Wiring a stator with 16 magnets and 24 coils

[lieutenant]

So I decided to make a one rotor generator with 24 coils and 16 magnets (on the same side) I want to wire it in STAR 3-phase. The question is how to wire it properly?

 Actually I already did wire it this way: http://www.otherpower.com/images/scimages/6/statorwiring.jpg but seems like it is not working quite well. I got only about the same amount of output from one phase when measured as from one coil tested alone before. Shouldn't one phase give 8x the output of one coil because there are 8 coils series'd? I dont have a " bridge rectifier"(?) so I cant test the total output of STAR-connection yet.

If someone could help, id be very thankful.

[electrondady1]

perhaps you could do a new diagram showing how you managed to put 24 coils over 16 magnets.

[lieutenant]

Well, seems like i just did a classic mistake by labeling the end wire A as C and C as A :// Thx for the reply though.

[DanB]

It doesn't really make sense to me, 16 magnets, and 24 coils - unless you have fairly large coils that are somehow overlapped.  With 24 coils & 3 phase, then 6 of your coils (Every 3rd one around the circle) needs to see the same 'magnetic' situation and I doubt yours are.  We get that when we use the ratio of 3 coils/4 magnets....  but 24 coils /16 magnets doesn't work out unless you've done something unique with your winding.

[Flux]

This has to be an overlapped coil winding to work. The basic 3 phase derivative from 16 poles will have 48 coils. You can leave half out and use 24 coils with double the number of turns over the same polarity magnets.

You can leave half out again and only use 12 with 4 times the number of turns and then it will fit on a single layer without overlapping.

If you have 24 coils in a single layer it's not going to be much good.

Flux

[oztules]

Flux,

I have recently seen this one ...."You can leave half out and use 24 coils with double the number of turns over the same polarity magnets." ....on the AWP windings. (7 coils cooked from corrosion). It is a six layer winding using 30 magnets and 90 coils (55 turns), but I found that it is really 30 magnets and 45 coils (110 turns).

It could have been wound with only 45 coils but they elected to wind two coils on top of each other rather than 1 of twice the turns.

The good thing about this setup, is all the coils are wound the same direction.

Is this consequent wound?

.........oztules

[Flux]

Yes, with 30 magnets and 90 coils it would be based on 3 single phase windings each with one coil per magnet. The 3 windings will be displaced 120 degrees electrically to produce 3 phase.

We only need to consider the single phase case here so a full winding would have 30 coils and because each alternate coil was covered with a magnet of different polarity you would need to reverse alternate coils ( SFFS etc).

If you remove half the coils and leave 15 all over the same magnet polarity it still works ok and the coils remaining are connected in series without reversal ( NSNS etc). You can increase the turns to compensate the voltage and you have the extra winding space to use thicker wire.

If this is a closed iron circuit ( which it is)and you feed a dc current into the winding you will have identical poles over the centres of the coils. The flux has to complete the path and it will return via the spaces between the poles produced. There will be consequent or phantom poles formed between the poles at the centre of the coils and it will behave exactly as a 30 pole winding although you have 15 coils.

Both winding schemes are used commonly, N.America tends to favour the consequent pole winding while Europe probably has more small machines wound with full coil in a 2 layer winding. There is negligible difference and often rewind companies will change the method to suit their facilities. For hand winding many go for the consequent winding, whereas those with winding formers may go for the 2 layer winding where all coils are the same.

Flux

[lieutenant]

Well now I actually understand that 16 magnets and 24 coils isn't the right ratio. But I think that I have so much space left on the rotor that I could buy 16 more of these magnets Would it work then since the coils/magnets ratio would be 3/4 ??

If its wise to use 32 magnets and 24 coils is a whole different story..

[Flux]

Not enough information to make any judgement. 32 magnet 24 coil will work if proportioned right but when you are in a mess, just changing things at random is likely going to get you in deeper.

If you can post some pictures or give an accurate description of what you have and details of voltage, prop size, magnet size etc then perhaps we can help.

If with 32 magnets the space between them is comparable to the magnet width and the holes in the coils are about magnet size plus or minus a bit then you could make something of it, but what voltage or power or speed I couldn't guess at present.

If you have that much space between magnets you could make your stator work with all magnets on one rotor one polarity if the coils shape and proportions are right, but I won't add even more confusion as you are struggling already and unipolar designs are not discussed here.

Flux

[lieutenant]

Thx for the reply. So I'll add some more information. First the background. The idea of this whole project was just to make my first windturbine to learn the basics (trial and error) and then if it would work nicely I could build a better/bigger one. And now seems that I've already gone wrong

So I decided to make a VAWT, savonius since its easy to build and the wind conditions are quite variable. So the stator and rotor diameter is 50cm. I did read from somewhere that I should use 16 magnets and 24 coils which was OK because I had so much enameled wire (21 AWG??) and about 8 magnets (20x10mm round N42) already. Seems like the 24/16 design was a whole different thing than normal 3-phase :S So I've wired the coils like in the picture link posted above (star) except I have more coils..

I measured with a multimeter that I get about 4V and 0.3A at 50-70 RPM. Which is nothing IMO :/

Hope this info helps a bit. And hopefully someone can give me some advice

[kurt]

lieutenant, you are not giving the proper answers to the questions so anybody can help you.... how about giving coil size id and od and thickness, are you using a steel backing plate for the magnets if so how big is the plate?? how big a circle are your magnets laid out in outside edge to outside edge. that mite get ya started.

[lieutenant]

Ok, thx for advice. Here they come: Can not measure the coil size exactly right now, but it has 20m of 21AWG wire and height is about 6cm, width 3cm, 1cm thick. Not using a steel backing plate and the magnets are laid about 5cm off the outer edge. Coils are laid in one layer about 2cm off the edge. What I'm hoping for now is an answer for the question that does doubling the amount of magnets from 16 to 32 help. I don't really have the mood/time for changing the whole design anymore :/ The main thing is to get it working somehow nicely. I'm not expecting much output from a savonius anyway.

Thanks.

[Flux]

Worse than getting blood out of a stone, but if I have managed to make something of what you have then it should work with 32 magnets.

With 32 magnets you will have spaces about the diameter of the magnets and that will be ok.

I don't know why you use rectangular coils with round magnets but again it will work. It would seem that you have spaces between the coils from your dimensions but that doesn't matter. Still no idea of the hole size in the coils but again whatever it is it will work in some way.

You need the magnets to be alternate polarity, depending on what you have done you may have to reverse some of the existing ones.

I can't decide what the magnets are mounted on, it seems to be a single rotor with no return disc. If the magnets are on a steel disc, give it a try it should do something.

If they are on a non magnetic disc then scrap it and forget the whole thing it will never be worth the effort.

If you can make a single rotor with magnets on a steel disc with magnets of alternate polarity and manage to connect your coils  correctly it will produce something. Whether you can get a standard voltage at a speed to suit a Savonious I know not or care not, I can only help with the alternator.

Good luck.

Flux

[lieutenant]

Well I actually used the round magnets because I already had a few of those. I know its not so efficient as using block magnets. The hole size of coils is something like 2x1cm. And yes, atm the magnets are on a non-magnetic disc. But under the stator I have some steel. Don't know if it helps. I might actually find some steel from the garage and make a steel disc for the magnets, so that could be worth a try.

All I want now, is to get this somehow working and then just move on to other things. Maybe I'll just go for the 16 more magnets even if this project looks quite doomed. Those aren't really so expensive. Atleast I think I've now learned how 3-phase works and won't make the same mistakes next time

Again, thanks for the very pragmatic reply ! It really helped.

[Flux]

The coils will be ok with holes that size.

Try to get the magnets on a steel disc. The steel disc under the stator will increase output voltage but the eddy current drag in it will most likely mean that you will need to remove it. Perhaps with the low speed and high torque of the savonious it may not be so bad but all iron in a changing magnetic field needs laminating in some way.

Flux

[lieutenant]

Yep, the disc under the stator gives some more drag, but its not really dramatic. Not sure if its 100% steel/iron, might be an alloy also. I've considered of removing the disc already, but I'll leave it on for now. Lets see if I can find some steel to make a nice round disc of it to mount the magnets on. Well see what happens....