Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: Non-conservative forces in a liquid dielectric

Ludwik Kowalski wrote:

On Saturday, May 31, 2003, at 04:22 US/Eastern, Pentcho Valev wrote:

Ludwik Kowalski wrote:

On Friday, May 30, 2003 Pentcho Valev wrote:

This is just one of the numerous myths in thermodynamics,
Ludwik. When a gas pushes a piston, isothermally, it converts
heat into work without creating disorder elsewhere.

Please keep in mind that thermodynamics is not my specialty.
You are talking about an engine, a cyclic device delivering net
work. I know how to calculate the p*dv work, done on an agent,
when an ideal gas is allowed to expand very slowly, at constant
temperature. The same work must be done by the agent, to
compress the gas slowly to the initial state at constant
temperature. The net work in each cycle is zero.

Yes, your constant-temperature electrostatic engine
(presumably performing net work in each cycle) works
differently. You allow the plates to come closer to each other
in a vacuum, (work done on the agent) and you separate
the plates in a liquid dielectric, such as pure water. You argue
that work done by the agent (in separating plates inside the
liquid dielectric) is smaller than work done on the agent,
during each constant-temperature cycle.

Is this a correct description of your proposal? In my mind you
are describing a perpetual motion machine of the second
kind. Please show me what is wrong with this "accusation."

Nothing, except that the perpetual motion machine of the second kind is
just one of the two possible solutions to the problem. You refer to two
steps of the four step cycle - when the plates come closer in vacuum
and
then when they are separated in the liquid dielectric (steps 1 and 3).
The net work extracted from THESE TWO steps is positive - there can be
no doubt about that. The reason is that the attraction between the
plates in the dielectric is experimentally shown to be lower than the
attraction in vacuum. But we should take into account the other two
steps - when the capacitor is immersed in the dielectric and then when
it is withdrawn (steps 2 and 4). It may happen that we SPEND net work
in
these two steps, and the work spent counterbalances the net work gained
from steps 1 and 3. Then there is no perpetual motion machine of the
second kind. However there is a second possibility. As we let the
capacitor down and immerse it into the pool (step 2), the capacitor,
through a pulley, lifts some (maximum) weight. Then, as we withdraw the
capacitor from the pool (step 4), the same maximum weight is used for
withdrawal. If so, we do not spend any net work in steps 2 and 4 and
the
net work gained as the plates get closer in vacuum and are separated in
the dielectric (steps 1 and 3) becomes NET WORK EXTRACTED FROM THE
CYCLE. This can of course be called perpetual motion machine of the
second kind.
As far as I know the rules are as follows. We must accept
data coming from real reproducible experiments, no matter
how many theories they disagree with. But gedanken
experiments are different in that respect; we accept their
conclusions only when these conclusions do not conflict
with accepted theories. Do you agree with these rules?

It depends on what you mean by "theory". Is the statement "perpetual
motion machine of the second kind is impossible" the result of a
theory?
I am afraid it is not even an experimentally tested proposition.
Rather,
it has been advanced to show that those who attempt to test it are just
as mad as those who try to extract energy out of nothing.
In my view, gedanken experiments should not differ essentially
from
real experiments - there can only be inessential reasons for not
performing them in practice. In the present case, the easiest thing to
do is to partially immerse the capacitor in the pool, punch a hole in
one of the plates, near the pool's surface, and see if water would leak
through the hole. That is extremely easy but I am unable to do any
experiment - I hardly survive. I have been asking physicists to do this
experiment for several years - no effect. This shows how efficient the
curse "perpetuum mobile of the second kind" is.

1) Please describe forces involved in steps 2 and 4. This
should allow us to speculate about the net work per cycle.

In step 2, the distance between the plates is SMALL and we slowly let them
down and immerse them into the pool. The oversimplified picture between the
plates can be presented like this:

P- (+)(-) (+)(-) (+)(-) +P

where the dipoles are those from the surface of the rising water. Many forces
are acting but the one of interest is capillary - it is a force of attraction
between the (+) pole of the dipole on the left and the part of the left plate
that is still not immersed. This force pulls the capacitor downwards and is
mentioned in textbooks - this is the only force that could be different in
steps 2 and 4.
In step 4 the distance between the plates is LARGE and we slowly
withdraw the capacitor from the pool. Now the picture between the plates is

P- (+)(-)..........................................(+)(-) +P

The same capillary force pulls the capacitor downwards but it seems to me
that, for the same height of the capacitor with respect to the ground, this
force is smaller in step 4 than in step 2. At least there is no reason to
believe that it is greater in step 4 than in step 2. This could easily be
tested by simply weighing the capacitor as it is partially immersed, at a
fixed height with respect to the ground, for two distances - small and large
- between the plates. If the weighing shows that the downward force in step 4
is not greater than in step 2, we have a perpetual motion machine of the
second kind. It is another matter that this machine is useless.




2) Also describe the geometry of the "plates with holes"
needed to test your prediction. Somebody on this list might
be able to perform the experiment you want.

I am afraid this is difficult - preliminary experiments are needed. But I can
present once more the message from the UCSD colleague giving some
information. I suggest repeating his experiment without cutting slots -
rather, a small hole (1 mm in diameter) should be punched in one of the
plates, near the pool's surface. Here is his message:

---------------------------------------------------------

Pentcho,

I had a factor of 4 pi wrong in my calculation of the field.
As a consequence, the effect is readily observable.
A colleague who is an experimental physicist put together a capacitor
made of 1 in square aluminum plates spaced 1.5 mm apart. Water rose
about halfway up the plates at this separation due to capillary action.
When a potential of 500 V was applied across the capacitor (E = 3300 V/cm)
the water rose by about .4cm -- quite visible.

We're going to cut a slot in one of the plates to make the rise easier
to see, so I'll be able to report to you on whether we get a perpetual
motion machine. I'm betting we don't.

(Later)
Unfortunately, my colleague cut a fairly wide slot in BOTH plates. The
result is that the water rises on both sides of the slot, but not as much
in the slot. Still, NO tendency for the water to fall out of the capacitor
into the slot, creating a waterfall.

I'm hoping to convince him to make another with only one plate slotted with
a narrower slot.

In any case, this is a real and observable effect, and shows no signs of
producing perpetual motion or a violation of the second law -- no surprise to

me, and I suspect not to you either.

End of UCSD colleague's message------------------------------

Pentcho