3 phase coils

[gordon01639]

Hi I am new to windpower but am experimenting with PM alternators, I have noticed that when wired as a three phase alternator the ratio of magnets to coils seems to change for example 12 magnets 9 coils. I was wondering why this is? what difference if any does it make to the alternators performance? if anyone can furnish an explanation I would be grateful.

Gordon.

[harrie]

I will give it a try, 3 phase duel rotor

12. mags, 9 coils== three coils per phase
    
16. mags, 12 coils = four coils per phase

20. mags, 15 coils = 5 coils per phase
    
24. mags, 18 coils== six coils per phase.
    
    

must have even number or mags, to get even number of North poles, and even number of south poles. Rotors with mags need to be layed out so mags of the same pole pass thru legs of coils in phase at the same time in all three phases. doing a search in the Google will give you alot of information, I have been on this board going on three years, and I still reley on the Google for information. hope this helps,

great fun, Harrie

[Flux]

For a single phase winding it it usual to have one coil per pole,but you can leave half the coils out and double the number of turns on the remainder with much the same results.

Normally a 3 phase winding has 3 single phase windings spaced at 120 electrical degrees,so it would have 3 coils per pole.

Let's consider a 12 pole machine, the basic single phase winding would have 12 coils. Three phase would have 36 coils. As I said before the single phase case could be wound with 6 coils so the 3 phase could be wound with 18 coils. Both types of winding are common and in every case the coils of the 3 phase winding need to overlap.

When you try to build axial air gap type alternators, overlapping coils causes problems due to the shortage of space at the centre, so it is common to leave out even more coils. If we leave out half the coils of the 18 version and use 9 we find that they will all lie on a single layer without overlap.

This is common for the types of alternators built here. The thing to watch is that for this trick to work your poles must be a multiple of 4 and for every 4 magnets you use 3 coils.

Don't try it with pole numbers that are not divisible by 4 or you will not be able to do it in a single layer winding.

It is not conventional so if you look at text book on electrical engineering you will find no mention, so the only information you are likely to find is in the archives here or on windmill sites.

Flux

[Seaspray0]

Three phase is fairly easy to concept if you looked at it this way...

Place three pickup coils near the tire of a car where they are 120 degree's apart from the hub (looks like a Y shape).  Place a mark on the tire and spin the wheel.  You can see that each coil will pick up the mark but at a different time (equal distance of time between them since the coils are even spaced around the tire).  This is the basis of 3 phase.  Lets put the north pole of a magnet towards the mark on the tire with the south pole 180 degree's on the other side of the tire.  For every revolution we'll make one sine wave of voltage in each coil (same voltage but at a different time).  If we double the number of magnets to 4 (two north, two south spaced around the tire), we can double the frequency while still maintaining the equal time between the phases.  If we tripple to 6 magnets, then we run into a problem because we no longer maintain the equal time difference between the phases for the sine wave (same sine wave but no longer seperated by equal amounts of time).  You would have a magnet passing the coil simultaneously for all three pickups.  To have three phase you must maintain the equal time spacing between the phases which is why the ratio of coils to magnets is not the same.  The ratio of 4 magnets to every 3 coils works well.

[marvin]

I guess I'm confused also.

I have 50 25mmX25mmX3mm N45s and had planned to use them to build 3 dual rotor PMAs.

My plans were to mount 8 on each, approx. 12" disc at 45 deg intervals and use 24 coils, one every 15 degrees. Every 4th coil could be wired either series or parallel to increase voltage or current respectively paying attention to phasing so they wouldn't cancel. I had hoped to use smaller wire (#20) and high turn count so paralleling would probably be necessary to get the current up.

I haven't started yet so any input will be greatly appreciated so I don't build something useless.

thanks,

marv

[gordon01639]

Thanks for the help, I made a model on a piece of hardbord and can see exactly why now that you have a 3 coil to 4 magnet ratio if you use this ratio then you always have the same pole of the magnets lined up with the same phase of the coils.

I will keep posting on here as I progress with my prototyping.

 Gordon.

[finnsawyer]

I don't see the point of using 24 coils with 8 magnets per rotor in a dual rotor set up unless you overlap the coils.  Keep in mind that a coil will only be producing voltage when the relative motion between it and a magnet is such that it is either moving onto or off of the magnet.  The coils between magnets will be producing no voltage.  With your proposal only eight coils can be centered over magnets at any given time and hence moving off of a magnet.  The other 16 will be between magnets and in a dead zone.  With 24 coils it is possible to construct a type of single phase single rotor alternator using 16 magnets (a 3:2 ratio in general).  I explain the concept in my diary.  Or one could do 12 coils with the 16 magnets in a dual rotor.  Since my proposed design also has coils passing through a "dead zone" it has some similarity to your scheme.  You might check it out.