Question for Underground Lightning Ground regarding stator design

[windy]

Question for Underground Lightning Ground,

Second:  To get the most out of a given amount of magnet material, you want the gap between the magnets (in which the copper sits) to be about as thick as the sum of the thickness of the two magnets that are facing each other and creating the field through the coil.  Starting from a thin set of coils and a small gap, as you add more copper, thicken the coils, and move the magnets apart to make room, you add generation from more copper more than you lose generation from less mag field due to the larger gap.  But that flattens out, then turns down.  When your coils are thick you gain less generation from adding still more copper than you lose to weakened mag field from increased gap.  The maximum (assuming the mechanical clearance gap is zero) is with the thicknesses equal.  (But it's not too critical, because the curve is horizontal at the max and gradually turns down going away from it.  So you can go tens of percent, and goof around with some clearance between the coils and the poles, without substantial loss of power.

One thing I noticed in the spreadsheet that you posted is that the coil thicknesses are calculated to be 1.5 inch thick.  The air gap between magnets would have to be about 1.75 inches because you can't expect a 1/32" clearance between rotor and stator to work on such a large alternator.  But the coils are too thick anyway - the flux per pole is getting low with the magnets so far apart.  As a rule of thumb, keep the air gap roughly equivalent to the magnet thickness.  For a 1" gap, need 1" magnets.  Not a hard rule, but if you play with the numbers in the spreadsheet you'll see the flux drop well below 1 Tesla as you move them apart.  I tweaked them and got better coil sizes that match better the magnet size and has a higher flux through the gap.  These numbers look pretty realistic. 

When trying to design my stator for my 20 foot variable pitch windmill, I was told to try to keep the air gap thinner than the thickness of one magnet. You state that the air gap should be equal to or less than the sum of both magnet thicknesses. I am using 1.5" wide 
x 3" long x 3/4" thick neo magnets on 1/2" steel rotors. I don't plan on shorting the stator to stop it.

I am just confused that there is that much difference in the air gap between the two designs.

Any comments on the differences?
windy

[Ungrounded Lightning Rod]

Question for Underground Lightning Ground,

... To get the most out of a given amount of magnet material, you want the gap between the magnets (in which the copper sits) to be about as thick as the sum of the thickness of the two magnets that are facing each other and creating the field through the coil. ...

... As a rule of thumb, keep the air gap roughly equivalent to the magnet thickness.  For a 1" gap, need 1" magnets.  ... 

When trying to design my stator for my 20 foot variable pitch windmill, I was told to try to keep the air gap thinner than the thickness of one magnet. You state that the air gap should be equal to or less than the sum of both magnet thicknesses. I am using 1.5" wide 
x 3" long x 3/4" thick neo magnets on 1/2" steel rotors. I don't plan on shorting the stator to stop it.

I am just confused that there is that much difference in the air gap between the two designs.

Any comments on the differences?
windy

I had understood that the peak (for the ideal case of the "air gap" exactly full of copper coil and rubbing on the magnets, the pole pieces being entirely the magnetic material, the disks thick enough to be effectively perfect "magnetic conductors", consecutive poles  substantially farther apart than  the gap between the facing poles (in an axial flux the gap between consecutive poles would be about the width of the poles) to be where the thickness of the gap was equal to the sum of the thicknesses of the magnetic material in the two poles, rather than SparWeb's rule of thumb of a gap equal to the thickness of each pole's magnetic material, i.e. half their sum.

But I inferred that from others' postings some years back and didn't work it out myself.  (I also haven't actually BUILT such an alternator and experimented with this).  So I could be wrong.

Then again the "gap full of copper" equalling the sum of the magnet thickness, if true, is about getting the most from a given set of magnets, while the "no more than one pole's thickness for the gap" may include other design considerations (such as having a substantial real air gap in addition to the coils so they DON'T rub B-) ).  Adding a clearance gap would put the most-out-of-the-materials point elsewhere.

Maybe if SparWeb is watching he can help clarify this.

(Meanwhile, even if I have this detail wrong, the prescription is the same:  One double-radius alternator gets  more out of a given set of rare-earth magnets than two single-radius alternators.)

[electrondady1]

the air gap is usually  the thickness of one magnet.

[Adriaan Kragten]

In all of my KD-reports about axial flux PM-generators I give a calculation of the flux density in the air gap. This calculation is rather simple if it is assumed that the magnetic resistance of the magnet is the same as for air. The maximum flux density is realised if the north and the south pole are magnetically short-circuited by a steel arc which isn't saturated. For this situation the flux density is equal to the remanence Br for the given magnet. On the website www.supermagnete.de you can find a list in which Br is given as a function of the magnet quality. In this list you can see that for a magnet with quality N40, Br is in between 1.26 T and 1.29 T (Tesla), so the average is 1.275 T.

The flux density in the air gap is reduced proportional with the ratio in between the magnet thickness and the sum of the magnet thickness and the air gap. So if the air gap is the same as the magnet thickness, the flux density is halved and becomes about 0,64 T. However, this is only true if the steel in the armature isn't saturated which is the case for about 1.6 T. So the armature sheets must be chosen thick enough.

If you have two steel armature sheets with magnets at the inner side, then you have two magnets for one air gap. If each magnet has a thickness of 10 mm and if the air gap is 20 mm, the flux density is half the remanence. So the larger the air gap, the more place is created for copper in the stator but the less strength of the magnetic flux in the air gap is realised. It might be that an air gap which has the same width as the total magnet thickness for one air gap, is the optimum. For my axial flux generators I use only one armature sheet with magnets so in this case the magnet thickness must be equal to the air gap if you want a flux density of half Br.

This calculation is only correct if the magnets are not positioned close to each other so if almost all magnetic flux is flowing through the air gap in between the magnets. If the magnets are mounted close to each other, a part of the magnetic flux is flowing directly from one magnet to its neighbour and is so not passing through the air gap in between the magnets of the two armature disks.

[windy]

Thanks for all the comments.

I checked on the magnet spacing and they will be 1" apart at the closest point. So I should be fine with a 3/4" air gap.

While I am asking questions on stator designs, when building the stator, can I install a small piece of copper tubing against one of the coils so I can install a temperature sensor to monitor temp.

I was thinking of using a 1/4" OD tubing about 3" long and bring it out through the edge of the stator. Will the tubing create heat and give me a false reading?

windy

[electrondady1]

ok just to clarify the terms used on this site. the air gap is the space between mag rotors.  mag spacing is the distance between magnets on the same rotor. i would think the most successful alternators constructed over the last decade on this page have used an air gap the same thickness as 1 magnet.
 if your mags are 3/4 " thick. that should be the width of the air gap .

[SparWeb]

Hi
Didn't see this until tonight.  I spent most of the weekend outside, tilting a tower up and down.  Don't get excited: radio tower in this case

ULR is on the right track; we can do a lot of rough sizing with our thumbs... rule of thumbs I mean.

The thickness of the magnet disk needs to be enough to carry the magnetic field inside and not let it leak out.  Half the thickness of the magnet is the minimum I believe.  Adriaan is not wrong; there will be some losses if you use the minimum, but not a lot and what I see from Windy's writing, it seems that 3/4" thick mag's on 1/2" thick plate is well above the minimum and works OK.

Keeping the gap between magnet disks to the minimum allows space for the stator and a clearance to prevent rubbing and hopefully not much else.  The physical space clearance you allow between the stator's faces and the magnets spinning is determined by how stiff you think the rotor disks are, and how confident you are with your hub, etc.  Maybe you can tolerate 1/8" physical space, maybe less maybe more.  Consider that these turbines are big gyroscopes, and as they pivot to face left and right, the things want to pitch up and down.  The pitch is a pretty substantial force, don't underestimate it, especially on a 20' rotor. 

Critical question that affects the gap: will the blade hub be attached to the alternator, or will the alternator be driven remotely from the blades?  If you attach the blades to the alternator (the way the Dan's do), then these pitching loads will be like trailers going over speed bumps and you'll need more than 1/4" gap.  If you attach the alternator to a driveshaft taken off the blade hub (the way Midwoud is doing it), then the pitching moment doesn't warp the rotors, and you may be able to get the gap down to 1/8" or less.  Below that is dependent on your fabrication skill.

The next question is how thick the stator can be... that's where Windy's question really comes from.  I take it you're still considering possibilities designing your stator, and if it's coming out to be fairly thick, then let's look at this carefully.  Are you getting the phase resistance we worked out for you a few months ago?

My guesstimate for magnet spacing:
magnet thickness = 0.75 inch
2x magnet thickness = 1.5 inch = maximum space between opposing magnet faces

stator thickness = 1.5 inch

Air space left for clearance (stator centered between magnet faces)

(1.5in - 1.0in) / 2 = 0 inch  = no clearance

So let's try something else, using the minimum clearance that I think you could get away with if you are really careful.

1.5in - 2*(0.125in) = 1.25 in = maximum thickness you should make your stator.

So if you go back to the spreadsheet and try to get the thickness of the stator down to 1.25" then let's see what else happens.  Does some other parameter go out of whack?

Here's another look at the big picture: http://sparweb.ca/2_Gen_Ax/AXIAL_FLUX_HowItWorks_V4.pdf
My one contribution to the internet that's been plagiarized a thousand times.  I googled it, and this time it turned up in a dozen spots.  The original is way down the list.  All these websites aggregating others peoples' stuff get more hits. 

[windy]

SparWeb,

In regards to your questions, my design has all of the pitch mechanism on a separate 1/2" steel plate attached to a 2" shaft. The two 1/2" rotors will be bolted to the back of the pitch mechanism plate with steel spacers in between the rotors to hold the air gap. The bolts will go through all three plates and tightened with six 1/2" diameter bolts . All three plates would have to flex to rub the stator. I don't believe they would flex, but could be wrong.

Regarding a comment from Adriaan,

This calculation is only correct if the magnets are not positioned close to each other so if almost all magnetic flux is flowing through the air gap in between the magnets. If the magnets are mounted close to each other, a part of the magnetic flux is flowing directly from one magnet to its neighbour and is so not passing through the air gap in between the magnets of the two armature disks.

I checked the magnet spacings on the rotors from one magnet to its neighbor and the spacing is 1" at the closet point. Reading his quote, does that mean that the air gap should be no more than 1"? Or am I reading that wrong. The rotors are already cut so I can't change the magnet spacings. The magnets will be 1.5" x 3" x .75" thick N42.

Regarding the stator design, I am still in the design stage. If I can increase the air gap to 1" or 1.25" then I think I can get get the phase resistance  to 2 to 3 ohms. Haven't tried it with the spreadsheet yet.

I was trying to design the coils using 10 gauge wire, but I seen in your link, http://sparweb.ca/2_Gen_Ax/AXIAL_FLUX_HowItWorks_V4.pdf  that 10 gauge is good up to 55 amps. I don't think I need to have the current that high if I can keep the voltage up higher at full output (10KW). Will have to try it with 10 to 14 gauge wire. Will post results back at my other post. https://www.fieldlines.com/index.php/topic,149463.0.html

Thanks again for the comments.
windy

[SparWeb]

...Will have to try it with 10 to 14 gauge wire...

That's the parameter to play with - exchange thinner gauge for more turns; presto new phase resistance.

Also consider workability of the thick wire.  It's pretty hard to bend 10-gauge solid copper wire around a 3" circle.  You can pair up two thinner wires whose cross-section adds up to the same size, and they will wind around the coil together more easily.  This is referred to as winding "two in hand" and you'll probably benefit from doing it this way.  Maybe even "three in hand".

About the 55 amp rating on 10 gauge.  I think that comes from the self-heating of the copper through resistance, and the temperature limit of the varnish on magnet wire.  Obviously when the current is high, the wire can get hot enough to break down the varnish insulation.  If you were to read into the details carefully (beyond that particular table) then you would find different varnishes with different temperature ratings - implying that you can put more current through the wire with high-temp rated coatings.  Working that out is a bit beyond me, though.

The hub sounds pretty stiff.  One less thing to worry about.

[Adriaan Kragten]

The calculation of the magnetic flux which is lost in between two adjacent magnets is difficult and different from the calculation of the magnetic flux which is passing through the normal air gap. For the normal air gap and two armature sheets with magnets at the inside, everywhere there is a north pole opposite to a south pole and all the field lines are about parallel to each other. Everywhere the strength of magnetic field is about the same. For two adjacent magnets (on one vertical armature sheet), one magnet has the north to the left side and the other has the north to the right side. At the left and at the right side of the magnets the difference in magnetic strength is maximal so there the magnetic flux is maximal too. But half way the magnet there is no magnet flux. So there is a magnetic loop in between both magnets which is strongest at the outside.

Another point is that when you use circular magnets, the air gap in between adjacent magnets is minimal at the line which connects the heart of both magnets. But the further you are away from this line, the longer the path is which a field line has to follow. The longer the path, the larger the magnetic resistance, so the lower the flux density. I have never tried to make calculations for the losses in between adjacent circular magnets but I think that these losses can be neglected if the distance in between the sides of the magnets is the same as the magnet diameter.

For rectangular magnets, the distance at the pitch circle must be at least the tangential width of a magnet. The magnets can be placed closer to each other which means that more magnets can be placed on a pitch circle with a certain diameter but in this case some of the magnetic flux will be lost.