Since the 11th century, many people have tried to beat the First Law with ingenious machines. There are good reasons for trying: if you could build a machine that could work forever without needing energy, that would solve the world’s energy problem in one stroke. You would gain immeasurable wealth, fame, and surely the Nobel Prize as well. Remember the tremendous excitement that emerged when Pons, et al. wrote about “cold fusion” in 1980? And that did not even involve perpetual energy production, but only a claim that nuclear fusion could proceed at room temperature instead of 5000oC – the temperature on the surface of the sun!
There are basically two different aims for perpetual devices: to achieve perpetual motion, and to generate work. A perpetual motion machine is typically not very useful other than its allure as a kind of magic show that can attract big crowds and so generate income from admission fees. In contrast, perpetual motion engines claim to generate work. (We call them perpetual motion engines because even perpetual motion requires that a certain amount of work be generated to overcome the friction forces, however small, that are present in all engines.) The hundreds of proposals made over the years for perpetual motion engines use many different forces to keep the movement going – including purely mechanical forces (gravity, expanding fluids, or springs) as well as magnetic, electrical, or buoyancy forces.
As said above, the First and Second Laws are postulates. This means no proof is possible, but it is merely based on many observations. Thus, in principle it could happen that tomorrow somebody builds an ingenious machine that can produce “free” energy. This would obviously be an enormous blow for the thermodynamic theory, but a blessing for humankind. Many such claims have been made, and many machines have been built either by people who intentionally produced frauds, or who were very serious about the matter and saw a mission to provide humanity with a useful tool. Also, many designs were made but never translated into real machines, and they are sometimes very difficult to prove wrong before they are actually built. Detailed mechanical analysis is often required before the flaw in the design can be found. The author knows no verified perpetual motion engine or machine has ever been built. Nevertheless, it’s fun to look at some of these concepts.
Before we do, though, it’s good to know that there are two kinds of perpetual motion engines, based on which of the two laws is being violated. Engines of the first kind typically claim that they can generate more energy (in the form of work) then the amount of energy that was put in, which clearly violates the principle of conservation of energy. Engines of the second kind are a bit trickier to describe. They try to convert heat into work without implementing any other change (achieving 100% efficiency), or purport to let heat flow from cold to warm, or attempt to convert heat into work without using two heat reservoirs at different temperatures. Simply put, second-kind perpetual motion engines draw energy from a heat reservoir and convert this heat into work without doing anything else.
Many perpetual motion engines of the first kind use the classic design of “overbalanced wheels.” An early example comes from the Indian mathematician and astronomer Bhaskara, whose design incorporates tubes filled with mercury. In the figure below, we see the operating principle of Bhaskara’s idea. He claimed that the wheel would continue to rotate with great power, because the mercury in the tubes is not at the same distance from the axis at opposite sides of the wheel. Bhaskara probably never built a real device, but similar ideas later were incorporated into the designs of other inventors’ engines, none of which ever worked. Fraudulent designs for perpetual-motion machines even made it to actual patents, which were later challenged in courtrooms. A famous example of a fraud was that made by Charles Redheffer in 1812 in Philadelphia. He claimed to have invented a work-generating perpetual motion engine, which seemed convincing until it was discovered that a man in an adjacent room was powering it.
Several famous names are connected to the idea of perpetual motion engines. Leonardo Da Vinci designed and built many devices and machines, including two devices to study the workings of perpetual motion. In his time, the principle of the conservation of energy was not known, but Leonardo had good insight into the working of machines and did not believe one could construct a perpetual engine. Simon Stevin, a Flemish scientist who lived from 1548 to 1620, actually showed that a purported perpetual engine based on a chain looped over a pair of asymmetric ramps would indeed not move without the addition of external energy.
An example of a perpetual motion engine of the second kind was provided by John Gamgee with his invention of the “Zeromotor” in 1880. His idea was to draw heat from the environment to let liquid ammonia boil; the ammonia vapor would expand and drive a piston. Afterward, the vapor was expected to cool down and condense, allowing the process to start again. Gamgee proposed this idea to the American Navy as an alternative to its coal-fueled steamships. The problem, however, was that ammonia at atmospheric pressure condenses only at temperatures lower than -33°C, and that temperature was not present in Gamgee’s system. Thus, we see here a violation of the Second Law: if you want to draw work from heat, you must have two different heat reservoirs, one at a high temperature, and the other at a low temperature.
Perpetual engine after a design of Bhaskara.
© 2007 William Andrew Publishing. Reproduced with permission from the publisher William Andrew Publishing
 Many patents can be found that claim to have invented a perpetual engine (for instance a patent for perpetual movement by Alexander Hirschberg in 1889, patent number GB 7421/1889). That these patents were granted was because in Great Britain patents filed before 1905 were not checked for whether the claims were realistic. This is unlike patents in the US where within a year a working prototype was required [van Dulken, 2000]. The American Navy was wrestling with the fact that their steamships were too limited in their routing because they could not get coal everywhere. Thus the Zeromotor was seen as a solution to this problem. The invention was even shown to President Garfield who was very positive about this approach.