(David Combs) wrote in news:cgh9d2$ecc$1
@reader1.panix.com:

[snip]
I have a dim recollection of something called a "carnot engine"
or something like that, and there's some inherent
max efficiency you can get, and never any more --
some work done to compress the gas is just plain
lost, gone, unrecoverable for useful work.

The perfect heat engine utilizes the Carnot Cycle (you can't do any better
than that for heat engine efficiency):

http://hyperphysics.phy-astr.gsu.edu...mo/carnot.html

Again, I never did really understand this stuff --
maybe someone who does can expand on it, or maybe
even show that I'm just plain wrong.

Here's a page that describes it in fairly simple terms:
http://oak.cats.ohiou.edu/~piccard/p...ot/carnot.html

At the end, it also shows how heat pump efficiency is calculated if the
Carnot cycle is used -- notice that the COP (Coefficient of Performance)
depends strongly on the temperatures involved. Given the right
temperatures, a heat pump is the most efficient way to heat or cool a
space.