Hacker News Reader

Comment by fmihaila | Hacker News Reader

> Well, perpetual motion is not impossible, obtaining energy from it impossible

This is incorrect.

From Wikipedia [0]: "A perpetual motion machine of the third kind is usually (but not always)[10] defined as one that completely eliminates friction and other dissipative forces, to maintain motion forever (due to its mass inertia). [...] It is impossible to make such a machine,[11][12] as dissipation can never be completely eliminated in a mechanical system, no matter how close a system gets to this ideal (see examples in the Low Friction section). "

[0] https://en.wikipedia.org/wiki/Perpetual_motion

View on HN

Parent is probably referencing https://en.wikipedia.org/wiki/Time_crystal Because a time crystal is a driven (i.e. open) quantum system that is in perpetual motion, it does not violate the laws of thermodynamics... does not produce work...does not spontaneously convert thermal energy into mechanical work...cannot serve as a perpetual store of work... A time crystal has been said to be a perpetuum mobile of the fourth kind: it does not produce work and it cannot serve as a perpetual energy storage. But it rotates perpetually.

In theory I think a closed system undergoing completely reversible processes could be considered perpetual motion (being reversible, entropy never increases). The second law states that those are the only sorts of isolated systems that don't gain entropy over time.

https://en.wikipedia.org/wiki/Second_law_of_thermodynamics

But time crystals aren't perpetual motion just by dint of not being closed systems at all, so they're kind of neither here nor there.

> But time crystals aren't perpetual motion just by dint of not being closed systems at all, so they're kind of neither here nor there.

Literally!

> Parent is probably referencing https://en.wikipedia.org/wiki/Time_crysta

Possibly, but that is not the meaning intended in the article, nor the usual meaning of the expression, so it's not relevant.

That was the weirdest shit I've seen in a good while.

He said perpetual motion, not 'perpetual motion machine'.

Do planets ever stop rotating?

> Do planets ever stop rotating?

Of course they do, on an appropriately long time scale. First they stop rotating around their own axis due to tidal forces, after which they lose the remaining kinetic energy to the gravitational waves emitted by their rotation around the star (again, on a very long time scale, but not infinite).

2nd EDIT: <removed gratuitous reference to moderation>.

Not gravitational waves (you're thinking caused by solar flares moving at the surface of the sun) - but the particle radiation from the sun will continue to slow planets over time.

> Not gravitational waves.

Yes, gravitational waves.

You are correct that in the case of the Earth, other causes will lead to its demise long before said waves would amount to any significant effect. But even in an ideal case where a single planet like the Earth revolved around, say, a black hole with the same mass as the Sun, which would emit no particles, it would still not orbit indefinitely due to the energy lost to gravitational waves emitted as a result of its revolution around another body. The amount of energy lost in one revolution is tiny, but is not zero (according to this calculation [0], for the Sun-Earth system it amounts to about 200W).

[0] https://en.wikipedia.org/wiki/Gravitational_wave#Power_radia...

Yes, two bodies create a gravitational wave that affects other bodies - but I don't see how those variations represent work - and they aren't observable on the two bodies themselves, so one wouldn't expect any effect upon those two bodies. Throw in the other planets and you might have a better argument, for example that Jupiter-Sun waves affect earth. Still, gravity "waves" aren't like water waves, as far as we yet know they don't represent (changes in) the impact of particles. However if you wish to argue that Higg's Bosons mediate gravity (some have, I believe) this objection of mine might be overcome. In other words, though I might be wrong, I just don't think we're at the point where we can deduce friction from gravitational waves, and we certainly aren't at the point where we can empirically measure such friction (as opposed to the "wave" - variation in gravitational field.)

PS, sorry: there should have been a question mark in my previous post asking whether you were thinking about flares.

> I just don't think we're at the point where we can deduce friction from gravitational waves

Where do you think the energy we measure in a gravitational wave comes from? The whole concept arose to explain the last parsec problem of orbital collapse.

There is a change in gravity, true - but a net slowing is another matter. That an explanation of X was needed doesn't justify any sufficient-if-true explanation Y.

> Yes, two bodies create a gravitational wave that affects other bodies - but I don't see how those variations represent work - and they aren't observable on the two bodies themselves, so one wouldn't expect any effect upon those two bodies

The lost energy results in orbital decay, eventually leading to the two bodies colliding. The page I linked to explains it in more detail.

Asserts it, yes - but my last sentence still applies. We don't have a fine-grained understanding of gravity; we'd need it for such a deduction. As it happens, I don't think we've ruled out that gravity itself - not just waves - slowly decays orbits. I rather expect this to be so, in fact, but on the basis of an eccentric theory of gravity.

BTW, I find it amusing that people take internet points away because they think the laws of thermodynamics apply only in certain circumstances.

They do only apply to closed systems.

(which information we are unable to ascertain about our universe)

It's better than that: the fact that the universe even exists implies that at some point there was either a violation of this law or it is part of a larger closed system within which the law still holds.

Or, there is no minimum entropy, and no matter how far back you look there's more entropy. Maybe the Big Bang is an illusion and before the big bang, there was even less entropy, and on and on infinitely into the past.

Which is like saying "I can swing forever as long as someone keeps pushing me." Clearly true, but pedantic and irrelevant to the discussion.

Pedantic needling is always a relevant response to internet point outrage.

Yes and yes. Any other softball questions?

Doesn't inertia guarantee the possibility of linear perpetual motion?

No because there is such a thing as the heat death of the universe, which with the knowledge we have at present seems to be the most likely end-game.

In other words: planets and comets will eventually completely disintegrate because their component atoms and even the components of those atoms will possibly disintegrate.

You're looking at 10^30 years or so, so don't wait up for it, but it is still substantially shorter than 'forever'.

That is such a pedantic and irrelevant answer.

Given the context (planets and comets) it actually isn't.

Planets are just like clocks, slowly winding down, especially the ones with moons through tidal friction.

And even that weren't the case the long run will still get you.

The problem with anybody proposing some kind of perpetual motion device is that perpetual really means 'forever', so no cheating.

If you want to change the scope you can do so but you'll have to drop the 'perpetual' bit.

If it lasts as long as the universe as we know it, I'm happy to call it 'perpetual'.

That's the trouble with definitions.

The way I look at it is the spin-down of planets and the heat death of the universe are reduction to extremes arguments about why perpetual motion machines won't work.

That's the very best you could do, in practice any 'perpetual' motion machine that is actually built will do enormously worse, so much worse in fact that I feel pretty confident placing the word between quotes.

The problem with this subject matter is that it brings the kooks out of the woodworks like not much else does (oh, maybe 'over unity energy', en even more absurd concept).

Of course any practical failures are due to insufficient funds or skill, and any theoretical limitations are just that: theoretical.

But so far in spite of there being in excess of a million dollars of prize money awarded for the first perpetual motion machine nothing seems to have passed the first level of testing.

"The very best you could do" is downright simple if you can afford a big rocket. I think it's fine to admit perpetual motion non-machines where perpetual means 'longer than the universe has existed', and to focus on the important part that there's no perpetual energy extraction.

A handheld box that outputs 1kW constantly for 2 billion years is very impossible despite bypassing all those arguments about eventual tidal/orbital/proton decay.

No, big rockets will return to their point of origin eventually. Compare with a comet, which also re-appears every so many years. The only thing we've managed to launch into inter-stellar space are the Voyager craft and they are simply coasting, like any other bit of inter-stellar debris, straight-line unless captured by some gravitational field. And they'll run out of power within a decade.

> A handheld box that outputs 1kW constantly for 2 billion years is very impossible despite bypassing all those arguments about eventual tidal/orbital/proton decay.

It is impossible in principle, but if it were possible chances are that it would still not be a perpetual motion machine, likely it would be producing an excess of 1KW and be capped at that much output to mask the eventual running down of whatever power source sat inside the box.

Applying all this to perpetual motion, it does not mean 'straight line unless interfered with', it means 'able to change configuration or acellerate relative to it's own frame of reference'.

Dead objects are dead, whether they are moving respective to Earth or not is not relevant and does not magically make them perpetual motion machines, that's just energy imparted at some point in the past.

> Voyager

Yes, that's what I mean by 'big rocket'.

>It is impossible in principle, but if it were possible chances are that it would still not be a perpetual motion machine, likely it would be producing an excess of 1KW and be capped at that much output to mask the eventual running down of whatever power source sat inside the box.

I meant something that is uncapped and does not run down.

Just because it eventually breaks does not stop it from being a perpetual motion machine. Nobody claiming to have a perpetual motion machine is saying it will never need maintenance. An argument that everything needs maintenance is not enlightening. What matters is whether the energy source is infinite, not whether physical durability is infinite.

> Applying all this to perpetual motion, it does not mean 'straight line unless interfered with', it means 'able to change configuration or acellerate relative to it's own frame of reference'.

> Dead objects are dead, whether they are moving respective to Earth or not is not relevant and does not magically make them perpetual motion machines, that's just energy imparted at some point in the past.

Did you forget that this line of conversation started with "Doesn't inertia guarantee the possibility of linear perpetual motion?" They're not talking about a machine with infinite power, they're talking about a dead perpetual motion non-machine.

You are redefining terms in ways that I'm not comfortable with.

Linear perpetual motion isn't.

Dead perpetual motion non-machine is indistinguishable from object at rest. If you don't agree with that you are disagreeing with just about all of physics and any speculation that follows is disconnected from reality as we currently understand it, which leaves a tiny little loophole but not enough to get your hopes up about.

If wishes were horses...

> An argument that everything needs maintenance is not enlightening.

Not everything comes in easy to understand ten word soundbites, any machine needs maintenance due to friction, and as soon as there is friction there is a loss of energy and so the system will run down.

A needle bearing is a really good bearing but it still wears out (and warms up slightly losing some system energy to heat) and will eventually cause your machine to run down.

A big enough flywheel on air bearings could run for a really long time but it too eventually would run down, and air bearings use some energy. Magnetic bearings would work too but would induce a little bit of drag causing some losses to heat.

A current in a superconductor will run for a really long time but - you've guessed it - eventually it will run down. Very low resistance != zero resistance.

Anyway, HN is one of the few forums where scientific discussions are grounded in facts and that makes it a good place to hang out. If you feel that you want 'equal time' for perpetual motion either by redefining the term in such a way that it becomes meaningless or in a way that it no longer qualifies as perpetual motion to begin with I'm fine with that but it makes the discussion rather pointless.

Let me try to explain my entire point in one go, because I don't think you understand what I'm trying to argue.

We have the real world, where something can go on nearly-forever but there are no magic energy sources.

We have unicorn world, where there are magic energy sources.

Arguments about friction and air resistance and tidal decay and heat death prove that nearly-forever is not literally-forever.

That's fine, but it doesn't disprove the unicorns.

If we were in unicorn world, our magic energy source would take 100% power input and turn it into 200% output, and even after losses to friction you'd still have 190% left over.

Saying that "eventually the magic device needs parts replaced" is completely true, but that's not what people care about.

When someone says "perpetual motion machine" they're talking about turning 100% power into 200% power. They're not worried about exactly how long it can run without maintenance.

In other words, you're taking 'perpetual' too literally. You're disproving the idea that anything goes forever, but what 'perpetual' actually means here is that it's extracting energy from an infinite source rather than a finite source.

> You're disproving the idea that anything goes forever, but what 'perpetual' actually means here is that it's extracting energy from an infinite source rather than a finite source.

There are no infinite sources of energy, only very very large ones.

> When someone says "perpetual motion machine" they're talking about turning 100% power into 200% power.

No, that's over-unity, a different concept altogether.

Anyway, it's way past my bedtime here (3 am almost), so I have to bow out here, apologies for that, it was an interesting discussion, I'm not sure if I've achieved anything but that's fine with me. FWIW I was active for years on a message board where 'perpetual motion', 'over-unity' and 'zero point energy' came up with great regularity, if you enjoy that sort of thing you might find similar minded folks there, it is called 'fieldlines'.

sleep well & best regards,

Jacques

> There are no infinite sources of energy, only very very large ones.

Yes, in the real world. But thought experiments are different.

> No, that's over-unity, a different concept altogether.

It's a subset. I'm pretty confident that if you have a hundred people define or describe a "perpetual motion machine", almost all of them will talk about pulling energy out of 'nowhere', or turning energy into more energy. And almost none of them will care if parts wear out, as long as those parts aren't supplying the energy. Maybe I'm wrong, but you're the first person I've seen that has taken 'perpetual' literally and to the end of time.

Good night.

Linear perpetual (unaccelerated) motion is the same thing as no motion.

Are you saying motion == acceleration?

He's making the point that relativity says that if something is "in motion" but never accelerates (including accelerating by turning), you can just put your point of reference with that object and define it as "at rest."

In practice, in the real universe, everything is subject to acceleration. You can only have an object that is infinitely moving without accelerating in a toy universe with only one object. At which point it is probably more intuitive to define it as at rest instead, since there's nothing else around for it to be moving in relation to.

I understand that everything is motionless (with respect to itself).

However, we are talking about motion. So, it is obvious that there is an implicit frame of reference other than the object itself.

It makes no sense for him to bring up that everything is motionless.

It makes perfect sense. Maybe not to you, but all motion is relative to some frame of reference. So if there is no acceleration in any dimension an object might as well be at rest until it collides with something (which would show either one or both objects to be in motion relative to a third reference frame).

The universe does not have convenient 'true' grid lines along its axis to indicate the one true reference frame. You would be hard pressed to indicate a frame of reference that is motionless compared to everything else.

The statement that motion is relative is true, but using it to argue that nothing has motion is what doesn't make sense. An object does not have to be alone in an entire universe to have no acceleration, or utterly negligible acceleration. And if it's not alone, it's trivial to have motion, even if you're not 100% sure which objects are moving.

> The statement that motion is relative is true

Ok.

> but using it to argue that nothing has motion is what doesn't make sense.

Nothing has motion worth discussing that is not able to (a) change it's trajectory or (b) change it's configuration.

As soon as those options are out you have an inert piece of matter, by definition not a perpetual motion machine.

> An object does not have to be alone in an entire universe to have no acceleration, or utterly negligible acceleration.

No, it can be without acceleration in an extremely crowded universe. And any acceleration that it does have will likely come from gravitational interaction with other bodies, or the impacts of particles or other objects (including light).

> And if it's not alone, it's trivial to have motion, even if you're not 100% sure which objects are moving.

That is where we depart. You are thinking of motion in a classical sense, possibly anchored in 'relative to the point of origin', but that's not how it works in physics where acceleration to a local frame of reference is the only meaningful one because it shows you whether or not the 'device' is the master of its own fate or merely along for the ride.

That's a huge difference, roughly akin to 'dead' or 'alive' and of course there are always people who would like to split hairs over that but we all know what 'dead' means. In physics it can be a little harder to prove that an object really is dead but it gets a lot easier once you catch on to the inertial frame of reference idea.

https://en.wikipedia.org/wiki/Inertial_frame_of_reference

A final note that may drop the coin for you: An object 'at rest' does not change relative to it's own frame of reference, but an object that has a powersource can change relative to that same frame of reference. The reference frame stays still relative to where you found the object the first time and then you track it over time. If the change is just a translation then the object is dead, if it can change trajectory relative to that frame of reference, rotate about an axis faster or slower than how you found it initially or change it's configuration relative to the frame of reference (for instance, by operating some kind of actuator) then it is still 'alive'.

> As soon as those options are out you have an inert piece of matter, by definition not a perpetual motion machine.

I think you're mixing up different comment threads. This was entirely about inert pieces of matter with 'dead' motion.

> You are thinking of motion in a classical sens

No I'm not. Look, if two billiard balls are 10 meters apart and that distance is growing by one meter per second, there is objectively motion there. It's inert motion, it's relative motion, but it's real motion. You can use any reference frame you want.

I'm willing to bet that existence of perpetually travelling energy or mass has been around for an infinite amount of time.

A superconductor is sort of a perpetual motion machine for electrons, isn't it?

> A superconductor is sort of a perpetual motion machine for electrons, isn't it?

It may seem like it, but no. As this link [0] explains, "A prime example is the superconductive metals, whose electrical resistance disappears completely at low temperature, usually somewhere around 20 K. Unfortunately, the energy required to maintain the low temperature exceeds the work that results from the superconductive flow."

[0] https://www.britannica.com/science/perpetual-motion

Yeah I didn't mean to imply you could get work out of it, just that it's a system with zero dissipation, so it would seem to be a perpetual machine "of the third kind" as referenced by the post above.

(If you don't count the work required to maintain low temperature, of course, which I don't think you should since you could place the device way out in intergalactic space where it would attain a temperature near 2.7K without you having to do any work to maintain it.

It's a misuse of the word 'machine' to use it on something incapable of performing any kind of work.

Reply on news.ycombinator.com