# Perpetual Motion - why's it impossible? **URL:** \ **Category:** Factual Questions\ **Created:** [November 25, 2001, 10:21pm UTC](https://boards.straightdope.com/t/perpetual-motion-whys-it-impossible/95191 "2001-11-25T22:21:12Z")\ **Posts on this page:** 15\ **Page:** 2
**Author:** ![LordVor](https://avatars.discourse-cdn.com/v4/letter/l/90ced4/32.png) [@LordVor](https://boards.straightdope.com/u/LordVor)\ **Post date:** [November 26, 2001, 3:48am UTC](https://boards.straightdope.com/t/perpetual-motion-whys-it-impossible/95191/21 "2001-11-26T03:48:01Z")
> [@](#): > > \*Originally posted by Qwertyasdfg \* > \*\*Also, for clarity, is it perpetual motion, or perpetual motion _machines_ that are impossible, (although I still think its irresponsible to use that word in scientific context.) \*\* Both, but perpetual motion is slightly less impossible. Perpetual motion means something in a stasis gets a push and then keeps going without any additional outside energy applied to it. To keep going, then the following forces would have to be cancelled: 1. air friction against moving pieces. To overcome this, you really need an absolutely perfect vacuum, which doesn’t exist, even in deep space. Or frictionless pieces, which don’t exist. 2. Friction between any moving parts in your “something”. This either means creating frictionless joints (again, currently impossible), or not have any part of your device touch any other piece of your device. 3. gravitational (tidal) forces. Easiest way to think about this one is the moon’s effect on the oceans. The “high tide” going across the earth provides friction against the earth’s rotation, and is causing the earth to spin almost imperceptably slower over the millenia, and will eventually freeze the earth with one face pointed towards the moon. The same sort of force would be acting against the pieces of your device, and would eventually stop it. Unless, of course, you manage to find one of the points in our universe (and there should be a couple), where all the graviational fields of all the objects that summarily effect it cancel each other out, and manage to transport your device there. I think you can see what I’m getting at. The loopholes that are available are nearly impossible to achieve, and using them make assembling such a device nearly impossible. Therefore, using rough math terms, using these loopholes and assembling such a device would be about half-way in between “nearly impossible” and “impossible”. There are probably other factors that I am forgetting, but you can see how hard it would really be to create something that truely lost no energy to anything outside the system. Now, a perpetual motion MACHINE would actually have to ADD energy to the outside of the system, while still keeping itself going. And there isn’t even a theoretical model for something like that to exist outside of science fiction stories. Such a machine would have to somehow import energy from outside our universe, or else it violates the First law of thermodynamics (See the later books of James Hogan’s “Giants” novels. Which, of course, makes the assumption that all of our physics models are flawed.) -lv ---
**Author:** ![CalMeacham](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/calmeacham/32/35_2.png) [@CalMeacham](https://boards.straightdope.com/u/CalMeacham)\ **Post date:** [November 26, 2001, 3:49am UTC](https://boards.straightdope.com/t/perpetual-motion-whys-it-impossible/95191/22 "2001-11-26T03:49:23Z")
Theoretically, perpetual _motion_ may be possible in an isolated system – two masses orbiting a common center of mass. But I suspect that, in the real world, interactions and friction will inevitably degrade even this apparently friction-free motion. It’ll just take a long time. Perpetual motion _machines_, as has been pointed out above, aree something else entirely. If you try to tae energy out of a system, either you have to replaceit, or else the system loses nergy and eventually stops. See Ord-Hume’s book **Perpetual Motion** for lots of schemes that don’t work. Conservation Laws, by the way, aren’t just empirical laws, or som sort of accountant’s books of the universe. They’e inevtably nd intrinsically bound up with symmetry relationships. Conservation of momentum happens when you have translational symmetry. Conservation of Angular Mmentum happens when you have rotational symmetry. In classical terms, there s no explicit dependence upon linear position or angular position (respectively) in the Lagragian for the system. Conservation of Energy is related to Time Symmetry. Finally, I think the closest you’re going to get to perpetual motion mchines are systems where you take advantage of small changes that you can capture in a ratchet" system – like like thse clocks that are “self winding” because they use micro-changes in barometric pressure to wind the clock. You can also usefully extract energy from quantum fluctuations – Robert Forward demonstrated this several years ago. In neither case can you get a lot of power from this arrangement, though, and in neither case is it a true Perpetual Motion Machine. ---
**Author:** ![JS\_Princeton](https://avatars.discourse-cdn.com/v4/letter/j/57b2e6/32.png) [@JS\_Princeton](https://boards.straightdope.com/u/JS_Princeton)\ **Post date:** [November 26, 2001, 3:49am UTC](https://boards.straightdope.com/t/perpetual-motion-whys-it-impossible/95191/23 "2001-11-26T03:49:47Z")
Okay, some clarifications are in order. Perpetual motion machines are the topic, but what’s really important is the machine idea. What is a machine but a transferance of force. When you look at the simplest machines (levers, pulleys, etc.) they are usually means by which force is redistributed or, changed in some way to make the job to do easier. Work is force times distance (roughly speaking) and work is energy. Well, this is the quantity we are concerned with. By changing either the force or the distance we can change the energy applied to a system to do work. The idea for a machine is to get this work to be effective… that is get it to be efficient. If all the energy applied to a machine goes into effective work, you can have perpetual motion in theory. Well, there’s only one energy transformation that’s 100% efficient and that’s matter/antimatter interactions. This doesn’t, however, really apply to machines because we are usually looking at producing some macroscopic task like spinning a wheel. Besides, looking at a perpetual motion machine based on this energy/mass conversion would be nearly ludicrous in light of the scales needed. At that, I still don’t think such an endeavor would be possible simply due to the fact that there would be a problem of reconverting the energy of the photons produced into noticeable mechanical energy (thought theoretically, my arm might be twisted otherwise). Most “proposed” perpetual motion machines involve electric or magnetic forces which are far below the scales we’re talking about for our mass/energy conversions. In fact, we can look at all energy transfomations as mass/energy transformations if we consider relativity (though that isn’t necessary to understand perpetual motions). The perpetual motion transformation basically requires a special type of violation to the second law of thermodynamics, which is basically a law of statistical mechanics in that it only gives probabilities (albeit ridiculous ones that are unfathomable in the normal sense of fathomability). That is why people explain the 2nd law in terms of things like “tends to”. Well, as it is a statistical law, there’s really only the macroscopic (as opposed to quantum) probabilistic nature of our universe standing in your way. You need a system that would tend to go against the second law of thermodynamics (on the very least locally). In fact, there’s really no reason that you couldn’t design a perpetual motion machine, it’s just that you’d need a perfectly closed system in a perfect vacuum to do it. Find one of those in nature and you’ll be abhorred! :)~ However, if you want, indefinite perpetual motion on timescales we humans deal with is possible. Just supply some external energy source and you can counteract that pesky heat loss. Eventually though, cosmologically, you’re either doomed to heat death, great crunch, or an assymptotic thermodynamical universe that would be forever more and more and more and more difficult to extract your work from. This is getting way out there, but then, perpetual motion machines are way out there. Try to build one and you’ll see why! ---
**Author:** ![Manduck](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/manduck/32/256_2.png) [@Manduck](https://boards.straightdope.com/u/Manduck)\ **Post date:** [November 26, 2001, 7:18am UTC](https://boards.straightdope.com/t/perpetual-motion-whys-it-impossible/95191/24 "2001-11-26T07:18:07Z")
Okay the way I learned it many years ago, is that a perpetual motion machine isn’t simply a device that keeps running forever. That only satisifies the “perpetual motion” part of the description. To be a “machine” it must also do some useful work, i.e. you should be able to use it to move something. If you could eliminate friction you could easily come up with a wheel that never stops spinning, because inertia would keep it going. But if you tried to use that wheel to generate electricity, for example, it wouldn’t work. The reason is that you can’t get more energy out of a machine than you put into it. If you make the machine perfectly efficient it would still only have enough energy to keep itself going without driving something else. To get more energy out than that, you have to put more in. ---
**Author:** ![octothorpe](https://avatars.discourse-cdn.com/v4/letter/o/b77776/32.png) [@octothorpe](https://boards.straightdope.com/u/octothorpe)\ **Post date:** [November 26, 2001, 7:57am UTC](https://boards.straightdope.com/t/perpetual-motion-whys-it-impossible/95191/25 "2001-11-26T07:57:05Z")
Hmm… > [@](#): > > _Originally posted by **Manduck** :_ > To be a “machine” it must also do some useful work, i.e. you should be able to use it to move something. **Manduck** , I understand what you are getting at here, but i don’t necessarily agree that such a machine would have to ‘produce’ useful work. Specifically, what if all the machine did was tell time? **CalMeacham** mentioned a clock that winds itself by way of minute fluctuations in atmospheric pressure (the _ **Atmos** _ manufactured by **Jaeger-leCoultre** is one such example). Disregarding the normal wear on the parts of the movement that will eventually need replacing, this ‘machine’ could (theoretically) run until the end of the earth. While the entropy thing will put paid to the clock actually running forever, this is no mean feat. Does anyone else have an example of a ‘damn-near’ perpetual motion machine? Many thanks to **CalMeacham** for reminding me of this clock; i humbly bow to you. * * * It’s not so much that i fear the perpetual motion machine, rather i fear the evil the telemarketers would put its use to. # ---
**Author:** ![octothorpe](https://avatars.discourse-cdn.com/v4/letter/o/b77776/32.png) [@octothorpe](https://boards.straightdope.com/u/octothorpe)\ **Post date:** [November 26, 2001, 9:25am UTC](https://boards.straightdope.com/t/perpetual-motion-whys-it-impossible/95191/26 "2001-11-26T09:25:40Z")
An addendum to my previous post (i.e. getting my words in before somebody else calls me a stupid nit): I know that the example of the clock is not true to the concept of the perpetual motion machine as it violates the major tenet of not being acted upon by an outside source (in this case changes in atmospheric pressure). A true perpetual motion machine must be totally self sufficient. The post was merely meant to point out that the machine could be sufficient unto itself and would not necessarily have to act upon another machine (i.e. do work). I just find the clock facinating and let my facination get the best of me. My apologies. As to other ‘near perpetual’ machines, thermometers & barometers come to mind. But with the same caveat as mentioned above. # ---
**Author:** ![ZenBeam](https://avatars.discourse-cdn.com/v4/letter/z/3ab097/32.png) [@ZenBeam](https://boards.straightdope.com/u/ZenBeam)\ **Post date:** [November 26, 2001, 6:21pm UTC](https://boards.straightdope.com/t/perpetual-motion-whys-it-impossible/95191/27 "2001-11-26T18:21:14Z")
I think a clear definition of what is meant by “perpetual motion machine” is necessary before you can say why they are deemed impossible. When I think of perpetual motion machines, I think of machines which somehow generate work, in the physics sense of the word, without any external energy source, and which could do so forever. [This site](http://school.discovery.com/homeworkhelp/worldbook/atozscience/p/423920.html) has some useful descriptions of what is commonly meant by the term “perpetual motion machine”, and separates them into two or three types. Here’s another site (which I’ve only skimmed, but looks interesting) which has four types defined, corresponding to the zeroth through third laws of thermodynamics. The devices “inventors” claim as perpetual motion machines typically are supposed to run forever _in spite of friction_. Simply having a wheel spinning in a vacuum doesn’t really cut it, in my mind. ---
**Author:** ![Scott\_Dickerson](https://avatars.discourse-cdn.com/v4/letter/s/f04885/32.png) [@Scott\_Dickerson](https://boards.straightdope.com/u/Scott_Dickerson)\ **Post date:** [November 26, 2001, 7:59pm UTC](https://boards.straightdope.com/t/perpetual-motion-whys-it-impossible/95191/28 "2001-11-26T19:59:39Z")
Some thoughts-- hopefully relevant. 1. Perpetual Motion per se: Various posters appear to be conceiving of motion as a certain generally specifiable condition that is opposed to being “really stopped.” But this presupposes a Newton-type view of the totality of space as a kind of fixed “thing” against which such motion can be measured directly. The prevailing contemporary view (it could change tomorrow, of course) is that space, in part and in whole, lacks properties of locale-identity (= that parts of space are intrinsically differentiable). Thus motion cannot be measured directly against space itself, but only relative to some other thing or things. One guy’s “it’s moving” is another guy’s “it’s stopped.” IF that analysis is correct, then any argument that perpetual motion (as motion, not “work”) is impossible even at the theoretical level must also rule out the possibility of perpetual immobility. (Maybe some relevance to quantum-level jumpiness?..) 2. Machines that do useful work without consuming any resources whatever: Wanna create one? Simple. Take any machine you want–a battery-power flashlight, maybe–and flip the power source 180-degrees in the 4th dimension, so that its “future” now lies (for it) in what we would call the “pastward direction.” Then hook the power source up to your non-time-reversed machine. The amount of available power from your source will increase the longer you use it. (Actually, BTW, you don’t flip the the power source “device”–ie, the batteries–because you don’t want something that was assembled at some recent date: it will disappear that much time into our future. What you want is something more or less eternal–the relevant chemical atoms themselves, which have existed for 6-10 billion years.) Note that some physicists maintain that time travel is not impossible, just a technological problem. 3. And then consider Hilbert’s Hotel, which has infinitely many rooms. You arrive with no reservation, and it’s filled–big convention in town. But there’s no problem: All the guests move one room, to the next higher number–leaving you room number 1, which is now empty. And this works no matter how many newcomers arrive at once!! So, on this principle: if anything like the “Many Worlds” interpretation of quantum uncertainty is true, and there exist an unlimitedly-large number of parallel universes, every universe could borrow energy from its neighbor with no overall losses–indeed, infinite gains! Whoa!! ---
**Author:** ![JRDelirious](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/jrdelirious/32/9531_2.png) [@JRDelirious](https://boards.straightdope.com/u/JRDelirious)\ **Post date:** [November 26, 2001, 11:33pm UTC](https://boards.straightdope.com/t/perpetual-motion-whys-it-impossible/95191/29 "2001-11-26T23:33:26Z")
> [@](#): > > \*Originally posted by octothorpe \* > \*\*Hmm… > > > [@](#): > > > > _Originally posted by **Manduck** :_ > > To be a “machine” it must also do some useful work, i.e. you should be able to use it to move something. > > **Manduck** , I understand what you are getting at here, but i don’t necessarily agree that such a machine would have to ‘produce’ useful work. Specifically, what if all the machine did was tell time? \*\* How would it communicate to the outside world what the time is, w/o expending energy? ---
**Author:** ![Manduck](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/manduck/32/256_2.png) [@Manduck](https://boards.straightdope.com/u/Manduck)\ **Post date:** [November 27, 2001, 5:01am UTC](https://boards.straightdope.com/t/perpetual-motion-whys-it-impossible/95191/30 "2001-11-27T05:01:28Z")
The clock that you describe doesn’t violate conservation of energy (as far as I can see), so it is not ruled out. I think theoretically, if you could eliminate all causes of inefficiency, such as friction, you could make a machine that keeps running forever; BUT you couldn’t use it to make anything else happen, because that would violate conservation of energy. The reason it is said that perpetual motion machines are impossible is because of the law of conservation of energy, and that’s why it will always be impossible no matter how far technology advances; because it would violate a basic law of nature. ---
**Author:** ![DPWhite](https://avatars.discourse-cdn.com/v4/letter/d/d26b3c/32.png) [@DPWhite](https://boards.straightdope.com/u/DPWhite)\ **Post date:** [November 27, 2001, 7:07am UTC](https://boards.straightdope.com/t/perpetual-motion-whys-it-impossible/95191/31 "2001-11-27T07:07:12Z")
> [@](#): > > \*Originally posted by Osiris \* > \*\* > > > [@](#): > > > > \*Originally posted by DPWhite \* > > \*\*A body in motion tends to stay in motion. This is a basic law of physics. For all pratical purposes, the earth is in perpetual motion around the sun (at least for the next couple billion years) > > > > * * * > > Just wanted to point out that in any system consisting of two objects orbiting each other which are not point-masses there is friction. One place where this is most evident is Io where it’s orbit around Jupiter causes a whole heck of a lot of heat. So even if both the Earth and the sun were black bodies there would be a small leak of potential energy into kinetic energy (heat) causing the sun and earth to move gradually close together. \*\* Just wanted to point out that not only is atmospheric friction a potential slow down, but the tidal forces of gravity between two bodies do the same thing. The moon is slowly moving away from the earth because tidal forces are slowing it down (earth’s gravity having a tidal effect on moon.) I think. I am still not a physicist. ---
**Author:** ![TwistofFate](https://avatars.discourse-cdn.com/v4/letter/t/cdc98d/32.png) [@TwistofFate](https://boards.straightdope.com/u/TwistofFate)\ **Post date:** [November 27, 2001, 12:42pm UTC](https://boards.straightdope.com/t/perpetual-motion-whys-it-impossible/95191/32 "2001-11-27T12:42:52Z")
[homer] because on this board, WE OBEY THE LAWS OF THERMODYNAMICS!! [/homer] ---
**Author:** ![Curt](https://avatars.discourse-cdn.com/v4/letter/c/cc9497/32.png) [@Curt](https://boards.straightdope.com/u/Curt)\ **Post date:** [November 28, 2001, 2:02am UTC](https://boards.straightdope.com/t/perpetual-motion-whys-it-impossible/95191/33 "2001-11-28T02:02:01Z")
Perpetual motion is impossible (in all senses) because of irreversibility. Simply put- if you turn an axle, it will heat up, but if you heat an axle, it will not turn. All energy transfers require irreversibility or no work can be done, I’ll spare you the mathmatics of the second law of thermodynamics, which is egghead stuff and tends to diconnect you from the reality. Yet… “hope springs eternal”… humans are ‘very smart’ and might be able to figure out a way around it? Yes and no. Its just as likely for gravity to reverse, hot to become cold and Jerry Falwell realizing cloned tissue does not a baby make, that perpetual motion of either the first or second order will be achieved. first order- a machine that runs forever. second order- a machine that produces EXCESS (usable) work. Neither are possible within the confines of our known universe. Perhaps we may somehow discover a way to suspend the laws of physics… but nothing short of that will work… personlly I hope so, because I hate to think we’re stuck travelling slower than the speed of light 🙂 -Curt ---
**Author:** ![ZenBeam](https://avatars.discourse-cdn.com/v4/letter/z/3ab097/32.png) [@ZenBeam](https://boards.straightdope.com/u/ZenBeam)\ **Post date:** [November 28, 2001, 5:41pm UTC](https://boards.straightdope.com/t/perpetual-motion-whys-it-impossible/95191/34 "2001-11-28T17:41:03Z")
Two bodies orbiting each other will eventually become tidally locked with each other (like the Moon is to the Earth, but the Earth isn’t to the Moon). When that happens, there isn’t any friction anymore (well, they’d have to circularize their orbits, but I imagine that happens eventually, also). They will still emit gravitational radiation, however, until they coalesce into a single spinning body. But hey, isn’t the universe now expected to expand forever? Everywhere you look is a perpetual motion “machine”. ---
**Author:** ![Britt](https://avatars.discourse-cdn.com/v4/letter/b/87869e/32.png) [@Britt](https://boards.straightdope.com/u/Britt)\ **Post date:** [November 28, 2001, 7:48pm UTC](https://boards.straightdope.com/t/perpetual-motion-whys-it-impossible/95191/35 "2001-11-28T19:48:09Z")
> [@](#): > > \*Originally posted by CalMeacham \* > \*\*Theoretically, perpetual _motion_ may be possible in an isolated system – two masses orbiting a common center of mass. > \*\* I am afraid that this system generates (OK is predicted to) generate gravity waves, which reduce the angular momentum of the orbiting objects until they crash into one another. [QUOTE] \*Originally posted by Qwertyasdfg \* \*\* Ah yes, one more thought, isn’t the motion of electrons perpetual in that, in order for them to stop “moving,” wouldn’t it have to reach absolute zero. \*\* OK he 2nd Law of thermodynamics states that the total entropy and the number of black holes in a closed system cannot increase. The conservation of energy states that in a closed system energy can neither be created nor destroyed. Therefore the minimum entropy state of the universe is an even distribution of thermal energy at approximately 2.7 degrees Kelvin. Electrons do not orbit, but they do have angular momentum. You do not need to stop the electron you slow it down until it degenerates in a bust of photons etc, which then tend towards the constant thermal radiation level. This happens well above 2.7 degrees Kelvin due to the electrical charge on the electron. For a non-charged particle with an angular momentum you will have to wait even longer for the heat of the universe to pass below that required to maintain the quark interactions. [Previous page](https://boards.straightdope.com/t/perpetual-motion-whys-it-impossible/95191.md?page=1)