[extropy-chat] Casimir Torque Project

[Adrian Tymes]

--- Hal Finney <hal at finney.org> wrote:
> This is consistent with the concept of a conservative force.  The
> force
> results from a change in the potential energy of the system.  In a
> conventional parallel-plates Casimir force test, the ZPE is lower in
> the region between the plates, and so moving the plates together is
> energetically favorable.  This is what produces the force.

No it isn't.  Just because and end result is energetically favorable
does not automatically, in and of itself, cause it to happen.

Consider a boulder on top of a hill.  If its path is not blocked, it
will roll down the hill because that would be energetically favorable
to gravity.

Now consider a boulder a meter or so away from a cliff.  Its path is
otherwise not blocked, and it would be energetically favorable if the
boulder fell off the cliff.  And yet it refuses to jump sideways a
meter or so to allow that fall to happen.

Likewise here: to see what happens, you have to consider the vectors
of force, not merely the potentials.

> In fact, despite your attempt at geometric arrangement, the pressure
> on the outer ring will be entirely central (directed towards the
> center
> of the ring).  The same thing would happen if you considered the ZPE
> as
> gas pressure.  Imagine having high pressure gas outside the system
> and
> low pressure gas in the cavity (white region).  Despite the
> asymmetrical
> shape of the cavity, the pressure on the ring will be perpendicular
> to it.
> The gas pressure will not act to rotate the outer ring.

This analogy fails too.  Gas particles are more or less uniform in
size, and usually far smaller than any cavity or pressure vessel they
act upon.  However, the virtual particles that cause the Casimir effect
are of wavelengths that are significant fractions (or more) of the
width of the gap they act upon.  There are only so many ways the longer
wavelength ones can fit inside - and they all lean in the same
direction with respect to the perpendicular.

> I would not be afraid of some equations if you want to offer them.

That, I will admit, has been a problem.  Almost all of the analysis
I've seen of the Casimir effect has been on the parallel plate model;
there appears to be some disagreement on the magnitude of the effect in
different geometries (like the one I'm using).  So I'm not certain
that, even if it works, it would produce significant power.
(Microwatts per cubic centimeter, considering a vast array of these
devices, might be too low density to be of practical use.)  The proof
so far is just that there's torque at all in one direction, and not in
the other, ergo there's net torque in one direction.  (It's possible
that the torque would be less than what is needed to overcome friction,
which would also cause the device to not work.)

> I am also curious to know the qualifications of the academics who
> have
> endorsed your design, in broad terms - you don't have to name any
> names
> or embarrass anyone.

Ph.D.s in physics or quantum physics at various universities.  A few
who have studied the Casimir effect in particular (say, for using it to
store energy - a kind of nanoscale spring or capacitor).

As it happens, my next session in the lab to work on the device is
about to start, so I'll have to cut this email off here.