[extropy-chat] Casimir Torque Project

[Adrian Tymes]

--- Hal Finney <hal at finney.org> wrote:
> As I understand it, this device consists of a disk that has a metal
> square in the center and a metal ring around the outside.  Between
> the
> disk and the ring are specially shaped black body structures which
> block
> the Casimir force.  The whole thing is one piece and rigid.

Incorrect: the ring is mechanically separate from the rest of the
device.  The black body structures and the metal square in the center
are rigid with respect to each other, though.

> Your
> design
> is intended to produce a net unbalanced torque force, such that if
> this
> device were suspended in a vacuum in empty space, it would
> spontaneously
> begin spinning, superficially at least violating conservation of
> angular
> momentum and conservation of energy principles.  Is that right?

That is correct, if by "superficially" you mean "from a point of view
that fails to recognize the actual source of energy" - which is an easy
oversight to make if one is not familiar with quantum mechanics.

> The problem that I see with your design is elementary.  You try to
> arrange the black body shields so that the unblocked places on the
> rim
> are attracted towards the center square with a net torque.  But you
> have
> forgotten Newton's third law!  Every action has an equal and opposite
> reaction.  This means that if the rim is attracted to the center
> square,
> then the center square is equally attracted back to the rim.  And the
> torque due to the center square is exactly the opposite of the torque
> due to the rim.  These will cancel each other out and it won't spin.

Actually, by that analysis, the post would spin the opposite way from
the ring.  Total angular momentum of the entire system would continue
to be zero, even if the pieces themselves move with respect to one
another: plus and minus cancel out.  (Consider two interlocked gears,
flat on a table.  Turn one gear one way, and the other gear turns the
other way.  There's an issue with where you get the energy to spin the
gears in the first place, but the fact of counterbalancing torque to
conserve angular momentum does not itself freeze the system in place.)

The post and shields should indeed experience some torque.  While these
will be fixed to a large external mass in practice, it is possible that
a sufficiently large array of these would cause that large external
mass to rotate.  But since this would be in the opposite direction of
spin from the rings, the rings would still spin relative to the posts.
Any generator assembly would probably be fixed relative to a post, and
thus be able to extract energy from the ring's relative spin.

This does not appear to be the core issue of your objection, though.

> Consider a large model of your device, which instead of using the
> Casimir
> force, uses rubber bands.  Stretch rubber bands between each
> unblocked
> point on the rim and each unblocked point on the central square.
> Your analysis still goes through!

No it doesn't.  When the point of the band that's on the ring travels
beyond the blackbody shield, the band would wrap around the shield and
produce a counterbalancing force.  As I stated in another post, the
Casimir force doesn't act like that.  Magnetic force does, and that may
be what you're trying to draw an analogy to.  But this is closer to
optics, where the lines do not warp unless there is some medium change
making them warp - but the curved path the rubber band would occupy to
get around the shield, is occupied only by empty space.

To analogize to the rubber bands, imagine if they could stretch but not
bend - and vanished if they were forced to bend.  In fact, try to
imagine rubber bands made of photons bouncing back and forth:
obviously, such a band that is "cut" will disappear as all of its
photons are absorbed by the cutting material.  (In this case, the
blackbody shields intersect and thus "cut" the band.)  Now imagine that
new rubber bands appear, pre-stretched, between the ring and the post
as new sections of the ring are uncovered with respect to the post.

Actual rubber bands don't act like that, of course.  But the force I'm
tapping doesn't act like actual rubber bands.  (At least, by current
academic understanding it doesn't.  If my experiment proves that it
does act that way - which would indeed cause the device not to produce
energy - then that's a significnat revision to our understanding of
quantum mechanics.  Which isn't to say that it won't happen, just that
the experiment would be worth doing even in that case.  Finding a way
to make light bend in free space, without any refractive material at
the actual point of the bend, has its own applications.)