Understanding Eddy Currents

[kitestrings]

I'll admit I find eddy currents both fascinating and mysterious.  When I recently removed stator coils from our alternator, I felt a notable drag as I slid each coil out of their respective positions.  Maybe it was just the mechanical resistance, or even minor vacuum as the tongue and groove separated?  I know that eddy currents are generally resisting change in the magnetic field(s), but does this make sense, or did I imagine it?


In a youtube video on the subject I saw a demonstration where a magnet is dropped thru a plastic tube.  A coil of CU wire is wrapped around the tube.  With the circuit open there is no resistance.  With the circuit closed it is slowed.  I assume this is akin to the cogging/braking action we get when we short or open the alternator circuit on an axial.  But, I had thought eddy was independent of an electrical path - like dropping the same magnet thru a CU or AL tube - obviously my individual coils are open.  I'm confused.

[JW]

But, I had thought eddy was independent of an electrical path - like dropping the same magnet thru a CU or AL tube

There was a stator  that was made by DanB that was made with thick copper rectangular wire" We expected to make lots of amps. Believe it or not. even with an open circuit the winding got really hot. And the stator failed in short order.

There was this experiment that I have seen that helps to understand to Eddy currents

Get a strong neo magnet and set up a downward slope ramp/slant made of wood it will fall/slide faster. Then get a aluminum the same as the wood slope and you will see that the magnet slides down at slower speed.

This experiment is clearly known about, and is a good way to understand eddy currents the best way to explain this the use of "wire in hand" use two or three small dia wire.

So its better to use smaller wire wired in parallel to the coils to minimize eddy currents.   

JW 

     

[kitestrings]

There was this experiment that I have seen that helps to understand to Eddy currents

Get a strong neo magnet and set up a downward slope ramp/slant made of wood it will fall/slide faster. Then get a aluminum the same as the wood slope and you will see that the magnet slides down at slower speed.

Yes, I've had fun with that...better still if you drop a cylinder thru a copper or aluminum pipe.  It seems to defy gravity.

After seeing the plastic tube/coil experiment though, I assumed with open coils, there would be no drag.

[kitestrings]

You never outgrow the fascination with this stuff.  Magic.

https://www.youtube.com/watch?v=sENgdSF8ppA

[mab]

After seeing the plastic tube/coil experiment though, I assumed with open coils, there would be no drag.

I think it's a question of how thick is the wire in your coils? If they're wound with thick wire you can get eddy currents within the diammeter of the wire - hence the resistance you feel even with the coil open circuit.

As JW says it may be better wound say, 4 in hand with wire 1/4 the cross section.

[Adriaan Kragten]

In my recent reports KD 678 and 679, I describe a simple 8-pole axial flux PM-generator which is used for the VIRYA-1 and the VIRYA-1.02. This generator has a 3 mm thick steel stator disk but no iron in the coils. As the stator sheet is made of iron, you get eddy currents in this sheet and this results in a certain sticking torque which increases at increasing rotational speed. As there is no iron in the coils, the sticking tor isn't fluctuating and therefore the generator has no preference positions. The rotor has therefore a low starting wind speed.

The unloaded sticking torque has been measured and the Q-n curve is given in figure 7 of KD 579. The Q-n curve is about a straight line which starts with a very small torque of 0.02 Nm which is caused by the bearing friction. If this bearing friction is neglected, it can be seen that the sticking torque due to eddy currents increases about proportional to the rotational speed.

The maximum rotational speed with the VIRYA-1 rotor is about 575 rpm. The sticking torque is about 0.105 Nm at this rotational speed and it can be calculated that this torque results in a power loss of about 6.3 W (see chapter 7 KD 678). This power is rather low if compared to the generated mechanical rotor power which is about 60 W. The rise of the stator temperature was only about 6 degrees at this rotational speed. So it is allowed to use a steel non laminated stator sheet for this generator.

[SparWeb]

After building my (one and only) axial alternator I tried this test:



The alternator magnet disks were spun on the lathe at various speeds, with no stator in the way.
Within the gap, I inserted different bars of aluminum.  I inserted the bar as far as I needed to get the full effect, basically deep enough that the field from magnet to magnet was passing through the whole bar.

Test 1:
Solid bar, 3/4" square.  This had the strongest effect.  It was difficult to hold the bar straight and not scrape the sides when spinning at 700 RPM. 
The difficulty was not due to attraction, just effort to keep the bar steady with such a strong force pushing against it.  The bar was being pulled with the rotation of the magnet disks.
The bar also started to get hot in my hands.


Test 2:
Bar made of 3 laminations of 1/4" thick pieces, together making a 3/4" square bar.  This I also pushed into the gap of the mag rotors up to 700 RPM.
When the laminations were oriented face-to-face with the magnet faces, the force was almost as strong as when the solid bar was in the alternator.
It also started to heat up.


Test 3:
The same bar, but turned 90 degrees.  The 1/4" pieces were edge-on to the faces of the magnets.  When that was passed between the magnet rotors, the resistance was DRAMATICALLY reduced.
I was able to run the lathe much faster (1000 RPM) and still held it easily.  The resistance wasn't zero, but still easy to hold, with no sign of heating.


Note that I was measuring the RPM of the lathe, not just going by the rated gear ratio, so during the solid bar test the resistance held nominal 770 RPM the speed down to 730, while the edge-on lams it ran at 760 RPM.  The same happened at lower speeds, but the handling forces and slowing effect all grew stronger at higher speeds - more than just in proportion to the speed.

[Adriaan Kragten]

These tests are in accordance with the orientation of the stator stamping of an asynchronous motor. A stator stamping consists of a package of thin iron sheets which are isolated electrically from each other. The magnetic flux flows in the direction of the sheets. The isolation in between the sheets prevents that eddy currents are flowing perpendicular to the sheets from one sheet to another. So the current can only flow in one sheet and is therefore very low as the sheet thickness is only about 0.5 mm. The eddy currents create its own magnetic field which is working in the opposite direction of the magnetic field of the magnets. Eddy currents therefore result in a sticking torque.

So a PM-generator made from an asynchronous motor has a low sticking torque especially if the winding is rectified in star. The sticking torque for delta rectification is higher as higher harmonic currents can flow around in the copper winding for delta rectification. So this increase of the sticking torque for delta rectification is not caused by stronger eddy currents in the stator stamping. Measured values for the uloaded sticking torque for such a PM-generator are given in my report KD 78 for star and for delta rectification.

[kitestrings]

Thanks, this was very helpful.  I was afraid I'd lost my mind...and not that this alone disproves that fact , but it helped me to understand what was happening.

And yes, Mab, our coils are fairly thick, 2 - #14 IIRC.