The Inventions of Willi von Unruh and Hans Coler

[F6FLT]

Are you saying that the Lorentz magnetic force F=qVxB when applied to the moving electrons in a conductor carrying current when integrated over the length of the conductor does not does not yield the force derived from Ampere's force law for current?  If so I challenge that.  Ampere's law creates force that does do work.

Ampère's law is not causal. It links two concomitant effects.

[F6FLT]

At the Coler frequencies skin depth is about 0.1mm so surface conduction definitely applies to his RF currents.

My steel plate is about 0.1 mm thick. That seems far too thick for any noticeable effect.

[F6FLT]

Quick test with low-grade aluminium foil (i.e. very thin): still nothing, no difference when I turn the neodymium magnet over, wrapped in plastic film to avoid electrical contact with the aluminium.

[F6FLT]

I get it. The drift speed of electrons is V=J/n.e where J is the current density and n is the number of charges per unit volume.
If I is the current in a conductor of cross-section s, then J = I/s, hence V = I/s.n.e
If d is the thickness of the sample and w its width, then (1) V = I/d.w.n.e

The electric field is E=VxB, i.e. E=V.B when V and B are orthogonal.
The Hall voltage is U = E/w = V.B/w, hence (2) V=U.w/B

Since (1) = (2)   =>   I/d.w.n.e = U.w/B   =>   I/d.n.e = U/B   =>   U = B.I/d.n.e

So we can see that U is inversely proportional to thickness d.

And all this was in fact trivial and obvious from the start. In fact, the number of charges involved is smaller when the volume containing them is smaller. So for the same current I, i.e. for the same number of charges passing per unit of time, if the conductor cross-section is smaller, it means they're moving faster!
For example, a current I of N charges passing at speed V is strictly equivalent to a current I of N/2 charges passing at speed 2.V .

Reducing the cross-sectional area of a conductor for the same current increases the drift speed of the electrons, and therefore the effect of the Lorentz force on each electron. This has no effect on the overall Lorentz force when translated into a mechanical force, since this force on a smaller number of electrons flowing faster is equivalent to that on a larger number of electrons flowing slower. But contrary to our intuition, it does have an effect on the Hall voltage.

[Verpies]

So let's explore ways to decrease the width of the conduction path in non-geometric manner.

[Smudge]

I get it. The drift speed of electrons is V=J/n.e where J is the current density and n is the number of charges per unit volume.
If I is the current in a conductor of cross-section s, then J = I/s, hence V = I/s.n.e
If d is the thickness of the sample and w its width, then (1) V = I/d.w.n.e

The electric field is E=VxB, i.e. E=V.B when V and B are orthogonal.
The Hall voltage is U = E/w = V.B/w, hence (2) V=U.w/B

No, U=E*w.

[F6FLT]

No, U=E*w.

Right 

I rewrite :
(1) V = I/d.w.n.e
The Hall voltage is U = E.w = V.B.w, hence (2) V=U/B.w

Since (1) = (2)   =>   I/d.w.n.e = U/B.w   =>   I/d.n.e = U/B   =>  U = B.I/d.n.e

As the final equation remains correct, the previous conclusion remains valid.
Are we ok now?

[Verpies]

When you look into this that rapid change introduces an electric field vortex that induces a voltage into the curved conduction path whose curvature hasn't changed.

Is this vortex caused by the Magnus Effect from the electron's precession in the flux of the nucleus?

[chief kolbacict]

Reducing the cross-sectional area of a conductor for the same current increases the drift speed of the electrons, and therefore the effect of the Lorentz force on each electron.

Sputtering can be thinner than foil, for example on quartz plates. Especially old quartz resonators contained a very large plate. 
All that remains is to solder the orthogonal leads.   The metal was silver and even gold.

[Smudge]

Right 

I rewrite :
(1) V = I/d.w.n.e
The Hall voltage is U = E.w = V.B.w, hence (2) V=U/B.w

Since (1) = (2)   =>   I/d.w.n.e = U/B.w   =>   I/d.n.e = U/B   =>  U = B.I/d.n.e

As the final equation remains correct, the previous conclusion remains valid.
Are we ok now?

Yes.

[Smudge]

Is this vortex caused by the Magnus Effect from the electron's precession in the flux of the nucleus?

It will take me some time to absorb what that experiment shows.  But the answer to your question is no, in my mind the Hall effect at the atom-electron interaction results in the electron's acceleration along its curved path and that creates the vortex (just as in a coil carrying time-changing current the electron drift velocity that is changing creates the E field vortex that we know as induction).

[F6FLT]

You're going to say that I'm rambling, but the special relativity explanation of induction is considerably simpler, clearer, more elegant and more general than these rather twisted arguments to make induction fit the Magnus effect.
On the one hand, the simple Coulomb force between moving charges, on the other, a complicated and ugly development.

[F6FLT]

Given the weakness of the Hall effect on thick conductors such as Coler's, this effect can be eliminated, or at least not chosen as the starting hypothesis for his contribution.

Incidentally, yesterday I discovered another possibility, that of chaotic magnetic fields, which can generate effects that are difficult to reproduce.
Contrary to what you'd expect, magnetic chaos can occur even in very regular conductor configurations where currents are flowing. For example, two nested coils or turns, or a long conductor surrounded by a slightly regularly undulating circular conductor, can generate a chaotic magnetic field.  See :
https://iopscience.iop.org/article/10.1209/0295-5075/80/60007
or
https://link.springer.com/article/10.1134/S0016793216070161
(downloadable with sci-hub).

These chaotic fields can form "magnetic islands", a common concept in plasma and solar studies, but which also seem to exist in free space. I see these islands as conductor-free zones surrounded by a magnetic flux. They would form in chaotic fields, neither too close to the conductors nor too far away (where the fields are stable and known), but at intermediate distances. The distances between the magnetic elements of the magnetstromapparat seem compatible to me.
Just a thought...

[F6FLT]

Let me clarify the idea. In the papers mentioned, as the currents are assumed to be constant, it's in space that the magnetic field is chaotic, in certain places.

A chaotic zone in the magnetic field means that a small disturbance can cause significant changes in the field.
If a local change is used as feedback, it could therefore give rise to an oscillating system of greater amplitude than the small disturbance that creates the chaos.

This doesn't explain the excess energy - we'll never know without knowing the energy source - but it would explain the mechanism. This is a speculation made about the Coler device, whose configuration of the various magnetic parts/coils could give rise to this instability.

How to experiment with this, I have no idea yet.