poles versus coils

[artv]

Hi all , been doing lots of reading ,I don't uderstand the realationship between how many coils per poles. Ed Lenz seems to use 1:1 ratio eg. 6 mags 6 coils three phase =18 coils ,here 12 mags 9 coils total for three phase.Or 8mags 6 coils?? I'm totally lost........I finished my stator core (~210' of 26guage steel 3/4" wide) , my mag rotor has 6 poles (I think)...ns,ns,ns. My stator core is 4 1/2" inside dia. outside dia. is 9",....rotor is is stacked magnets but same dimensions only there is a 1" space between the magnets. Trying to do test coil but I'm not sure how many coils I can put in, in the end. I don't know what size to make the test coil, or even wire size, or how many turns??...........the more I work on this the more frustrating it gets.........any help would be aprreciated............thanx ........artv

[ghurd]

First get an understanding of a 1-ph machine.

Then draw out the magnets as accurately as possible on paper.
Make a dozen copies.

Draw each phase of 3 phases, one at a time, on separate copies...

Do Not look at all 3 phases at the same time until after you comprehend one at a time.


A single layer wining (as usually seen here) is very different than an overlapping coil design (as seen in some of Ed's and some motor conversion PMAs).

What some people call the 'wasted space' inside the coils is used the overlapping coil designs.
G-

[oztules]

This link is from Hugh Piggott's site.

http://www.scoraigwind.com/CABLE/stator.html

Look at it ... and keep looking at it until your totally happy with how it works. Take careful note of which magnet is where, and what it's voltage amplitude is at any given time in the cycle.

The line graphs on the bottom give you the voltage in real time. for each phase. Note where a magnet is at time of maximum  voltage/amplitude, and every where in between.
ie whats happening when a magnet is over the hole, over the coil leg, in between coils etc.

If your iron is only used for a backing, then this simulation will be fine. If your using a slotted iron core..... then this simulation will not be useful


.........oztules.

[Flux]

Think of the single phase case. For 6 poles the logical conclusion is one coil per pole. As each pole is of opposite polarity you need to reverse the connections to adjacent coils so the volts add in series.

If you make it 3 phase then you distribute 3 of these windings equally at 120 electrical degrees. This gives 18 coils and the coils have to overlap to get them in.

For the single phase case you can omit alternate coils and then they are not reversed as each coil Will, be under a magnet of the same polarity. The missing polarity coils are formed as virtual or coincident poles and you have more space to wind. The net effect is more or less the same and this method id widely used in electric motors produced in N America, less widely used in Europe.

The 6 pole 3 phase case now has 9 coils and the coils need to overlap but you can overlap them in a different way if you wish.

If the number of poles had been divisible by 4 you could have left out half the coils again and the end result would have fitted on a single layer for 3 phase. This is where the 4/3 magnet coil arrangement comes from and it is easy to wind for axial machines where overlapped coils are difficult. It can't be used for your 6 pole magnet.

If you have chosen 6 poles then you are stuck with 18 or 9 coils for an air gap machine. If you look at electric motors you will usually find that each coil of the pole pitch is spread between more slots and may be in 2 , 3 or more sections. this is merely to keep the slotted core a sensible size and shape and it helps reduce various harmonics but this is not needed for an air cored winding.

For a small air gap machine I would not have chosen 6 poles but you can do it, things will be tight for space at the centre where you have to crowd the end connections into a restricted space.

Flux

[artv]

G-,& Oz,thanks for the replies,that animation should be put in the faq's.Oztules ,I thought by making slots in the core all you are doing is bringing the mags closer ,making for a more concentrated flux path? I was going to do test coil find the best one then cut slots .Whats the difference between having the coil laying on the suface of the core or being sunk down in ,other than the strength of the flux???                                                                                                                                          Flux, thanks for the explanation ,I knew six poles were a bad choice but it was all I have to work with.My pole faces are 2 1/4" in dia. spaced 1" apart,...if I follow the rule for coil size (the hole ~the same size as mag)I only have room for six coils on the flat surface but if I cut slots then I can fit in the 18??? Maybe I should just make this single phase sounds alot less complicated.One last thing if I do cut slots should they be skewed or just do straight ones for now.......thanks for the help think Iunderstand.....artv           

[Flux]

Yes making slots in the core does let you reduce the air gap and get more flux. With the old wound field machines it was essential to keep the field requirements within limits. For ferrite magnets again it keeps the size and weight of magnet within limits.

For neo there is still some gain but the issues are not quite so compelling unless you are trying to use the smallest quantity of magnet. You gain in some areas and loose in others. Invariably you will get a certain amount of cogging with slots unless you spend a lot of time experimenting, you can get close with calculation but for total reduction it's back to trial and error. The iron loss also goes up with the higher flux densities so you get a lighter and cheaper machine for full power but at a cost in terms of light load loss so it may not be the best choice in a low wind area.

You can still make a good machine with coils stuck on to a core with no slots and it will not cog and may have less start up drag so in a lot of cases it can be a good choice. there is little doubt that it is easier to wind a coil in slots than make a self supporting coil structure that you can stick on the surface but again it can be done even with overlapped coils.

The 3 phase arrangement has overlapped coils that can be a challenge to wind, but the 3 phases interleave and you wind all the area and not leave half unwound as for the single phase case.

If you go without the slots then keep the coils fairly thin, you then get a good gap flux and it is much easier to wind. The cooling is still quite good with coils stuck on to the steel core so you can use a thinner stator than with a dual rotor.

If you go with slots then you must not burr over the edges when you cut them or you must unwind the core and remove the burrs, there is a question about whether it will rewind correctly without staggering the slots.

Skewing if done correctly could reduce cog but incremental spacing of the magnets would do this without skewing and also the relative shape of magnets and slots has a lot of effect on cog so it's not a clear cut case.

Finally the 4/3 rules don't really apply to multilayer windings so you can make your coils span a pole pitch and make the hole such that 3 coils will overlap and just fill the space. you can still get 18 coils in without the slots. this is too tricky to explain here but for a 3 phase overlapped coil winding the coils can't be the ideal shape of the single phase case. You need gaps between the magnets half the magnet width and you seem quite close so a 9 or 18 coil would work perfectly well without the slots.

Just imagine you have 18 slots and wind the coils as though it was a slotted core and it should work out.

I hope this gives you some ideas, it would take forever to describe every form of winding but you should get the basic idea from "Windstuff now" site, Ed gives quite a few options.

Flux

[Boss]

http://www.scoraigwind.com/CABLE/stator.html
Awesome Flash animation; mesmerising, hypnotising
Thanks for sharing. Now if I can only take my eyes off of it

[oztules]

".Whats the difference between having the coil laying on the suface of the core or being sunk down in ,other than the strength of the flux??? "

In the animation, the voltage is in real time, synchronised to the alternator so you can see spacial relationships  between magnet position, and emf generated.... so the animation describes what is happening in real time.... thats what it is for.

For a sunk in slot arrangement..... slotted core, the voltage graph would be way out of phase..... ie max voltage would not be with the magnet directly over the coil legs, and min voltage would not be when the magnets are in the "hole" of the coil. The animation would be wrong, with emf going up when it should be going down, Vmax at the wrong time etc.

So the animation would need to be re-written to reflect this.





............oztules

[artv]

Oztules ,...does that mean the slots make the magnetic  flux kind of drag (slow it down so it is no longer in the same spot or phase)??                                                                                                                                                                               I tried 4 different test coils , the best I could muster was .2vac, that bites.I think it has alot to do with my crappy coils ,but more so the wobbly mag rotor.So I dug through my vast amount of goodies and found a stack of magnets I forgot I had.I made a new rotor that now has eight poles.So I can put six coils in ,two coils firing every 15 degrees of revolution ,giving me three phase correct? With the new rotor I have 1 1/2" space between rectangular magnets on the inside dia. on the outside they are 2 3/4" apart, with an overall outside dia of 9 3/8" still using the same  laminated core.Would it be better with six coils or should I try to put 12 in there,I'm just thinking 12 would run alot smoother??.Also the pole faces' are 7/8" wide so does this mean the coil legs will be half of the 7/8"s...........hopefully this rotor will give some better results...........still along way to go,..but I could have never gotten this far without the help of everybody here so again Thank-you ,it is very much appreciated.................artv