Talk:Perpetual motion

Minor Edit

I made a minor edit to the first paragraph of the lead. I altered it to remove the implication that 'perpetual motion devices' are impossible in relation to the laws of thermodynamics. My edit is accurate. I hope it doesn't cause any problems with personal bias, but that is what I am about to discuss.

Formatting suggestion

Ugh. Nevermind. Sorry I somehow missed that CLEAR explanation. Apologies all around... The Masked Booby (talk) 23:48, 1 February 2011 (UTC)

Edit request from 75.79.21.196, 18 February 2011

Resolved

{{edit semi-protected}}

Please verify, then add " ISBN 978-0198518907 " to Note 6 - Barrow, John D. - Impossibility:The limits of science and the science of limits

75.79.21.196 (talk) 19:42, 18 February 2011 (UTC)

Done. Rivertorch (talk) 20:08, 18 February 2011 (UTC)

I need Help with a project on perpetual motion machines

Will one of you guys or girls help me with a project on perpetual motion machines EthanKid17 (talk) 16:55, 3 March 2011 (UTC)

This article makes it very clear that there is no such thing that does or even could ever possibly exist according to the fundamental principles of physics as science currently understands them. I wish you luck, but I think there's nobody that can help you do anything other than say "it's impossible and all attempts will fail or are accidental or intentional deception". An interesting project might be to disprove a commonly-believed example if you have one in mind (but I wouldn't want to give money to any of the commercial ones...again, they're only get-rich-quick scams). DMacks (talk) 17:08, 3 March 2011 (UTC)

Yes i understand all that but what about these pictures http://www.lhup.edu/~dsimanek/museum/pm.gif and http://img.rlt.com/A/14100/2_perpetual_motion.jpg or http://img.rlt.com/x/14100.jpg

aren't they feasible. i only have to do it for a seminar. so i don't want to do anything to big for one person to do or create Ethan Krueger 18:35, 3 March 2011 (UTC) — Preceding unsigned comment added by EthanKid17 (talk • contribs)

None of them work. Well, you can turn them and then they gradually (or rapidly) slow down and stop. There are even several similar ones diagrammed in our Perpetual motion article. Google for the phrase "overbalanced wheel" used to describe them to see lots of examples and explanations. Googling for "overbalanced wheel" explanation might help get more info about them rather than just examples of them. There are lots of similar approaches throughout the History of perpetual motion machines. DMacks (talk) 19:11, 3 March 2011 (UTC)

Ok. Can i make one that works for anywhere between 15 minutes and 1 hour. and how big will the machine have to be to get to work for that long. Oh yea i almost forgot something DMacks there is a video i found that is somewhat a perpetual motion machine and i would like to know if it is considered to be one. if you want the video let me know i'll put in your user page that's only if you want to see it. EthanKid17 19:20, 3 March 2011 (UTC) — Preceding unsigned comment added by EthanKid17 (talk • contribs)

Maybe you can explain what the point of the "project" is. If the point is to build a perpetual motion machine, then it will be a failure. If the point is that you can make something that appears to be a perpetual motion machine, then it's possible and could even be educational. Nothing described here (overbalanced wheels, etc) will run for more than a few seconds. In order to get something that looks like perpetual motion, you need to have a hidden (or non-obvious) source of power. The simplest and most elegant "perpetual motion machine" I know of is the "drinking bird". Now if you made something starting with a drinking bird, but took off the obvious bird features, and perhaps hid the water in something that looked like a electrical coil - and then added a bunch of completely superfluous features, you could probably cook up something that wouldn't be recognized and would run for at least several hours. Anyway, that's just the first thing that comes to mind. Another fairly common trick is to have a "perpetual motor" that has an externally powered fan "to cool it", as is done in this demonstration of the "Newman Machine".Prebys (talk) 19:37, 3 March 2011 (UTC)

(ec) No - they don't work anywhere near as long as that - 20 seconds - tops. The overbalanced wheels have much more internal friction than (say) a simple bicycle wheel - so they certainly don't spin freely for even as long as a bike wheel does. To get something to spin for 15 minutes, you'd probably need magnetic levitation bearings and a much reduced air pressure. If you just want something that SEEMS like it's perpetual motion, (with the CLEAR knowledge that there ain't not - nor can ever be such a thing) I would try to buy a "Crookes radiometer" (which in reality is powered by sunlight - but it looks like it's perpetual motion to people who are uninformed about such things...until you try to run it in the dark!) SteveBaker (talk) 19:39, 3 March 2011 (UTC)

SteveBaker (talk) 19:39, 3 March 2011 (UTC)

Just thinking out loud now, but another really easy thing would be to use a pair of coils to inductively couple power in. Once coil could be below a table and plugged into the wall, and the second could be hidden in the base of your "device", sitting on top of the table. With that, you could power literally anything (including one of the "overbalanced wheels" in the pictures). You could do an elaborate show to prove there was no contact or connection to the table.Prebys (talk) 20:08, 3 March 2011 (UTC)

i would paste the assignment to wiki but i fear that i might get blocked for it. so if you want to the specifics i could email them to you after school when i get home. which is in like about an hour to a hour and fifteen minutes. everyone else in the seminar have groups of two and the odd number of students makes me to be the one alone. so i need all the help i can get either on building the device from the specifics that i send to you. i really think that the device has to look like a perpetual motion machine at least i think. but i will check when i get home.

BTW if you really do want to the exact specifics let me know on my user page please and i'll give you either my e-mail or you can give me yours. but i will most likely give you mine. but that will have to wait till i get home. but in advance thank you for any help that anyone can provide me. EthanKid17 20:30, 3 March 2011 (UTC) — Preceding unsigned comment added by EthanKid17 (talk • contribs)

This discussion probably should be moved to EthanKid17's talk page, since it has nothing to do with improving the article. Just a suggestion. Rivertorch (talk) 21:28, 3 March 2011 (UTC)

Or not on wiki at all, since I don't think the teacher will appreciate us giving substantially more help than we already have. DMacks (talk) 21:35, 3 March 2011 (UTC)

Well, perhaps. I don't suppose what a teacher appreciates or doesn't appreciate ought to govern what gets discussed on WP talk pages. This was the proper place to ask the question in the first place, but I figured that since the discussion was already well underway it might be continued more productively in userspace. This page is already long and in need of more archiving, and this thread has nothing to do with the article. Rivertorch (talk) 21:56, 3 March 2011 (UTC)

There are an enormous number of ways to fake a perpetual motion machine demo - that's why so many frauds have been inflicted by people over the years. But basically you're either going to need:

1. An 'unobvious' external energy source (like sunlight in the Crookes radiometer case or one of those sterling engines that can run from the heat given off by the palm of your hand when you hold it up - or the nodding bird that's running from the temperature differential from a glass of water).
2. A more deliberately hidden energy source (like using induction coils below the table)
3. Something like a really heavy pendulum that has enough stored kinetic (or whatever) energy to keep moving for the duration of the demonstration without perceptually slowing down.

I question the value of having students do something like this though. SteveBaker (talk) 14:13, 4 March 2011 (UTC)

ok the oil pumps in texas do they have some part of a perpetual motion machine in them. I will be moving this discussion of mine to my userpage and i could use anyone's help that has knowledge in this area. and my teacher doesn't mind. EthanKid17 16:37, 4 March 2011 (UTC) — Preceding unsigned comment added by EthanKid17 (talk • contribs)

NO! There are no perpetual motion machines...perpetual motion is completely impossible! Didn't you read our article? Anyway, those are "Pumpjacks" (often called "Nodding donkeys" by the locals) and they commonly powered by an electric motor, or by a perfectly ordinary internal combustion engine running on diesel or propane if they are too far from an electricity supply. Some of them in really remote areas run on the very unrefined oil that they are pumping. In every case there is an easily visible energy source. They are about as far from a perpetual motion machine as you could possibly imagine! SteveBaker (talk) 17:46, 5 March 2011 (UTC)

yes i know that it is impossible to make one the world doesn't have enough money to even start designing one. the project i am doing is more about the learning process and when i run into failures or problems how can i improve on the problem/solution to make it better. any information and facts that you guys or girls give me i will give you the credit on that information so i don't take credit for the facts you give me.

Edit request from 175.180.185.43, 4 March 2011

{{edit semi-protected}} please add the text below to the section

Thought experiments

........

Techniques

Another thought experiment that yields a perpetual motion machine of the second kind: Take a resistor which generates Johnson–Nyquist noise and connect this resistor to an ideal Diode ( a Diode with a on-voltage of +0V or some uV above this). The uni-directional current through the diode can charge a capacitor. That capacitor charge could be used to do work outside the system of resistor, diode and capacitor. At this time, a suitable Diode is not available, therefore it is impossible to demonstrate such a device. However, in superconductors a similar effect can be found. A persistent current that can be interpreted as rectified Nyquist’s noise. http://arxiv.org/abs/cond-mat/0310073

175.180.185.43 (talk) 08:37, 4 March 2011 (UTC)

Johnson-Nyquist noise is a thermal spectrum. The diode would preferentially select the more energetic electrons thereby cooling the resistor. The physics involved is essentially the same as a thermocouple.Prebys (talk) 13:19, 4 March 2011 (UTC)

I've closed this edit request because, regardless of this thought experiment's veracity, it can't be included in the article without a reliable source. Adrian J. Hunter^{(talk•contribs)} 14:57, 4 March 2011 (UTC)

Wouldn't single energetic electrons lack a definable Kelvin temperature? Also, in what sense would it heat up something cool if it is converted into electricity? Only the resistive component of the electricity would generate heat. The heat transfer would be only a fraction of the total energy transfer.siNkarma86—Expert Sectioneer of Wikipedia
^{86 = 19+9+14 + karma = 19+9+14 + talk} 15:02, 4 March 2011 (UTC)

(ec) No! This is just a simple variation on Maxwell's demon...and it fails for precisely the same reason. The demon with the shutter that it can open to let energetic particles through is just the diode in your example. The demon creates more entropy than it can eliminate - and so will your diode. You can't use the "thought experiment" label to arbitarily claim the existence of a perfect diode if that perfection is not possible...that's how Szilard (and later, Bennett) shot down Maxewell's demon and restored the primacy of the second law. They realized that you can't just claim a "perfect" demon - even in a thought experiment...and for the same reason, you can't claim to be using a "perfect diode" if the laws of physics don't permit such things.SteveBaker (talk) 15:13, 4 March 2011 (UTC) <-Consider this: The Maxwell's demon is always only a thought experiment. The concept of the Diode is real. Different kinds of diodes have different on-voltages. What is needed here is not the ideal Diode with an on-voltage of +0V, but an on-voltage that is below the noise voltage of the resistor. The Diode shows a way to realize something that comes close to the Maxwell's demon. In the article about the Maxwell's demon you will read that the Maxwell's demon actually does not violate the second law of thermodynamics. Thus this perpetual motion machine of the second kind is possible, and now the realisation only depends on finding a Diode with an on-voltage that is below the noise voltage of the resistor. The available energy will not be enough to cook an egg, only enough to prove the point.175.180.185.43 (talk) 16:17, 4 March 2011 (UTC)

Bottom line is that what you have here can no more violate the second law of thermodynamics than my refrigerator can! SteveBaker (talk) 15:13, 4 March 2011 (UTC) <-violating the the second law of thermodynamics is not the point, the point is to show that perpetual motion machine of the second kind is possible.175.180.185.43 (talk) 16:17, 4 March 2011 (UTC)

This is why Wikipedia disallows "original research" (See: WP:NOR) - we couldn't write about this idea of yours even if it worked. It's not the business of editors here to figure out why every crazy perpetual motion idea doesn't work and thereby prevent the nut jobs from putting junk into the article. (Not that I'd label you as a "nut job" - this is a thoughtful and well-reasoned concept.) If you are convinced that your idea is true - then this is the wrong place to promote it. Instead, you should write it up as a proper scientific paper, get it peer-reviewed and published - and claim your Nobel prize (finding a loophole in the second law of thermodynamics would make you a shoe-in!)...and then we'll write about it in this article using your paper as a "reliable source"...unless of course it gets shot down at the first hurdle (which I'm 100% certain that it will!).

Good luck anyway! SteveBaker (talk) 15:13, 4 March 2011 (UTC)

I think I'm not the IP?siNkarma86—Expert Sectioneer of Wikipedia
^{86 = 19+9+14 + karma = 19+9+14 + talk} 15:45, 4 March 2011 (UTC)

The bottom line is that you have to always include all the physics in the problem. You can't take a physical process like Johnson-Nyquist noise and combine it with an unphysical thing like a "perfect diode". Another good example is the Brownian ratchet. It appears to work if you combine the physical process of Brownian motion with an unphysical ideal ratchet. Feynman showed that a ratchet sensitive enough to Brownian motion would undergo Brownian motion itself, rendering the device useless.Prebys (talk) 15:42, 4 March 2011 (UTC)

This is correct. Heat due to friction will occur in any real diode. However, the diode itself is a passive device. It makes no attempt to do any work, and thus, the diode in cannot itself be responsible for entropy increase. If you think about it really, if heat were to be converted into electricity in a diode, that would itself constitute cooling, and thus the temperature drop, assuming it is definable here, certainly must occur, causing entropy to increase. Also interesting is that the reverse process, though not an exact time-reversed process, creating heat by using electricity to pass current through a circuit element of non-zero resistance, also increases entropy.siNkarma86—Expert Sectioneer of Wikipedia
^{86 = 19+9+14 + karma = 19+9+14 + talk} 15:58, 4 March 2011 (UTC)

In any case, an imaginary Maxwell "Demon" is not necessary to allow the diode to ratchet energy in one direction. All you need first is simply an energy source, such as heat. This otherwise useless energy becomes the basis for the work necessary to convert it into electricity. In practice, less than 100% will be converted into usable electricity. The COP of converting this heat into energy also gives us the energy cost in conversion. Basically you take the amount of heat converted into work and divide that by the amount of heat not converted into work. So if 20% was not converted into work, while the remaining 80% was, then the COP would be 4. Instead of a Maxwell "Demon" powered by nothing, what really happens in this system is 20% of the heat was missed rather than converted into work. No new entropy is really created if the collecting device consists of 100% passive components, except for any active systems required to assemble and deploy the passive components in the first place. As long as this COP is less than infinite, we will have not violated the second law of thermodynamics.

You could somehow try to recover the remaining (missed) heat, and reflect it back to the system again, but you will still not convert 100% of it. The system which dissipates the waste heat can be any sort of machine, whether it is a motor or not. A fraction the same energy could be passed through one motor after another, and one car to another, and one house to another, etc. but you will always lose some energy each time, so practically, this back-and-forth exchange cannot continue on forever. The amount of positive work that can be done with a unit of energy is that energy times 1/(1-x), where x stands for the percent of waste heat recovered in each cycle. This assumes that all the negative work is done by the dissipation of heat. The net work done on the whole environment (i.e. motors, cars, houses, etc.) will always be less than that initial energy.

The way we use energy in today's economy (which is characterized by the amount energy stored in moving parts in the economy at any given time) is just like how energy is wasted in an overdamped oscillator. The vast majority of energy that is removed from storage does not return, and more than half of the energy used up is not even stored in useful devices in the first place. More than half of our energy is dissipated as heat before we even get to use it to do useful work! Our economy behaves as though it were an overdamped oscillator because right now we cannot recover even 15.9 percent of the waste heat. A future economy based on collecting waste heat from devices could change that, despite having the inevitable consequence of having an efficiency less than 100%. However, in such a future economy, energy available would not be an unapproachable limit to the positive work which could be done, but rather, it would be an unapproachable limit to the amount of energy that could be stored in the economy's moving parts at any moment in time (i.e. it would be a limit only to net work). So for example, even if just 50% of waste heat could be recovered, the usefulness of our energy reserves would double. If 80% of the waste heat could be recovered, then this would quintuple the usefulness of our energy reserves.

In this light, the only totally renewable energies are based on atmospheric and hydrospheric energy independent of contributions of sunlight, tidal forces, tectonic, geothermal, chemical, nuclear, and other, yet exotic, mass-energy conversion processes. While renewable sources could be sufficient to power our artificial devices, our state of existence is also contributed by the other forces. However, for a finite time as long as the other conditions remain stable, a small amount of input energy would be all we need to keep our machinery going as long as we retain an atmosphere which would harbor much more than the waste heat the we have put out. We would just have to recollect that energy through the use of infrared photovoltaics, wind turbines, and/or mechanical dampeners such as buoys.siNkarma86—Expert Sectioneer of Wikipedia
^{86 = 19+9+14 + karma = 19+9+14 + talk} 16:31, 4 March 2011 (UTC)

You need to understand something about "waste heat." Heat engines can only work by moving energy from the "hot side" to the "cold side", with an efficiency of each step in the cycle absolutely limited by 1-(Tc/Th). (Tc and Th are temperature of cold and hot side resp., and must be in an absolute zero-based scale like Kelvin or Rankine.) The heat must be continuously removed from the cold side to keep it cold, to keep the efficiency from falling. For example, it is easy to say "look at all the waste heat coming off of a car engine, we should be able to use that somehow." But most attempts to use it will reduce the efficiency of the engine's heat sink (the radiator, exhaust pipe, and the engine block itself), thus raising the temperature of the "cold side", thus hurting the efficiency of the engine's primary cycle. Sometimes "co-generation" tricks work. A common example is the car's heater, which is just another radiator in a convenient place. Another example is "smart buildings" that use the heat from a server room to heat other rooms. But practical applications that don't hurt the "main cycle" more than the recovered energy is worth are pretty rare. Jeh (talk) 20:02, 4 March 2011 (UTC)

The truth is that less than 100% of the waste heat can be transduced to a different temperature by removing it from a larger number of degrees of freedom and dumping it onto smaller number of degrees of freedom. Such transduction is not done with thermodynamic "engines" which involve PV work. Conduction and convention cannot do this simply due to the simple fact of geometry that the volume increases with the distance traveled, and so heat typically spreads to a larger number of degrees of freedom (i.e. temperature decline). In contrast, radiative transfer, might do this so as long as the medium which carries the heat has enough transparency to this radiation which would allow this heat to be transduced through concentration via mirrors and/or lenses. Failing that impractical method, converting heat directly into voltage and current, which actually constitutes a temperature drop, can allow this energy to bypass the gradient between hot and cold, and then such energy can be transduced back into heat (temperature increase) using a resistive heating element. It is also the case that some of this waste heat comes out still having a significant temperature difference with respect to the ambient temperature, which may generate enough pressure differentials, which as potential energy would convert into kinetic energy in the atmosphere, which can then be captured by a wind turbine, which can account for some (albeit small) fraction of the energy used in generating the hot reservoir from which the waste heat came. Regardless of all this potential to recycle waste heat repeated times, it still does not violate the second law of thermodynamics to do so because "the truth is that ONLY less than 100% of the waste heat" can be converted, COP is never infinite, and the system will stop oscillating.siNkarma86—Expert Sectioneer of Wikipedia
^{86 = 19+9+14 + karma = 19+9+14 + talk} 20:22, 4 March 2011 (UTC)

This thread is getting really long and no longer has any relationship to the article. It should probably be moved to a USENET group.Prebys (talk) 20:36, 4 March 2011 (UTC)

There is no mechanism to satisfy your proposal that is consistent with any standard use of the wiki interface.siNkarma86—Expert Sectioneer of Wikipedia
^{86 = 19+9+14 + karma = 19+9+14 + talk} 20:46, 4 March 2011 (UTC)

I was politely trying to say that this thread - although interesting - has nothing to do with improving the article anymore. It should be terminated here and (if it continues at all) be taken to a more appropriate venue.Prebys (talk) 22:14, 4 March 2011 (UTC)

Hopefully the argument has finally ended with my large post above.siNkarma86—Expert Sectioneer of Wikipedia
^{86 = 19+9+14 + karma = 19+9+14 + talk} 22:28, 4 March 2011 (UTC)

This is actually a useful line of discussion. The laws of physics as we currently understand them obey the laws of thermodynamics, so any argument based on those laws that appears to generate perpetual motion simply must be wrong - although the flaw is sometimes quite subtle. I think it's worth adding a "Gedanken experiment" category to the "Apparent Perpetual Motion" section, which I'll do when I get some time. It will include at least the Brownian Ratchet and Maxwell's Demon. Anything else?Prebys (talk) 16:04, 4 March 2011 (UTC)

Thought experiments

Based on the discussion above, I've added a gedanken experiment subsection to "apparent perpetual motion". It currently includes Maxwell's Demon and the Brownian Ratchet. Any other suggestions. In particular, does anyone know if there's a WP:RS for the Nyquist idea above? I googled a bit, but all I could find was one kook paper.Prebys (talk) 16:38, 4 March 2011 (UTC)

---

Could this be the WP:RS for the Nyquist idea above?: http://arxiv.org/abs/cond-mat/0310073 It does not exactly mention a "Diode", but rectified Nyquist’s noise.

arxiv.org is self-publishing service, so it doesn't qualify as a WP:RS for WP:FRINGE topics.Prebys (talk) 18:29, 5 March 2011 (UTC)

There is another thing: All the discussions above go around creating a temperature difference with a Maxwell Demon or similar that should eventually drive a Heat Engine to obtain work. But in the Nyquist idea above the potential for work is obtained by creating a voltage potential through rectification of an electric current. The point of this exercise should be to establish that although it seems really impossible to build a perpetual motion machine of the second kind with a heat machine, we should still acknowledge there might be other ways to do that. 175.180.185.43 (talk) 13:01, 5 March 2011 (UTC)

No, we shouldn't. WP does not print editors' speculation (that's called original research). Not unless you can find a reliable source that's in agreement. And in fact, since this is WP:FRINGE territory, WP would prefer to see several RSs in agreement, since RSs have to be interpreted by WP editors (or else we'd be infringing copyright), and our interpretations might be wrong; the chances of this diminish when we have several RSs. More generally, we don't put "this could turn out to be wrong, or to have some exceptions" on all statements of scientific knowledge, and there's no reason to make an exception here. Jeh (talk) 16:55, 5 March 2011 (UTC)

I agree 100%. As I said above - it's (mildly) interesting to hear about proposed perpetual motion machines and figure out why they are broken (the fact that they are broken is really not under any kind of rational dispute)...kinda like solving a crossword puzzle. However none of that discussion - fascinating or otherwise - belongs in our article. We are absolutely forbidden (for reasons that should be obvious from the kookie posts above) from engaging in original research or synthesis of 'truth' from multiple sources.

Bottom line: We absolutely cannot under any circumstances put any of this into the article. That's because you don't have a reliable source. Specifically: An article in a reputable, peer-reviewed science journal or book that says "If you connect a resistor to an ideal diode, you get a machine that breaks the second law of thermodynamics"...or whatever it is. Until you have such a source, this completely (and obviously) wrong idea doesn't get within a million miles of being in our article. That's not my call - not your call - it's written large in the fundamental guiding principles of this encyclopedia. SteveBaker (talk) 17:38, 5 March 2011 (UTC)

As I have stated in my above (non-kookie) posts, multiple times in fact, the proposed device will not violate the second law of thermodynamics. At best, it could recover some otherwise wasted heat and recycle a fraction of this a finite number of times, but it never may use all of it. It is not a perpetual motion machine of the second kind.siNkarma86—Expert Sectioneer of Wikipedia
^{86 = 19+9+14 + karma = 19+9+14 + talk} 02:58, 6 March 2011 (UTC)

But as SteveBaker said (multiple times), it can't go in unless you can find a reliable source for it.Prebys (talk) 14:49, 7 March 2011 (UTC)