N/S across the airgap, cause then the two legs would cancel out. thats the same thing as having a coil narrower than one of your mags.

now, if you put the mags N/N and S/S across the air gap, this would actually work, except that the flux would not want to jump across the air gap toward a 'like' mag. the steel in the stator would help attract the flux, but not as much as an opposite mag.

so, all in all, sounds like one of my ideas, nice try, but no cigar.

allan

[Aelric]

yes that is what I was wondering and thank you for the explanation,  I think I understand it better now.  Will keep thinking maybe come up with a few more hairbrained ideas hehehehe.  Thanks for the reply.

[Aelric]

k one more idea, what about dual stators, just put another stator opposite the first, so that both sides of steel laminate have a stator on them

ex.

magnet rotor | coils | laminate | coils | magnet rotor | blades

just a thought

[Flux]

Yes both ideas will work.

If you add a second winding on the back with another magnet rotor you just have 2 single rotor generators on one shaft. Best to keep the magnets as in a dual rotor, north opposite south unless your core is thick enough to contain the flux from each side in half thickness.

If you wind the coils round the core as a torus you need to set your rotors so that north faces north.

Proven use the torus design, it saves a lot of resistance in end windings but is not so easy to wind by hand.

If you do a Google search on torus alternator permanent magnet or something similar you should find information on both schemes. If I can turn up the reference I will post it.

Flux

[finnsawyer]

Do you wind each coil around its own coil of steel?  If so, I suspect there will not be too much effect, as I seem to remember seeing laminations used in a similar fashion in auto alternators (the field wires pass through a frame made up of laminations).  It may, however reduce cogging due to "air gaps" in the steel coil.  As the coils pass over the rotor there is a tendency for the flux to hang on, be bent, and be dragged along.  The gaps in the laminations should reduce the bending and keep the flux oriented along the axis of the coils.  If you've got the time, I'd say try it.  Just be sure to report back your results here.

[Flux]

I hope this reference works. Be warned if on dial up, it's a big PDF

http://www.em.tut.fi/tuuliryhma/weg_articles/icem2000.pdf

Flux

[Aelric]

actually my idea was to take a bit of say 1 or 2 inch wide coil steel, and then wrap the coils of copper around that (possibly cutting the coil of steel and prewinding the coils of copper then just sliding them onto the steel coil so it would be in effect coils wrapped around a large coil) imagine a ring of steel with small copper coils wrapped around it.  My biggest worry was that the magnetic flux would not be directed the proper way, because as I understand it the steel plate in the stator design most common on this site, the steel plate directs the flux in a straighter path (I could be very wrong but thats how I think it works) towards the opposite pole on the magnet behind the coils, inducing voltage and current on the copper wire.  

One other idea I had at work today along the same lines, what if you used this torus design and rather than having magnets in front and behind the coils, what about on top and on bottom?  Guess the best positioning would be simply to have the most magnet covering the most surface area of coil.

About the other design with the coils on both sides of the steel plate between the magnets, would this seperate the magnets too much to be effective?  I know the airgap is extremely important to keep to a minimum, wouldn't increasing it by adding more coils reduce the overall effect of the magnetic flux on the copper?  Or would the extra copper wire end up compensating for the loss?  In short would it be more or less efficient than a normal stator, if it is less then there is no point exploring the idea hehehehe, don't want to go backwards :-)

Anyway, I am very glad to have such great comments and suggestions so far.  I know I am probably re-inventing the wheel, but it really helps when you have knowledgeable people contributing input.  

Thanks

[Flux]

Here is a picture that might explain things.

Each phase has as many coils as poles, with alternate ones reversed, it is not just one continuous winding.

Flux does not cross from one rotor to the other, it loops from pole to pole on each rotor through the toroid.

It could be done as a radial with a long thin cylinder toroid with magnets inside and outside. Imagine a Gramme ring dynamo with poles in the centre as well.

With the case with a single rotor winding each side of the core, just treat it as two machines back to back. The air gap for each is the bit where the windings are, they are both single rotors, don't confuse it with a dual rotor where the flux crosses from one rotor to the other.

Flux

[finnsawyer]

If you had said ring or torus instead of steel coil it would have been clear.  As it is, this won't work.  The flux has to pass through the coils at right angles to have any effect.  In this case the flux is parallel to the coils.  If you were to take the ring and split it in two with an air gap and have the magnets on a rotor pass through the gap you would then get excitation of the coils.  It's not clear how good that would be, but it is clear that one could pile considerable copper wire on each half ring and that the magnetic path for the flux could be preserved at all times.  The magnets, of course, would have to be arranged so that two magnets are passing through the gaps at the same time and arranged with the poles opposite so the fluxes add.  Basically you need an even number of magnets with the poles alternating along the rotor as usual.  Furthermore you don't need a ring.  One could simply cut the frame of a transformer in half.  I do see some mechanical issues with building this, but I believe it could be done.  Have a nice day.  

[Aelric]

Thank you both for your input, I actually just 2 days ago ordered Hugh Piggots book.  I know I have a lot to learn :-)  just wanted to run a few ideas by some people.  Thanks again.