Today, another kind of perpetual motion. The University of Houston's College of Engineering presents this series about the machines that make our civilization run, and the people whose ingenuity created them.

Whenever energy production becomes a problem, people begin inventing perpetual motion machines. I've certainly seen that intensifying during the past few years. What few people realize is, there are two distinct kinds of impossible perpetual motion. 

That's because two physical laws limit energy production. The more familiar one says that energy can't be created from nothing. We create useful energy only by consuming energy in some other form. Suppose we're shown a machine and told that it produces energy without burning fuel, slowing the flow of water, or otherwise using up some natural resource. We needn't bother tracing its operation. We know from the start -- that machine will never work. 

Of course energy can be taken from the environment in subtle ways. Think about those funny toy birds that repeatedly dip into a glass of water. That looks like perpetual motion, but they're actually solar powered. Warm air evaporates the water the bird takes up, and unbalances it for its next dip. On the other hand, many perpetual motion machine inventors say, "gravity is an ever-present force, therefore it will produce energy." Alas that is not so. A thing can fall only once; then it must be hoisted back up. 

Most perpetual motion machine inventors deceive themselves by accidentally hiding an out-of-nothing creation of energy. It becomes hard to find when we trace the details. So the trick is to ignore details and simply ask, Where does the energy come from? If there's no answer, the machine will never run.

The other kind of perpetual motion machine is harder to diagnose. A second physical law says that the world will be run down in some way by the presence of any engine in it. If someone proposes to build an engine that leaves the world less run down than it was, that person fools himself. It cannot be done.

But how do we measure run-down-ness? We actually have a measure called entropy. Entropy is a property of all matter, just like, say, volume or mass. We can't measure it directly, but we can calculate it. When we add the entropy of everything the engine interacts with, the sum has to be on the increase. If it isn't, then the engine, even though it creates no new energy, cannot possibly run. Here's an example:

We heat water to 160 degrees in a solar collector, then use it to drive an engine that's cooled by 60-degree air. We think we can get 20 kilowatts of electricity from every 30 kilowatts we collect, but we fool ourselves. Calculations show that the engine won't produce entropy (therefore cannot run) unless its power output is very small -- less than 4 kilowatts -- no matter how clever our design. 

So be wary. Every few months we read about some new energy scheme. Always there's a catch -- never a magic bullet. Now we find that there's more than one way to fool ourselves when we try to get around the hard walls that nature has erected.

I'm John Lienhard, at the University of Houston, where we're interested in the way inventive minds work.

So many thermodynamics texts, of course I have my favorites. I recommend two classics: J. H. Keenan, Thermodynamics. (New York: John Wiley and Sons, Inc., 1941 and further editions); and W. C. Reynolds and H. C. Perkins, Engineering Thermodynamics. (New York: McGraw-Hill 1977 and further editions.) A fine early physics text speaks cogently of the two kinds of perpetual motion: P. Epstein, Textbook of Thermodynamics. (New York: John Wiley and Sons, Inc. 1937). 

For a contemporary non-technical discussion of these matters, see: J. H. Lienhard, How Invention Begins: Echoes of Old Voices in the Rise of New Machines. (New York: Oxford University Press, 2006): Chapters 4, 5, 6.

 
Sadi Carnot (1824) pointed out that, just as water's ability to do work is degraded after it passes through a water wheel, so too, is the capacity of heat (or caloric in his language) for doing work degraded when it "passes through" a steam engine.