Magnetic Saturation?

[oneirondreamer]

I have 2 coils in my test rig, both 18 G wire, one has 25 turns, the other has 50 turns.   At the same magnet over coil velocity, they produce almost exactly the same voltage?   Is that because effectively the 25 turns contain enough copper that it's absorbing almost all the magnetic flux, so more copper is just diminishing returns?    I'm using 1x1x4 ceramic magnets, so there's not as much magnetic force available as if I was rare earth.   

FYI, the variability between waves is due to the fact that in my test rig I have a variety of magnet spacings from testing to find optimal magnet spacing.    The final magnet array I am using is 1.2 m diameter, with 72 poles, each pole consisting of 2 1x1x4 ceramic magnets.    This is very heavy, weighs around 300 lbs including the steel but seems to run just fine on my 7m^2 VAWT (no coils in yet, just Proney Dyno).   My goal is to have it producing 90V at 3m/s so that it will work with an Aurora inverter.   It will have 36 sets of 3 coils, and in this version one of my thoughts is to lead out each coil separately, so that the ALT can be configured to provide anything between all the coils in series, wired star (for max V out) to all the coils in parallel for max Amps out. 

Thanks all!

[arc]

Sounds interesting, are you planning on using wind power?
Getting that mass (300 lbs.) moving may be difficult but man-o-man, once it's turning it will stay turning for a while.
With good bearings (you'll need them) it will tend to act as a battery (or perhaps more like a capacitor) and iron out fluctuations in wind speed.

Arc out

[electrondady1]

that makes no sense drew.
two times as many turns  and the same output?
with out seeing the coils, i am assuming one is twice as thick as the other.
so the air gap between rotors is much larger?
if they are not twice as thick then the coil leg width must be twice as wide and you may be getting cancellation by having the legs over two mags.

no magnetism is absorbed by the coils.
the magnetism is pulled from one rotor to the other.
the lines of flux go through the coils and change the polarity
the coils become an electro magnets in sympathy.

if you are using the same conductor there is no way you should be able to put twice as many turns on one of the coils.
what is the coil thickness and what is the air gap between rotors.
i have seen your alternator photo on your  other post.
the space between the magnet rotors must be smaller than the space between the magnets on each rotor.
other wise the magnetic flux will travail sideways between adjoining magnets rather than vertically between mag rotors.

[stag]

Electrondady.  I see the logic in that the air gap needs to be smaller than the magnet spacing, but on my alternator, the ceramic microwave mags are crammed in and actually touching each other and you have seen from my posts that it works very well. There must be reasons for this. could you maybe explain. Thanks.

[oneirondreamer]

The average gap between magnets (side to side) is greater than .75" with them being around 5/8" at the closest part and greater than an inch at the farthest part.   My testing is with the air gap spacing at .75 inch.   From this I understand that side to side leakage should be minimal.    My 25 wrap coil legs are about .25" thick and .25" wide, my 50 wrap coil leg is about .25" thick, and .5 wide.  Both are wound on the same mandrel, so inside dimensions are the same.   The geometry is such that there should be no point where a leg is being even partly covered by 2 magnetic poles.   In the photo I posted the magnet arrays are still under assembly, that's why they are spaced so far apart.    The coils are held centered in the array, with about .25 inch on either side to the magnets.

If the microwave magnets stag is using are the ring type, a reason minimal gap between magnets may be that the portions of the magnet that are too close together are very small, relative to the total surface area of the magnet.

I hear what electrondady is saying, but I'm not sure I've asked the question right.

In a situation like this, it seems to me that there is an optimal volume of copper that should be used?    Using the larger energy rare earth magnets, that volume seems generally to be as much of the air gap as is possible, there is so much flux available it always can overwhelm any force created by the magnets? The lower flux value of the ceramic magnets may mean that putting too much copper could mean that you are going to get the same watts value but with more resistance than is good?   I get the idea that the emf induced in the coil creates it's own magnetic field which interacts with the field from the magnet.   This again leads me to think that the reason I'm seeing these numbers is that 25 wraps is closer to optimal than 50 wraps?    One problem with that test is that I stopped the test rig to change coils, and while I've got an optical position sensor to give me speed of the motor, it's resolution at this high a sample rate (1000samples / sec) goes a bit wacky and so it's hard for me to get perfectly repeatable speeds.   I should have checked to measure peak to peak distance to ensure speeds were exactly the same, as a small difference in speed could account for some of the discrepancy.   

I'll try to publish some stuff from my larger data set today, where I used 7 different coils, all wound on the same mandrel, and changed them while the machine was running, so speed should have been exactly the same.

[stag]

Oneirondreamer. Good point! My magnets are ring doughnut type.

[electrondady1]

your using . 5 " between the stator and each of the rotors?
that seems huge.

[oneirondreamer]

From magnet face, across to the opposite magnet face, is .75" or less.   The distance from magnet to magnet side to side is .5 at it's minimum, and greater than 1" at it's maximum, so average gap between magnets on a plate is greater than .75".   The stator will be approximately .25" thick, so the gap from the stator to a magnet is .25".    I may reduce the gap, but want to know that at .75" it should reach 90V at 3m/s wind. 

[just-doug]

have you ohmed the coils to see if you have a wire quality issue?

[oztules]

Assuming you are trying this with coils unloaded.
If so,
1. The EMF generated in the coils has no current component, so there will be no back MMF generated by the coils to try to interfere with the magnetizing field.
Until current flows in the coils, the coils cannot act like an electromagnet, as an electromagnetic field is a result of the AMP TURNS... no amps... no field generated.

2. All things being equal, doubling the  N turns in the same space will give twice the EMF unloaded as the N turns was.

When you load this, two things happen.

a. The higher resistance of 2N turns will lower the output voltage via voltage loss= I x R in the coil. So output will be open EMF-(IxR)... and R will be half in 1N turns... But unloaded I=0, so no effect.

b. Particularly with weak fields from the magnetizing source (ceramics with large gap), your back MMF (magneto motive force) will have greater influence in trying to repel the magnetizing force, and will current limit earlier as the amp turns increases... but unloaded this is mute.

3. The copper mass does not absorb any field. It is not magnetic, and as such the field passes straight through it at no load, as if it was not there (ignoring eddy currents). So there is not a secret "mass" of copper that is optimum from this perspective... As above when loaded, your back MMF will come into play, but not unloaded, and thats AMP TURNS .. not copper mass.

4. As I understand it ( may not be completely correct) if we look at a Cathode Ray Tube (Older tv), we use a magnetic field to move an electron beam across the face of the tube which impacts the phosphor, and we see a picture. So we know that a magnetic field can move charge ( electrons in the beam in this case). Knowing this, if we now move a conductor through a magnetic field, we move charges around in the conductor.

In our wire we can use the right hand rule to see which way the charge moves. This displacement of charge results in an electric field produced between where the charge was, and where it is now, so we have a potential... our EMF ( the plus moves one way, the electrons move the other.. depends how you think of it). The moment we stop moving the field, all the charge goes back to a neutral position, and we have no electric field again (EMF) , but just a neutrally charged  conductor. There is no force holding the opposite charges apart anymore.

When we reverse the field, the charges move to relatively opposite positions, and we get a negative EMF whilstever the uniform field is moving past the conductor.... and then collapses to equilibrium again when the field ceases. If we move the same magnetic field  faster, charge is separated to a higher potential, and the resultant EMF is higher, or if the moving field is stronger, we can unbalance the neutral conductor more, with the same velocities. If we have more wires in series inside the field, seeing the same velocities, we can add these for more output... ie more turns....... higher EMF for the same conditions.

At least thats how I look at it, and warmly welcome better explanations..... but it seems to explain it fairly well. I haven't seen a good explanation  as yet. Hopefully someone here will chime in after seeing this drivel.

5. I don't seem to be able to explain your results with the information at hand. I'm missing something, or I don't understand the geometry at play.



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

[oneirondreamer]

My tests of both coils were with a 2 ohm resistor, not open circuited, thanks for your thoughtful reply oztules

[tecker]

The center hole is too big for the magnet. and there some alignment problem with your geometry

[oneirondreamer]

Tecker?  Yes, as I said earlier in the post, the magnets are not all aligned properly for the coil, as I was trying different magnet spacings, but the peak V is all that I was trying to show here.     The centre hole in the coil allows both legs of the coil to be on an opposite pole of the magnet array, for the peak V and the leg width is thin enough that a leg can not be bridging over 2 different magnetic poles.

[tecker]

Ok not a dig I recognize the sharp peaks and the slight pinch on the wave as it rises.