Why are perpetual motion machines impossible?



Why are perpetual motion machines impossible?

In: Physics

There are two core problems with perpetual motion.

First, any interaction will slow the device down. Friction, pressure, or simply just trying to extract energy from the device will slow it. Energy **must** be conserved, and cannot be created or destroyed, only transformed between various forms. This is a non-negotiable aspect of physics called the 1st Law of Thermodynamics.

Second, all objects inherently lose energy to their surroundings over time, even without any interactions. Energy is, gradually, lost to a process known as Entropy Generation, and as a result while the energy is not destroyed, it is transformed into something *permanently* useless. This is *also* a non-negotiable aspect of physics, and is called the 2nd Law of Thermodynamics.

Because of the 1st Law; perpetual motion devices, even if they could exist, cannot be used as a perpetual power source.

Because of the 2nd Law; perpetual motion devices straight up cannot exist anyway.

Because the whole idea people have when they’re talking about perpetual motion machines is the idea that you can extract energy from a machine without any energy being put in. This is impossible–you can’t create energy from nothing. If you didn’t actually need to extract energy from the thing then you could have a machine that ran for a very long time indeed–use low friction bearings, have the whole thing running in a vacuum so there’s no air resistance, etc–but even then, there will always be *some* losses in the system that will cause the machine to eventually stop.

You get energy out of a machine. Anything that you’re taking energy out of, will eventually run out of energy.

A perpetual motion machine would require getting energy out of a machine and also *not* taking any energy out of the machine. That’s just a nonsense idea.

The other answer is excellent, but just jumping in for a ELI5 answer.

Energy cannot be created or destroyed.

Let’s say that to run the machine, you need to put in 100 units of energy. To do anything useful, (like using the machine to power your lights or whatever) you would need 100 units of energy to make the machine go, plus 20 units for the lights.

So. 100 in. 120 out.

Where’s the extra 20 come from? Nowhere. It can’t come from anywhere. There’s no other source of energy.

So that doesn’t work as a generator.

So what about just as an experiment. You want to run a machine just to power itself and nothing else. Pure perpetual motion.

So. 100 units of energy in. Take the 100 units that you get out and use them to keep running the machine. Right?


No energy transfer is perfect. Moving parts lose some of their energy, mostly as heat to the air or with friction between the parts (rub you hands, they’ll heat up). Can you hear the machine running? That’s some of the energy coming out as sound too.

Even with a great, highly efficient engine you’ll only get 80 units out for every 100 you put in.

So. 100 in. 80 out. Not enough to run the machine. It stops.

Some people have built pretend perpetual motion machines that can actually run for a little while by doing this. But not forever. They always stop.

I’m afraid there isn’t really a “why”. The laws of thermodynamics aren’t derived from something deeper – they’re just based on observations. We think that energy is always conserved (the first law) because that consistently happens in experiments. We think that systems can’t be perfectly efficient (which is roughly the second law) because that also consistently happens in experiments. These two laws together imply that you can’t keep something running *forever* without putting energy into it – though plenty of things keep running for an extremely long time, such as the motion of planets.

There is a theoretical connection – a particular case of Noether’s theorem – between conservation of energy and the fact that the laws of physics don’t change over time. If one is true, then the other must be true. But again, we don’t have any deep reason to believe the laws of physics don’t change over time, we just have lots of experimental evidence.

On a very small scale – like that of individual atoms – the laws of thermodynamics become more of a statistical thing – they’re very likely to apply, but not always. But there are a lot of unanswered questions about how the laws of thermodynamics operate in microscopic systems, and this is an active area of research. A big question is whether it’s possible to create devices that can exploit repeated violations of the laws on a very small scale to create violations on a big scale. On the one hand, you’d think if it were possible we would see examples somewhere in nature, but on the other hand you can come up with ideas for such devices (such as “Maxwell’s demon”) and it’s not exactly clear why they couldn’t work.

There isn’t actually a hard reason like with why stuff is blue or why boats float. The universe sort of conspires to make them impossible every time. And it’s so consistent about this that we mostly just assume it’ll keep conspiring.