long coil question

[TheCasualTraveler]

     I have a question about coils, again. In my research I have read different opinions about whether or not there is any gain from the top and bottom of the coil or if it is just added resistance. I don't know the answer but assuming the two side legs produce most of the power, and for argument all other things equal, would long skinny coils work more efficiently than round or square ones from the point of keeping the top and bottom legs short (less resistance) and the power producing sides longer?

     Also, I remember something of a post saying that in longer magnets there was a problem with the flux pattern toward the middle of the magnet. I think the point being a loss of power if the magnet is too long. As often happens, I saw all this when researching something else and can't find it now. Anyway, is this true? Below is a diagram illustrating what I'm talking about and also know that I consider this for radial, not axial flux. Also, I'm NOT talking about a laminated stator or skewed magnets or skewed coils, just a simple radial flux (like a brake drum machine) but with longer magnets than the common 2".

[Flux]

This little trap comes about as a result of a basic fundamental ( and useless) equation from your physics book  E=Blv. it is for a voltage induced in a wire moving at constant speed in a magnetic field.

In real life volts are useless, we need power a combination of volts and current. To have current we have to join the ends of that wire and there comes the snag.

Regard the thing in terms of flux linkage and it will make sense. All parts of the coil are necessary, nothing is active or passive. Ultimately if you remove all the other restraints the coil with greatest area for least wire length is the one to go for and that is circular. Practical factors usually dictate that we use rectangles for most applications.

Flux

[hvirtane]

I think that you are right.

When the magnet lines are passing over the ends of the coils they are moving such a direction that they are making no power?

If the length of the coil is big compared to the width, the amount of 'the idle wire' is little compared to the wire, which is producing power?

Isn't this the main reason, why in electrical machinery coils are normally long?

- Hannu

[finnsawyer]

Which would be the better deal for the same money, a piece of land 100 by 400 feet or one 250 feet by 250 feet?  That is, one having 40,000 square feet of area or one having 62,500 square feet of area?  It would take the same amount of fencing to go around them, but very few would settle on the long narrow piece if given the choice.  So geometry matters.  You can enclose more magnetic flux for a given amount of wire with a square or circular magnet thereby maximizing the change in flux while minimizing the resistance of the wire.  These choices, however, affect the diameter of the rotor.  Using 12 of your 1 by 4 inch magnets would require a smaller diameter rotor than using 2 by 2 inch magnets.  This also creates another problem.  When the magnets are rectangular and oriented radially the space between them is wedge shaped, severely so for the 1  by 4's on the smaller rotor.  This in turn limits the number of turns of wire you can fit between the magnets.  I suspect the what you see in actual builds developed over time for very practical reasons, and these seem to have settled on magnets no more rectangular then 1 to 2.

[electrondady1]

this has always been a point of interest to me and i came to the same conclusion

as hannu.

it  led me to wards radial alternators to get longer legs.(love em).

well , that and overlapping coils.

2x2=

1x4=

.5x8=

 all the same surface area.

but which will produce the greatest output for our purposes.

i don't have the equipment or deep enough pockets.

to do the real test.

and there is another factor which  flux alluded to.

that is, flux leakage due to the length of the perimeter.

[electrondady1]

just want to mention danb is now using round mags for some of his large projects.

 which offer the smallest perimiter to area ratio.

[fcfcfc]

Hi: I would have to agree with your thinking regarding the shape. Wire that is "pointing" in the direction of movement is useless. The wire with the maximum power induced will be all those "sections" that are at 90 deg to the direction of movement, cutting directly across the field. So I would think longer and narrower is better disregarding practical issues. Neo's are magnetized through their thickness so the field is roughly the same across the surface no matter what the shape, within reason. The problem you will have as the magnets get "skinny" is getting coils with enough turns and various spacing issues depending on your coil layout. So once again, practical issues can end up "shaping" the final design as opposed to just theory. My design I am working which is finally at its end point was changed about 6 times. Wave winding was what I wanted to do but I kept trying to think of a way to do its with coils because I just didn't want to have to do the HUGE labor intensive task of a wave winding. I finally figured out a way to get just about the same result with coils regarding "dead" wire VS "active" wire. My magnet poles will be 1.25" x 3", basically all touching together. Not as long and skinny as I wanted (4")put price was the practical issue, since the duel rotors will require 120 magnets, 60 poles. I will be using #11 square, 8 turns per coil, 60 coils per phase so 480 turns per phase. The system is ending up as a 48VDC job, very low cut in RPM.

[DanG]

I just read you Oct 2007 "specific questions.." post & there was some talk about not using square conductors at 12 or 11AWG. Yes, of course they will will work and work well but to get long service out of it may be difficult. Cutting back on magnet strength to begin with might be a way to work up to perfection, eh? : )

Remember you are going to have very little cooling available, no forced air cooling or fancy loomed turns back to the legs leaving air gaps for passive heat shedding, very little 'extra' wire to act as non-heat-producing sink & dissipation area for the swept area density you are trying to achieve and an extreme minimum thermal mass compared to stationery or traction motors. Going for low RPM's will help a lot but still there will be whole rafts of potential just sitting and spinning like in a river eddy, disconnected from the 'current' underneath, especially under the highest flux areas.

You may see thermal expansion problems lifting loops, even heaving its way out of its own shellac trying to shed the encapsulate from heave and shrink cycles. Without internal jigs and bracing a cast stator will be pushed toward warp or embrittlement. Your winding jigs must be near perfect to not pre-load torsion into the turns, maybe even an oven heat treatment of coils while curing an alkyld resin should be called for. There are compromises built in to using round wire that the huge potential of large Neo negates easily and don't make for easy conversation...

[fcfcfc]

Hi DanG: Thanks for the reply. I am still "kicking" around whether to go with #11 square or #15 round 3 in hand. They are almost identical in resistance, 1.033 VS 1.060. In the thermal expansion area, which was the issue I was thinking of too, I thought the greater number of small wires might have a bit more "give" then a solid square wire of that size. Ultimately I have to see what will fit in the space. The final design calls for coils laying on top of each other so there really is not much "air" space left. I am going to put thermal snap disks, 3 of them, at 120 deg spacing imbedded in the stator. These will close around 150 DegF. I would use these to "kick on" some sort of forced air cooling using a vacuum cleaner motor which would pull air through the stator gaps. This would help to hold down the winding temps a bit. I also have been looking at specific resins like this one:

http://www.derakane.com/derakaneControllerAction.do?method=goToProductDetailPage&actionForwardNa
me=derakane441400&productCode=536005

It's heat deflection temp is not that high compared to their high temp resins, but it is more flexible so it is less prone to cracking brought on by repeated thermal cycling. It has stronger adheshion as well.

I don't know what you mean when you say cut back on magnet strength..??.. I would have gone with 3/4" thick Neos but they are just too much money considering how many total magnets I need. For cost reasons I will be using an N45 3/4" x 1/2" x 3" long main magnets and 1/2" x 1/2" x 3" N42 Neo's for blocking magnets. That's where the final 1.25" x 1/2" x 3" pole measurements come from. This should give me just about 1T in the gap at 1/2" or .8 to .9 T at 5/8" gap, a strong field...

[finnsawyer]

The maximum voltage in a single turn will occur when it exactly straddles two neighboring magnets.  The flux in the part of the loop covering the magnet with the north pole may be decreasing, becoming less positive with time.  Simultaneously, the flux in the part of the loop covering the south pole magnet will also be decreasing, becoming more negative with time.  But the total flux contained in the loop at that point will be zero.  It is the time rate of decrease which according to Faraday's Law that causes the induced voltage.  The actual dependence of the voltage with time will depend on the shape of the magnets.

The end lengths of the coil turns is not important as far as generating the voltage.  There are issues with the long magnets.  The wedge shaped space between the magnets that limits the coil size is one.  Putting this aside, consider a single loop of wire the area of which just matches and fits the area of that 1 by 4 inch magnet in a typical alternator having 12 magnets.  The rotor must advance 30 degrees to create a voltage pulse.  Compare that to the case for 2 by 2 inch magnets, which also must advance 30 degrees for one voltage pulse to be generated.  Both magnets put out the same magnetic flux.  So, if the time is the same for the rotor to advance from a magnet being centered over the loop to the neighboring magnet being centered (30 degrees of travel), then for Faraday's Law the change in flux is the same and the time is the same so the voltage pulses would have the same average voltage.  But, it takes 10 inches of wire to make the loop for the 1 by 4 inch magnet and only 8 inches for the 2 by 2 inch magnets.  So, the resistance for the 2 by 2 inch case would be 80% that of the 1 by 4 inch case.  You'd get more power from the 2 by 2 inch case.

There are other issues. One is the increased leakage flux from the closely spaced narrow magnets.  There will be more tendency for the flux to flow directly through the air from one magnet to the next.  Flux that does not flow through the coils from the rotor to the stator backing can not produce voltage.  From your point of few any flux lines having a component parallel to a magnet's face that pass through both sides of a turn of the coil will produce no voltage.  The effects cancel for that component.  It would only be the perpendicular component that would induce a voltage.  The square magnets would be farther apart giving less leakage flux.  You could also probably fit more turns of wire between them (not as much of a wedge effect), wire that is being used more effectively to start with.  One the other hand you are trying to ram twice as much flux through the rotor.  A thicker rotor might be needed.  

[fcfcfc]

Hi: Well in this case, that is my specific design, I really have no side flux leakage even though my magnets are touching, due to the blocking magnets. My inside diameter will be 24" so the wedge effect is small. They will touch on the inside and maybe be about ~1/16" apart 3" up from there. MY poles are 1.25 x 3".

Regarding the induced power in wire. So, if I understand you correctly, you can take 1 inch of wire and 100 inches of wire, both straight, and pass them through a field strength M at velocity V, both at 90 degs to the field and the 100 inches gives you absolutely nothing more in terms of voltage or amps than the 1 inch of wire..??.. Second, passing a conductor through a constant field strength at a constant velocity V produces no induced voltage, or power if you wish, in that wire..??.. Only voltage is generated when entering or leaving the field..??..

[finnsawyer]

I'd like to see a diagram showing how those "blocking" magnets are placed.

As far as the second part, why one inch and 100 inches?  Why not 100 inches and 1,000 inches for the wire length passing over a one inch magnet?  In order to read a sensible voltage you must finish the loop.  That is, when you connect a voltmeter or scope to the wire you have created a loop.  It could have any crazy shape, but Faraday's Law states that it is the time rate of change of the flux passing through that loop that causes the voltage.  And how you orient the loop matters.  You get a different result if the plane of the loop is perpendicular to the face of the magnet than if the plane of the loop is parallel to the face of the magnet.  I suggest you carefully study the two cases using both your way and the way I suggest.  It may surprise you to know that properly applied, both ways give the same result, but the time rate of change of flux method relates more directly to the geometry of the actual alternators and is easier to do.  

[electrondady1]

fin,

 i'm glad your still on board

i have been thinking about this for the last hr. or so.

 first,

 with narrow mags it would be be possible to place more poles in a given space

increasing the frequency.

so at what point does the width of the mag stop being a factor in the amount of flux density it has .

i'm thinking that you want the copper over the pole  long enough for it to have the  

maximun effect

and at that point begin to change polarity.

[electrondady1]

perhaps that should read at what point does it "start" to be a factor.

 

[Flux]

This issue is a mess. You can continue squeezing magnets in until they touch and you will increase output, but you will be making very poor use of the magnets.

It is ok adding more magnets but unless the coil can link the total flux for that magnet you get poor value for money. Yes add more magnets, that increases total flux. Yes it does increase frequency and that increases output. To be really useful you must increase the diameter if you add more magnets. If you have magnets touching ( or virtually so) you can't effectively use any width of coil leg ( no winding space)

For conventional 3 phase windings in slotted cores the best spacing is roughly with magnets about twice as wide as the space between them.

For the odd part wound 12/9 arrangement you do better with more space between magnets and space at least magnet width is desirable. With radials this is a compromise and space about magnet width about half way out along the magnet seems ok.

If the loop within the smallest turns of a coil is smaller than the magnet then you don't link all the flux. Because there is a trade off between induced volts and resistance it seems justifiable to make the hole in the coil smaller than the magnet but if you end up with magnets touching then turns that should be surrounding one magnet will be linking flux from an opposite magnet. You can effectively only use coils with legs one wire wide and that gives no winding space. This is much more of a factor than the issue of leakage flux between close poles ( true only for neo).

Within reason the coil shape is of no importance as long as you link all the flux.

Contrary to what some still want to believe the length of the coil is of no importance. If you make it excessively long and thin you end up with low values of area for a given length of wire. Volts are ok but power will be less.

Commercially high speed alternators are long and thin with few poles, low speed ones are short and fat with many poles. Most of this is linked with the mechanical constraints. Windmill alternators are best short and large diameter but not with large numbers of long thin poles with little gap.

Long thin poles are probably a pain with axials more so than radials.

Flux

[fcfcfc]

Hi: OK, your not getting this finn.. Forget the 1" magnets. Pretend its a HUGE magnet 10' x 50' OK. The field within is a uniform. Now, you take a conductor, a length of wire and start waking at a uniform speed. From before, you are saying that there is no induced voltage in the wire because the field is not changing, even though the wire is moving through it..??.. That was one question. Now, if you had two wires in your hands, one 100 time longer than the other, you were saying that the one that is 100 times longer gives you nothing over the one that is 100 times shorter. Both wires are straight no turns...

For the moment, forget the blocking magnets, these two points first....

[TheCasualTraveler]

     Thank you all, very interesting comments and as I suspected this is an area of contention. Getting back to what I was originally wondering, a new question, in light of what Flux said about round magnets and coils, given the diagram below,





     Note this represents one pole of  a RADIAL flux air core machine. Assume that the 4 round magnets combined strength, area etc. are equal to the long single magnet and also assume the total length of wire in the coils is the same for the 4 round coils combined and the long single coil. (again, this is not for skewed magnets)

                                      THE QUESTION:

     Which would give the most power output, the long coil or the series of round coils?  Or, would they be comprable?

[fcfcfc]

Hi: I posted the blocking magnets over in mechanical... enjoy...

[Lumberjack]

Allow me to clarify and perhaps simplify the question a bit.

Assuming:

coil A: 1 inch by 2 inch coil with 40 turns of wire and 1 by 2 magnets

coil B: 1 inch by 4 inch coil with 20 turns of wire and 1 by 4 magnets

Both coils will produce about the same voltage, sort of...

coil A has 80 inches of horizontal and 160 inches of vertical wire.(240 inches wire)

coil B has 40 inches of horizontal and 160 inches of vertical wire.(200 inches wire)

coil B will have a slightly lower resistance then coil A and produce a bit more current.

If the turns on B were the same as A then the voltage will double however the current will be less so overall you would only get about a 60 percent gain.

[Flux]

I can't imagine why you would want to try the round coils in series, the second arrangement would be far better with the round magnets touching and one rectangular coil linking it,far shorter wire length.

No, use round coils with round magnets, rectangular coils with rectangular magnets and if you approximate square or rectangular shapes with round magnets just crowd them together and make the coil follow the approximate shape.

Flux

[finnsawyer]

Excuse me, but you're not getting it.  If you connect a voltmeter to the wires, which of course you have to do to measure voltage as you walk over your lot size magnet, you will see no voltage on the voltmeter no matter what the relative lengths of the two wires, because the magnetic flux contained within the loop so created does not change with time as you walk unless you get too close to the magnet's edge.  The same holds true for a loop encompassing both the short and long wires.  You are forgetting the effect of the magnetic field on the leads of the voltmeter.  They are wires too.  In the case of a uniform magnetic field voltages are induced in those wires which cancel out the other voltages giving a great big goose egg if the magnetic field is truly uniform (constant in value and always pointing in the same direction).

[TomW]

Geom;

I think folks lose sight of the fact that voltage is only induced by a change in field density not from the intensity of the field.

No flux density change, no voltage produced.

Be pretty cool if we didn't need that then we could create power by laying conductors on magnets with no motion.

Anyway. Just a thought.

TomW

[fcfcfc]

Hi: Ah!! yes the connecting leads... BUT suppose one end of the main wire ran to one "side, edge" of the magnet and the other end of the wire ran to the other edge of the magnet. Now we are going to put the VOM and connecting leads outside the field somehow (it doesn't matter how), so inside there is only the straight wire moving at a constant speed through the uniform field but still perpendicular to it... SO, is there a current induced in the wire..??..

[TomW]

fc;

Near as I understand no, without flux density change there is no voltage induced. "uniform field" has no flux density change. At least until it initially enters or upon leaving the field both points will induce some voltage spikes.

Thats my understanding of it anyway. Lets see if the experts agree.

TomW

[Lumberjack]

It is change in the magnetic field that causes the wire to conduct. The speed of the change governs the voltage induced in the wire.

[fcfcfc]

Hi: Then if we agree that it is only a change in flux that induces voltage, the only remaining question is, does the length of the wire effect the induced voltage assuming the velocity and field strength remain the same, using the above configuration. This giant magnet of course would have be narrower so the wire ends just go to the sides as before, but the field within will be made the same for purposes of this question along with the velocity ..??..

[TomW]

fc;

Now, that one I cannot answer for certain.

Assumption follows.

Long ago in a galaxy far far away, I studied this stuff. As near as I can recall the only factor of the wire that had a bearing on voltage was # of turns, all other parameters equal.

This implies that a .25 inch "leg"  will induce the same voltage as a quarter mile long leg. The length must have a bearing on something. Perhaps it is current flow due to the amount of copper? That doesn't fly if we apply the series circuit rule on current. Longer wire so more resistance so less current with same voltage. Ouch, my head hurts.

Actually a very good query. Hope someone who knows will chime in. I sure do not know for certain myself.

TomW

[ghurd]

I think the leg length is built into the formula with "square inches of magnet".

A 2 x 1" (2^2") neo takes 6".  

A 1.4 x 1.4" (2^2") neo takes 5.6".

A 2^2" round magnet takes 5".

Close enough to get started?  20% is better than the rest of the stuff, where I just make a guess.

G-

[finnsawyer]

Be gone for a day and see what happens.  Obviously, real magnets don't have magnetic fields that are uniform everywhere in space.  Beyond the confines of the magnet the magnetic field loses strength and also changes direction.  That's why it is possible to get a sensible voltage around a loop having parts external to the magnet.  The contributions of the relative change in the magnetic field (B) are not as strong there allowing net voltage to appear and a current to flow if connected to some device.  But the voltage out will depend on the orientation of the resulting loop. Generally the coils are made greater in diameter than the magnets for actual alternators for this reason.  Also, in actual alternators efforts are taken to channel the flux or flux density to keep the maximum flux within the coils.  That's why the concept of a magnetic circuit.  It makes use of the properties of iron to channel the flux and maximize its value.  With the flux channeled it makes sense to use the time varying flux form of Faraday's Equation.

The equation you advocate is certainly correct and valid as there is only one physical world, but it needs to be handled carefully to get the correct results for alternators.  It would require more work.

Here's an example of your idea at work.  A copper disk spinning over and centered on a round magnet.  A voltage will be developed between the center of the disk and it's rim, and you can measure it and get current out if you wish.  But you need to use at least one brush at the rim and it isn't that great an alternator anyway.  Since the leads to the voltmeter wouldn't be moving relative to the magnet, the voltage you read would be that between the center of the disk and the rim.  Another example of your equation at work would be the Hall Effect.  I once had the job of soldering leads to small blocks of semiconductor.  Since the effective charge, either negative (electrons) or positive (holes), would always be the same one could measure the Hall voltage to determine the speed at which the charges moved in the semiconductor as well as their sign.  How cool is that!      

[fcfcfc]

Hi: That "muddies" the water. Lets keep the question clear. As above, wire longer all else equal in terms of field strength and velocity, what changes besides the obvious that the wire will have more resistance. My gut says the voltage has to rise.... not just because the resistance is bigger but there is more Field effecting a larger cross sectional area of copper. If you say that the total power remains the same, I think there is a breaking of the conservation of energy rule, in a way. It has to take more "WORK" to "chop through" a greater area of copper at the same field intensity as it would a smaller area of copper, so more energy should be available minus the extra heat loss to resistance.... again just my gut here...

[joestue]

uhhh, that is hard to answer due to a poorly worded question.

if by doubling the wire length we keep the total flux the same, nothing changes, aside from the length of the end windings. and therefore resistance.

however. real axial machines need pie shaped coils and wedge shaped magnet to make optimum use of the field strength, optimizing all this has been done many times by various authorities and the optimum inside diameter to outside diameter ratio is .6 to .65, assuming pie shaped coils and magnets. To my knowledge no one has done a full computer analysis of square/coils/magnet air core design. my feeling is that it will result in the same results ignoring leakage flux concerns.

end windings are waste resistance, but compromises must be made when leakage flux is 1/2 of the flux produced by the magnet.

[Lumberjack]

Ignoring any variation(s) in the magnetic feild(s), doubling the length of the wire doubles the voltage and doubles the resistance.

Other influences:

speed - determines attack and decay profile of generated waveform

Dwell - (time in flux) aka (amount of time wire spends over magnet) - If dwell is too short the coil will not develop full voltage. If dwell is too long then The coil will quit seeing a change in feild strength and stop producing voltage. This is also expressed as design speed and magnet/coil width.

[TheCasualTraveler]

     Flux, you said, "I can't imagine why you would want to try the round coils in series " Thats just it, I was imagining. You see I still have only a cursory view of coils and magnets and their relationship and all the research through these threads leaves me with more questions than answers. You have however helped me come to this conclusion,

     Radial design machines often use long magnets or a series of magnets in a row (skewed or straight) along with a long coil because the physical dimensions of the device make this a convienent setup. But as far as the shape of the coil is concerned, whether radial or axial, for best results you simply match it to the shape of the magnet and treat a row of magnets positioned as 1 pole as a single magnet.

     I think one reason I'm having trouble understanding this is that I have viewed the flux lines around a magnet as 2 dimensional. I've gone back and looked at these patterns and tried to see them in the third dimension just as the coil would. I think I have enough of a basic idea of this now to put it behind me as I only want to understand it from a basic construction point of view, and the theory part is for better minds than mine.

     This thread seems to have taken off on a few tangents and thats fine, have fun discussing it and I'll see myself out. I have to go get started on furling. Uhg!