Virtually Rotating 4 Overlapping coils

[Jim Mac]

I still aim to conquer the rotating magnetic field virtually (with no physical movement) and I wanted to try this idea for the last year but never got around to building it out..  So I am going to build it now to satisfy my itch...

We know that 2 AC waves 90 degrees out of phase creates rotation..  And I have tried many times to make 2 phases act like an alternator without success..  So here is my next possible theory..

THEORY TO BE TESTED:  Since the 2 phases are essentially "Overlapping" at the half way point,  the coils MUST also overlap at the half way point for smooth rotation to occur.  It may be that when the coils are separated individually we get a mix of partial rotation and partial standard transformer induction that works against itself.

So this is the build.  I will first be using Air Coils for simplicity..  I can upgrade later if warranted..  There will be 4 coils, but each coil will be exactly 180 degrees of the circumference.  So 2 arc coils but against each other to form a full circle.  Then an identical full circle from 2 other coils is offset exactly 90 degrees.  The cross coils are wired in series making 2 circuits.  Then these 2 circuits fed with 2 separate AC phases 90 degrees apart.

You can see in this image.  There is a RED, PURPLE, BLUE, and GREEN coil.  

   

This setup will allow me to test 2 different setups.  1 will be powering the coils as stated above and inducing into output coils.  The 2nd test after that will be rotating a magnetic field from a rotor over it like an axial flux generator. 

   

[Jim Mac]

After a few hours fighting the form and wires, it's ready to test..

   

24 awg.   188 turns per coil.  2 Coil's per phase.  I should be able to hook the phases in parallel or series and still be within working limit of the drivers.  So I have options.

I haven't measures resistance yet, but I predict somewhere between 7-9 ohms per coil.   With no core in the stator, the magnetic field will not be optimally strong, but I can use cores on the output coils (maybe)..  

I plan on testing it tomorrow after work, so I will update in the evening hopefully

[Jim Mac]

I couldn't wait till tomorrow to verify rotation...  

If this isn't proof of rotation then I don't know what is!



 

One piece of info I saw I can share now.  Unlike a standard coil,  the quicker the frequency, the more power it draws, even without load..  This seems opposite from a standard coil..  And all signs point to increased current and voltage with frequency increase.

Now If I can manage to wind this configuration inside a core-  we may get some legit results..

[unimmortal]

The frequency correlation aligns with the Don Smith solid state generator. Explains why he was using NST to get up around the 35khz mark.

Great work on the wave man, you've been looking for this solution for a while now.

Edit: so this wiring method removes the bloch wall, or rather puts N/S poles from adjacent coils over the bloch wall of each coil.

[Jim Mac]

I am convinced this is the technique that must be used to get alternator-like rotation from 2 phases..  The number of poles must correspond to the number of pickup coils.  We can use any even number of poles..

Here is an example with 4 poles..  We take 4 coils, N-S-N-S in a circle.  Make a 2nd group the exact same and shift it 90 degrees.  So it's like this:

   

Each group of 4 coils is powered by a separate phase.  So each group is shifted 90 degrees AND the phases are shifted by 90 degrees.  

The groups can overlap on a single plane like my original build shown above,  OR I think they can be separated..  If the 2 groups are separated, the output coils sandwich between them, then there is no need for complicated overlapping, AND this way allows you to easily use cores.  I think it can still work with the overlap method which would be necessary if using a stator from an induction motor or the likes.

See this image for simplicity.  But imaging these red and blue coils were in a circle pattern connecting..

   

If we power each row with it's own AC phase, shifted 90 degrees from each other, this is the rotation we seek.  Notice how the bottom middle coils are positioned exactly between the bottom coils.  Mutual induction back to the bottom coils can not raise input because we have 2 different coil directions equally affected by the output coil.  Now when looking at the phase of the output compared with the coil that is directly aligned with it, you will see the output is exactly 90 degrees out of phase with the coil aligned Center to Center..  So again, mutual induction is also NOT able to raise the input of the top groups input either!

So my assessment is,  The secret to 2 phase 90 degrees is that the 2 coil sets need to also be set at 90 degrees.  

My build gave up this data,  but I see some errors I made and many improvements which need be done..  It was a mistake making 4 large lanky coils..  We want them symmetrical and wound tight..  We Do Not want 4 air coils with HUGE middle sections..  We want many many tight poles so we can incorporate many output coils.   And obviously we need cores to concentrate that flux and not let it go to waste..

I am planning my upgrade build from what I learned with this one...  More to come!

[unimmortal]

Following along very keenly Jim. Although your last image has me slightly confused. Are we virtually rotating top and bottom coils, and the coils in the middle are the pickups? What is the orientation of the pickup coils with respect to the vr coils? all axial, all radial?

If you can work bucking coils into the output section, that'd be interesting.

[Jim Mac]

Yeah, that last drawing is confusing..  You can disregard it..  It's difficult to draw out when using different phases..  

The air coil test was inconclusive.  I stated that it drew more power at higher frequency, but then when I repeated the experiment, it didn't act the same.  The rotating field is very weak from the large air coils..

So I needed to add cores.  This next test-build is using 4 individual coils and utilizing a different wiring scheme.  I wanted to see if it would eliminate Back-EMF..  But it didn't...    Without full output coils and encasing the flux, I am being hindered by this pesky Back-EMF which is making the input current decrease with higher frequency.  

Next I will try the same wiring with 4 transformers..  I want to see if harvesting all flux with none escaping will make a difference..

So here is the latest demo..



I also want to try this same wiring technique on a universal motor rotor..  All in all, it's difficult to argue it's not rotating pretty smoothly.

[Jim Mac]

I "BELIEVE" I have the whole "Back-EMF" problem solved..  In the process, that magical extra current in the ground is appearing again..

Basically, to avoid Back-EMF the current going into the coils has to be travelling in 1 direction only.  As soon as we start flipping polarities, Back-EMF creeps in..

So now I have 4 isolated phases of Full AC with bare minimum Back-EMF when raising frequency..  We will see where it leads..

[Jim Mac]

As promised in the past...  Here is a video of the audio amplifiers and the circuit operating with NO GROUND going to the Audio drivers...

Also notice, ALL 4 speaker outputs are going through diodes, so NO CURRENT can return to the amplifier leads.  

+ as an added bonus-  Watch when I measure the voltage on the audio driver inputs...

 

So answer me this:  Since the audio driver outputs are ALL going through schottky diodes (only allowing power to Leave)  How the hell are they able to operate with no wire connecting them to the negative of the source supply?

[Tom1956]

Now that you have the circuit working.
Can you please post a circuit diagram, it will make your system so much easier to understand.
Thanks.. Tom