Nanotech free energy

[Ziggurat]

Only if the engine is frictionless and weightless.

Your point? The second law is supposed to hold even with frictionless engines (which will minimize entropy creation, but cannot reduce total entropy), and the weight of the engine is irrelevant.

 

[roger]

Not at all. The valve is a normal, macro-scale valve (I provided you with links). Just like the engine is a normal, macro-scale engine. Again, we know valves work. I'm a mechanical engineer, trust me, they do. If you don't want to trust me, try my links.

Of course I can change scales. We can have a small hatch and a big valve. Or rather, millions upon millions of small hatches and one big valve.

We know values work with macro gases, in macro scale.

As for your second paragraph, this strikes me as a "it's turtles all the way down" idea. We have the problem that the few gas molecules can't exit your device in the face of macro gas pressure. So, you introduced a valve. I say it can't work in micro. You say, if I read you correctly, that you are connecting 1 billion (say) of your devices to a single, macro valve.

So, how does this connection work? If this is all just hooked up with a bunch of pipes, you will build > atmospheric pressure on the left of the valve, which means that now every one of your billion devices sees > atmospheric pressure. And it won't work.

Of course, you could put a valve between your device and your valve, but now we have a cascading series of turtles.

 

[Merko]

ALL: The device works as stated in the OP. However, the door will, as stated in point 6 in the OP, sometimes open randomly from vibrations. In fact, this happens much more often than a molecule hitting it.

Sometimes when the door flips open from vibrations, it will hit a molecule as it swings back, hitting it like a baseball bat hits a ball, likely pushing it out of the box. This effect cancels the positive effect detailed in the OP.


My ponderings in post 82 can also be explained. Yes, the door will also 'clear' an area inside the box, by 'batting' molecules further into the box, away from the hatch, and thus reducing their chance of exiting. However, molecules hitting the door at a high speed from the inside will equally be 'caught' by the door, they will lose speed and transfer it to the door, and so they will stay in the vicinity of the door which increases their chances of exiting.


As far as I'm concerned, the case is closed.

 

[roger]

I may be wrong but I don't think roger is saying that. I think he's covering both cases because he realizes that there's been some sloppiness about slipping between arguments that depend on a "nano" scale and others depend on a "macro" scale. I think he's pointing out that there's been a lot of convenient handwaving choosing "nano" versus "macro" when it suits the argument someone wants to make at the moment.

Yes, this is what I was saying.

I added a diagram to an earlier post. I have seen no argument that allows pressure to build up with only a few energetic gas molecules in the center, with uncalibrated nano doors on the input and a macro valve on the output. Just claims that it "obviously" works.

 

[roger]

As far as I'm concerned, the case is closed.

I hope you don't mind that some of us keep discussing it. It's a fascinating device, and the details are fun to think about. I know I've learned stuff in this thread. Good times.

 

[Merko]

So, you introduced a valve. I say it can't work in micro. You say, if I read you correctly, that you are connecting 1 billion (say) of your devices to a single, macro valve.

I don't know where you got the idea from that it would be a nano-sized valves. For sure I never introduced any (again, check my links).

So, how does this connection work?

Uh, it's just a box. You know, a container with walls. One side of the box, we have a hatch (or n hatches). The other side, we have a valve. Not that the valve is really needed, I just introduced it because you didn't like the engine. The valve, if you follow my link, is really just another piece of wall unless the pressure difference gets big enough. It does absolutely nothing until then. And if the pressure would increase (it won't, of course), that would have meant our device would have worked.

I hope you don't mind that some of us keep discussing it. It's a fascinating device, and the details are fun to think about. I know I've learned stuff in this thread. Good times.

Oh, I might stay around to see if there's something I can explain. But I'm satisfied that I understand the proposed system now. Finally, after 10 years I can sleep again.. no just joking. Haven't really thought about it that much, but I did get the idea 10 years ago.

 

[RecoveringYuppy]

If you're tired of the conversation don't feel obligated to address this. I'll continuing with other people, or not, as they see fit.

My ponderings in post 82 can also be explained. Yes, the door will also 'clear' an area inside the box, by 'batting' molecules further into the box, away from the hatch, and thus reducing their chance of entering.

This seems to be one of those convenient changes in assumptions midstream to make an idea work. At other points it's been assumed that the molecule momentum/door inertia ratio was high enought so that the collision of molecule with door could still leave the molecule with rightward movement. Yet here we have the door reversing the motion of similar gas molecules.

 

[Modified]

So, how does this connection work? If this is all just hooked up with a bunch of pipes, you will build > atmospheric pressure on the left of the valve, which means that now every one of your billion devices sees > atmospheric pressure. And it won't work.

The whole point of the hatches is that if they could work, they would work against pressure to some extent. That is the "magic" part, a pump that works by extracting heat from a uniform environment.

 

[Merko]

I think he's pointing out that there's been a lot of convenient handwaving choosing "nano" versus "macro" when it suits the argument someone wants to make at the moment.

There's been no handwaving at all. Only the hatch has ever been 'nano'.

 

[Merko]

This seems to be one of those convenient changes in assumptions midstream to make an idea work. At other points it's been assumed that the molecule momentum/door inertia ratio was high enought so that the collision of molecule with door could still leave the molecule with rightward movement. Yet here we have the door reversing the motion of similar gas molecules.

Sure, but there's no problem with that at all. The molecules are indeed heavier than the door. They can hit the door and continue to the right. But they can also be hit by the door which will increase their speed. One possibility does in no way exclude the other. Ok, maybe the bat/ball analogy isn't perfect when the ball is heavier than the bat, but consider a light, fast moving bat and a big, heavy ball.

 

[DavidS]

How do you generate sufficiently high pressure with a single molecule?

(a) Put it in an ittty bitty box so it bounces off the walls -- reversing its momentum -- so often that you have to "push in" momentum at a sufficiently high rate -- that is, apply a sufficiently large force -- which, over a sufficiently small area is a sufficiently high pressure -- to keep it there. That's what most of us usually mean by "pressure".

(b) No need. Just collect/separate the molecules you want to use to generate the pressure one at a time into a box as big as you want/have/need, then let your entire collection work as a team. That's how we "generate... pressure" from molecules every day.

The simulation I ran was just a classical simulation, with the gas as small balls which undergo random, perfectly elastic collisions, with a door on a damped spring separating two rooms. The "molecules" did tend to enter the door more than exit it. I certainly don't think my simulation properly represents what happens in the real world, but the reason it doesn't work should have a better explanation than just saying "the door won't work, the rates will be equal, because otherwise it would work and violate the second law of thermodynamics".

Though on reread I guess it does have that tone, I didn't mean to suggest that you didn't properly simulate all the physical processes you intended, even including everything we normally consider meaningful in a "classical" context. My issue is that the question of the device's feasibility hangs on whether there are processes, "classically" ignored or negligible, that become relevant an the nano-scale that would permit or prevent its operation. While the question of what those processes might be -- or how they can be simulated -- is unsettled, simulation doesn't offer much evidentiary potential precisely because of the uncertainty whether the "simulation properly represents what happens in the real world".

That is, we just can't put much faith in simulation until we settle on what needs to be simulated.

Your simulation suggested the gate "worked". Most of us didn't think it should, but we were having trouble coming up with the reason it shouldn't. The simulation couldn't really "validate" the device, because we couldn't really validate the simulation. No matter how perfectly you simulated the processes we were considering, we couldn't tell how well you simulated processes we hadn't yet identified as potentially important.

We'd be little better off if the simulation said the particular device it simulated would *not* work. Besides the same question of whether the simulation properly represents all the nanoscale phenomena we have and expect to come up with, it's tricky for such a simulation to show that a somewhat different device could work.

Hmmm... I guess I do owe Dilb an apology. I confess I was too intent on sniping the point above and didn't properly give you credit for what value simulation really could offer. Though simulation in this case isn't useful as a "go-nogo" touchstone, even a known-to-be-incomplete simulation can serve an illustrative purpose and lend insight to the rest of the investigation.

 

[roger]

I don't know where you got the idea from that it would be a nano-sized valves. For sure I never introduced any (again, check my links).

Oh, I introduced it. Basically, the valve will be nano, or macro, I'm sure we both agree. My question is which, because I believe there is a fatal flaw with either design. You say macro, so I will say no more about nano.

Uh, it's just a box. You know, a container with walls. One side of the box, we have a hatch (or n hatches). The other side, we have a valve. Not that the valve is really needed, I just introduced it because you didn't like the engine. The valve, if you follow my link, is really just another piece of wall unless the pressure difference gets big enough. It does absolutely nothing until then. And if the pressure would increase (it won't, of course), that would have meant our device would have worked.

I really do understand the macro valve you are proposing. I get that if you put this in the pipes connecting your water supply to your faucet, it will just flap open when you turn on the faucet, using the energy of the pressure of the tank. I do get that. They exist, you can buy them from any plumbing supply or industrial supply house. Bog standard technology.

My issue remains that I believe there is an impedence mismatch between your nano sized chamber and the engine. Surely it's a moot point since we know the input door doesn't work. But I'd like to understand if I am right or wrong.

If there is an impendence mismatch, this flap/valve does not change that issue. You now have a billion nano chambers connected in front of a macro valve. The valve will only open when there is sufficient gas pressure, we both agree, where sufficient means somewhat greater than the atmospheric pressure outside of your nano chamber. Your nano chamber's outputs are connected to this valve, so they each see at their output this proposed greater than atmospheric pressure pressure. The impedence mismatch still exists. If the nano device wouldn't work when connected directly to the engine due to the pressure, they still won't when this valve is in line.

And, I say there is an impedence mismatch because your nano device is, necessarily, nano. If your front latch worked, it would have worked because it was not being subjected to normal gas pressures inside the chamber, but only to an occasional molecule bump. Your idea hinged on that fact, I thought. That means that the output side of the chamber also does not have normal atmospheric pressure, but the occasional molecule bump. You have this chamber connected (eventually, via the engine) to normal gas pressure in a macro environment. Hence, an impedence mismatch. One or two gas molecules at a higher energy state then the rest of the room, trying to fight their way out of a chamber against normal gas pressure. That's not going to work, so far as I can see. I admit I haven't done (nor do I have the training) the math to prove this.

In any case, introducing the output valve doesn't change anything, because to work (which they do), the valve needs higher gas pressure on the input than at the output. You haven't decoupled the chamber from the high gas pressure at all, as far as I can see. Perhaps I misunderstand the atmospheric conditions you postulated existed inside your nano chamber?

 

[DeviousB]

(b) No need. Just collect/separate the molecules you want to use to generate the pressure one at a time into a box as big as you want/have/need, then let your entire collection work as a team. That's how we "generate... pressure" from molecules every day.

N.B. Molecules, hence there is more than one. If you have more than one molecule on the inside of the flap, it is not a 'single molecule' system and you cannot treat it as such.

 

[RecoveringYuppy]

Ok, maybe the bat/ball analogy isn't perfect when the ball is heavier than the bat, but consider a light, fast moving bat and a big, heavy ball.

And from post 82:

it also 'clears' an area on the inside of the box from molecules.

Well, I certainly like the light bat, heavy ball wording better than the wording from post 82.

I (and roger I would presume) am not accusing you of any kind of malicious deception here. Just pointing out that the discussion is rife with inconsistencies that lead to poor intuitive conclusions.

But post 82 is the kind of inconsistent, biased wording that can lead to major mistakes in an intuitive or qualitative argument. Under one circumstance (letting molecules in) the door is a lightweight contraption that can get out of the way of a single incoming molecule without sending it rebounding and unlikely to ever contact two molecules simultaneously. Under another circumstance the door is capable of clearing an area of molecules.

This could be why you can't get comfortable with detailed balance.

 

[Merko]

That is, we just can't put much faith in simulation until we settle on what needs to be simulated.

My guess is that the simulation didn't take the door's random flapping due to vibrations into account, and if it did, it didn't make the door flap often enough. Dlearly it's not too easy to calculate exactly how often it should flap, unless we assume that it will flap exactly often enough to 'bat out' enough molecules to counter the gate effect, which I now assume it will.

I'm still not sure the system would have to satisfy the detailed balance if it worked though. Then it would not have been a system in equilibrium. But Ziggurats suggestion still solved it, because as we all suspected, the device doesn't work, and then the system is in equilibrium and we must have detailed balance. So as I see it, the detailed balance in itself is no proof it doesn't work, but it's a way to make us understand why it doesn't.

And, I say there is an impedence mismatch because your nano device is, necessarily, nano. If your front latch worked, it would have worked because it was not being subjected to normal gas pressures inside the chamber, but only to an occasional molecule bump. Your idea hinged on that fact, I thought.

Yes, correct, though I don't understand the term 'impedance mismatch'.

That means that the output side of the chamber also does not have normal atmospheric pressure, but the occasional molecule bump.

This is where you go wrong. The reason why there is no pressure on the nano-hatch is that it is nano-sized, not some special property of the gas or anything. Pressure is a statistical property that applies only to 'large' objects. Such as the valve - but not the hatch.

One or two gas molecules at a higher energy state then the rest of the room, trying to fight their way out of a chamber against normal gas pressure.

What makes you think one or two molecules would have a higher energy state than the rest of the room? In fact, the molecules will all (on average) have the same energy state, no matter if they are inside or outside the box. If the device worked, there would simply be more of them inside.

 

[Merko]

But post 82 is the kind of inconsistent, biased wording that can lead to major mistakes in an intuitive or qualitative argument. Under one circumstance (letting molecules in) the door is a lightweight contraption that can get out of the way of a single incoming molecule without sending it rebounding and unlikely to ever contact two molecules simultaneously. Under another circumstance the door is capable of clearing an area of molecules.

I disagree, because this would only be relevant if we're trying to make a quantitative argument, and I was indeed only interested in the qualitative argument. In other words: I was not interested in how often molecules would push the door open and still enter, only that they did. I was not interested in how often a light bat 'clears' a heavier molecule on the inside, only that it sometimes does. If we have a 'positive' effect - no matter how small - then we do not need to know how likely it is, only that it exists. If it exists, we must either have a negative effect canceling it, or the net effect will be positive. This is why I was only satisfied after the mechanism with such a negative effect was found.

 

[RecoveringYuppy]

What makes you think one or two molecules would have a higher energy state than the rest of the room? In fact, the molecules will all (on average) have the same energy state, no matter if they are inside or outside the box. If the device worked, there would simply be more of them inside.

Velocities in a gas are distributed on a skewed bell shape curve. You'd expect a wide range of energies in the gas. That's why I've been suspicious of the door being vaporized.

ETA: http://www.chem.ufl.edu/~itl/2045/lectures/lec_m.html

 

[roger]

What makes you think one or two molecules would have a higher energy state than the rest of the room? In fact, the molecules will all (on average) have the same energy state, no matter if they are inside or outside the box. If the device worked, there would simply be more of them inside.

Okay, I think I see where we are miscommunicating. The 2nd law can be described using statitistical mechanics - a bunch of gas molecules bouncing around all have different velocities, and when we consider them statistically, it is a near certainty that in maco dimensions the energy will be evenly dispersed. Hence, you can't make a stand alone heat engine that sits in an empty room (I know I'm telling you what you already know - just making my position clear). Well, there's a tiny chance all the energetic molecules ends up in a pocket, so maybe once in a trillion years or so the heat engine will work. But, mostly not.

I saw your device as a way to collect those more energetic molecules. I must have misunderstood the principle you were taking advantage of. My thoughts were - we can't exploit the deviations in energy at a macro scale, but if we go to a nano scale, where the heat engine is exposed to single molecules at a time, we can exploit the deviations.

So, if you had a valve that was tuned to the average energy level of the room, it would admit any molecule moving faster than average. You end up with a heat collector, which you then offer to a heat engine to generate energy from the heat differential between these molecules and the room. You are only collecting a few molecules at a time - at a higher energy than the room.

The problem with that scheme is that we have no way to precisely tune that valve to the average energy in the room w/o expending more energy then we would subsequently gather. Hence, my questions about how much energy these (presumed) nano gates would consume. And, when you try to exhaust those molecules into the room, the billions of gas molecules already in the room will overwhelm (this is where my physics goes wonky) the system - millions of gas molecules with an average energy will come flooding into your collector, cooling the few energetic molecules already in there, and negating the differential. And no heat gradient means no heat engine.

I guess I don't see how the door would have created higher pressure w/o exploing that concept, but it's not important I guess, since we know the door will not work. Anyway, clearly I was following a tangent to your original idea.

 

[Merko]

Velocities in a gas are distributed on a skewed bell shape curve. You'd expect a wide range of energies in the gas. That's why I've been suspicious of the door being vaporized.

Well.. the problem is that if this was the case, we should expect a lot of materials to be vapourised. Since there are a lot of materials where some atom of a molecule has only one bond.

I saw your device as a way to collect those more energetic molecules.

Well, that's one way of seeing it, yes. But they will be slowed down by the door. They will also be slowed down on their way from the door to the engine or valve, because they will collide with other molecules (it's a LONG way, for a molecule to travel!).
The laws of thermal transfer mean that the gas on the inside will have the same temperature as on the outside. So it doesn't really matter how energetic our molecules were when they entered the gate, in the way you're thinking. They'll even out. And so, the only difference is that there would be more of them - with the same energy distribution. These would batter the valve more than the ones from the outside, because there were more of them. There is no dynamic 'tuning' needed.

 

[RecoveringYuppy]

Well.. the problem is that if this was the case, we should expect a lot of materials to be vapourised. Since there are a lot of materials where some atom of a molecule has only one bond.

Ah. I think I see where your confusion lies.

We do see things vaporizing all the time. On a macroscopic scale though it usually doesn't matter because a single atom disappearing once in a very great while doesn't destroy the whole object. And everyday things like water evaporating from a glass despite room temperature being below boiling point happen because of that energy distribution curve in the middle of the link I provided.

Now I think I understand why we talking past each other when I asked about active sensing back around post 15 or whatever and why the ranges I tried to break the energy distribution curve didn't impress you.

If you're thinking that most atoms in a gas have the same velocity and that the outliers are few and far between that's not even close to correct.

And there's a better reason to dismiss this device than thinking you've found a "negative bias" somewhere. The real reason to realize this device doesn't work is to realize that the positive biasing case doesn't really exist. There is only one exact energy where the "positive biasing" case works. Being one exact energy on that curve means that it represents 0% of the area under that curve. Around that single point in the continuum there is a range of energies where the device will open and close erratically. On either side of that range there are two ranges where the door will either open or close dependably. And on the far right there is a range of energies that will destroy it, it's only a matter of how long it will take to destroy it (and given the inertia ratio of molecules to doors we've talked about before I suspect it won't take long to destroy it).