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: [Phys-l] entropy and electric motors



On 05/23/2008 07:04 AM, kyle forinash wrote:

This is helpful.

:-)

I guess I need to read the Nyquist paper (and I will)
to see what the connection between Johnson noise and the second law.

I recommend it. It exemplifies the sort of physics I like best,
i.e. simple yet fundamental physics with direct application to
real-world problems.

So would it be fair to think of the thermal noise as roughly parallel to
the Qout of a heat engine?

I dunno. There may be some connection there, but if so, I'm not
seeing it.

The disconnect comes from the following:
*) Heat energy, by definition, means your heat bath is at equilibrium.
That means it is in the maximum entropy macrostate, i.e. maximum
entropy per unit energy.
*) The electrical energy coming from (say) a battery is nowhere near
full thermal equilibrium. The distribution of probability versus
energy level does not follow the Boltzmann distribution. More
precisely, it follows the Boltzmann distribution *except* for
one *enormous* exception. For details, including an explanatory
diagram, see
http://www.av8n.com/physics/thermo-laws.htm#sec-metastable-t

The electric motor is not a heat engine, and ideas (such as Qout) that
apply to heat engines don't necessarily apply to electric motors.

This illustrates the point that the foundations of thermodynamics are
energy and entropy. Non-experts think it's about heat and temperature,
but it's not.
++ Energy is primary and fundamental.
++ Entropy is primary and fundamental.
-- Energy and entropy are well behaved even in situations where
the temperature is zero, unknown, irrelevant, and/or undefinable.
-- Often it is unnecessary or impossible to quantify "heat". It is
more practical to quantify energy and entropy instead.

To repeat: Anything you can do with "heat" you can do more easily
and more precisely using energy and entropy instead.


Do you happen to know where the loss comes from in real electric engines?

Mainly from "I^2 R" loss in the windings.

Engineering is generally a multivariate optimization process, with
lots of tradeoffs between the various variables. If all you cared
about was efficiency, you would use more wires and fatter wires in
the windings ... but you also want to optimize the power-to-weight
ratio and the power-to-capital-cost ratio, which argues for fewer
and thinner wires.

Keep in mind that this I^2 R loss is a nuisance loss, not required
by thermodynamics. It is like friction in your heat engine, which
is not required by thermodynamics. It causes an inefficiency in
_addition_ to whatever inefficiency is required by Carnot.