Interesting experiment with an transformer, 2 lamps, diodes and an magnet

[MeggerMan]

Femm seems to lock up if you put two materials too close together, arcs next to arcs seem to cause me no end of issues.
Check the matrix is not creating too many triangles in a small space.
I think there are some issues with two surfaces very close to each other and it gives some very odd results for the torque.  If you looks at my latest simulation the torque change from 0.5 to 5 n/M

Regards

Rob

[PaulLowrance]

It tried spacing objects and made sure there we not very many nodes (few hundred), but haven't tried using the arcs. You mentioned torque in FEMM. Is that something I could change that might help?

Thanks,
Paul Lowrance

[Jdo300]

Hi Paul,

I use FEMME all the time to do permanent magnet motor simulations and coils setups. If you post a copy of the .fem file, I'll be glad to take a look at it to see why you are having problems.

God Bless,
Jason O

[GM]

Hello everybody,

now I have done the measurements on my circuit and worked out the results.

I made two measurements because I used two kinds of coils.
Reason: In contrast to my previous mention the three layered coil around
the torodial core *does* show an effect, but in another way.

Where the 1-layered coil shows increasing kickback power and simultaneous
decreasing overall power consumption, the 3-layered coil shows nearly no
increase of kickback power but much more of decreasing of the overall power
consumption!

The first picture shows the measurements with the 3-layered coil.

All voltages where measures with the scopes RMS(eff) function which
read the RMS effectiv voltage for one signalperiode.

The voltages shown on the scopes screenshot have to be multiplied by
factor 10, because of the :10 dividide-factor of the scopes probes.


Picture 1, three-layered coil:
             Input power     Output power
w/o magnet   4.16 W          0.65 W
w magnet     1.52 W          0.64 W


The second picture shows the measurements with the 1-layered coil.

Picture 2, one-layered coil:
             Input power     Output power
w/o magnet   5.7024 W        0.41648 W
w magnet     3.016 W         0.79968 W


Worse luck! If the method of measurement and calculation is the right way,
then we have no over unity here (as I expected).
The values are showing a bad efficiency. :-/

Hmmm, subjectively I would say, the effect was more conspicuous with a
connected light bulb at the secondary coil using @ 15.8 KHz (as shown in my
previous experiements). This is a bit wired. I have to test more frequencies,
duty cycles and configurations.

In the animation you'll see a high scaled shot from the kickback-spike of
measurement-arrangement no. 2 (1-layered coil).
In the first frame you see current and voltage across the light bulb without
the magnet. In frame 2 with the magnet inside the core and in frame 3 I turned
the magnet by 180 degrees (which caused I higher overall power consumption
of the whole circuit than without magnet!)


What shall I say as conclusion? I think this arrangement doesn't have the potential
to be a over-unity device.

Regards, Markus

[Jdo300]

Hello GM and all,

Here is an idea for you. Take the power from the back emf spikes and store it into a capacitor. Use the capacitor to feed a resonant tank circuit and use the output from the tank circuit to pulse feed power back into the primary side of the coil on your transformer. Assuming that you can capture the back emf spikes for 'free' without loading the source circuit. Just my two cents :-). I will still try out this circuit to see what kind of results I get.

God Bless,
Jason O