Study of Generator Coil Acceleration Under Load (moderated)

[MileHigh]

mh,

could you create such a timing diagram that you mentioned to the best of YOUR knowledge of what is happening in this video? I would love to see your comment in a way that builds constructively.

Fausto.

How can I possibly create a timing diagram when it's not my experiment?  The timing diagram would likely help explain the observed phase shift between the pancake coil and the movable coil.

The closer the movable coil is to the rotating magnet the more flux passes through the coil.  In other words the coupling coefficient between the rotating magnet and the movable coil improves the closer they are together.  Since the movable coil drives a resistive load let's assume that it is purely a resistive load relative to the rotating magnet.  There is a phase shift issue which is not fully explained, but at least we can say that the resistance of the coil wire and the load resistor itself look like a purely resistive load at these low frequencies.  When the movable coil is far away from the rotating magnet then a lot of stray flux leaks out the the sides of the shaft which should look like a reactive load to the rotating magnet.  So as the movable coil moves back and forth there is an interplay between a reactive and resistive load from the perspective of the rotating magnet.

When you see a speed up in the Dremel when you add a load, all that you really have to do is make the measurements and calculate the total resistive power dissipation before and after the load is applied.  Almost certainly, you will find that the total resistive power decreases when the load is added.  You can't forget that when the total load resistance is very low, like this case where the coil resistance is 6 ohms plus the load resistor is 0.1 ohms, there may be an impedance mismatch.  Assuming this is the case, then there is an impedance mismatch with the Dremel on the low side - and therefore the resistive power dissipated in (and transferred to) the load goes down and as a result the Dremel speeds up.

Several people are suggesting "exotic" explanations before the most basic number crunching is done - make measurements and calculate how much total resistive power is being dissipated in the load and relate that back to the observed RPM of the Dremel.  You also have very good power consumption data for the Dremel itself which should not be ignored.

In the final analysis "acceleration under load," really "increased final RPM under load," is most likely because when you "add a load" you actually end up decreasing the mechanical load on the Dremel.  Therefore the Dremel speeds up.  If you can figure this out to your satisfaction you will realize that nothing special is happening and there is no "coil magic" taking place.

[Magluvin]

  You can't forget that when the load is only 0.1 ohms, let's assume that the coil resistance of 6 ohms plus the load resistance of 0.1 ohms results in an impedance mismatch with the Dremel on the low side - and therefore the resistive power dissipated in (and transferred to) the load goes down and as a result the Dremel speeds up.

Luc did say his coil resistance was so low it was not measurable with his equipment. Most meters go to .1 ohm.  I have a coil that looks similar and is .53ohm reading with a meter that goes to .01ohm. His may have larger wire and fewer turns than mine.

Mags

[gotoluc]

Luc did say his coil resistance was so low it was not measurable with his equipment. Most meters go to .1 ohm.  I have a coil that looks similar and is .53ohm reading with a meter that goes to .01ohm. His may have larger wire and fewer turns than mine.

Mags

Yes Mags, the coil I showed in the video (but did not demonstrate) has too low of a resistance to measure.
I'm quite sure mine has finer wire then yours as I used the same gauge (23AWG or 0.6mm) wire, which is same gauge as the coil demonstrated. It's just wound in the most unusual way.
Using a wire resistance chart I was able to calculate the coil to be 0.00135 Ohm.
It has 16uH Inductance (on core)

With an amazingly low magnet rotor frequency of 45Hz and coil position at the furthest position on the rod core, 2.3 in. or 58mm away from magnet rotor, when shorting the coil with its own 10AWG 12 inch long wire leads = 0.002 Ohm load resistor,  the magnet rotor goes to 46Hz and 37mV RMS is maintained across the coil = .685 Watt
Open coil voltage at 45Hz is 106mV

Luc

[Magluvin]

Yes Mags, the coil I showed in the video (but did not demonstrate) has too low of a resistance to measure.
I'm quite sure mine has finer wire then yours as I used the same gauge (23AWG or 0.6mm) wire, which is same gauge as the coil demonstrated. It's just wound in the most unusual way.
Using a wire resistance chart I was able to calculate the coil to be 0.00135 Ohm.
It has 16uH Inductance (on core)

With an amazingly low magnet rotor frequency of 45Hz and coil position at the furthest position on the rod core, 2.3 in. or 58mm away from magnet rotor, when shorting the coil with its own 10AWG 12 inch long wire leads = 0.002 Ohm load resistor,  the magnet rotor goes to 46Hz and 37mV RMS is maintained across the coil = .685 Watt
Open coil voltage at 45Hz is 106mV

Luc

Well I just wanted to make it clear that your coil was no where near 6ohms along with your resistor. Wasnt a clear comparison.

Funny so far with using the magnet to bias the core that Im working with at the moment, there doesnt seem to be much of a change at all in voltage output of the coil but has less drag on the rotor, when it is set up right. This seems to be a positive thing. Now to check while loaded. And readjust, etc.

Mags

[gotoluc]

That's a pretty good experiment.

I did it because I know many cannot build things but they like to see these effects.

Note that the conductive rod core constitutes a shorted 1-turn coil that always brakes the rotor.
It is possible for the slidable multiturn coil to interfere with this braking action.

Yes, a real bad core isn't it!... again, it was quick and dirty, just to show the effect.
I may build a better version using Metglas cores... will see if the time permits.

So for a clean experiment you'd need to use a non-conductive core.
Note that "non-conductive" does not mean "without magnetic hysteresis", which is an unrelated property of a material.

Yes, of course

P.S.
Do you have a non-inductive low ohm resistor to use as an input current sensor instead of the clamp-on current sensor ?

Again, quick and simple. When the real time measurements come the input will most likly be DC which is easy to measure. I should of just used DC since that Dremel is a universal motor.

Thanks for your comments.

Luc