Nanotech free energy

[Geckko]

The reason it wouldn't work?

The laws of Thermodynamics. Enough said.


If you want more detail, the force required to enter the chanbe will be equal to the force being exerted from inside the chamber. Energy required to get in will equal energy required to force molecule out (ignoring any leakage such as friction heat). Ergo, no net energy produced.

And another free energy machine bites the dust.

 

[DeviousB]

The problem with the machine in the OP is the door.

A molecule bumps into the door from the left, causing it to swing open and admit the molecule.

The door swings open because it has had energy imparted to it by the molecule. So where exactly does this energy go? If it is absorbed by the swing or the frame in some way (usually through heating), then entropy increases and the 2LOT is not violated, if we imagine a perfectly elastic door and spring (or doorframe and spring), then once hit by a molecule, the door will continue to swing open and shut and not function as an adequate partition between the two sides.

The design was first proposed by Maryan Smoluchowski in 1912, and is usually called "Smoluchowski's Trapdoor" rather than Maxwell's Demon in his honour. You can read a refutation of it here.

Smoluchowski was also the person who realised that atmospheric dust would cause Rayleigh scattering, hence answering the old, old question, why is the sky blue?

 

[RecoveringYuppy]

So what? I don't need such a guarantee. A molecule from the inside may exit, or it may not. At the same time, additional molecules may enter, or they may not.
What makes you think there will be more molecules exiting, than entering, in this way?

I think there will be the same number going each way.

For one thing, any molecule from the exterior that opens the door is going to be rebounding to the left after the collision not proceeding to the right in to the chamber. So the molecule that opens the door is virtually precluded from being the one that enters unless it engages in further collisions. Basically, after colliding with the door the gas molecule will now have some motion to the left. It's probability of it's motion changing due to collision is now exactly the same as any other gas molecule in or out of the chamber. It has no more chance of being the one through the door than any other molecule.

 

[roger]

The reason it wouldn't work?

The laws of Thermodynamics. Enough said.

True, it won't work, but I quibble with the enough said aspect.

People who think that free energy is impossible (not merko) think of the laws of thermodynamics as something we just made up, and there might be a way to violate them. I think they (the laws) are usually presented wrongly (for understanding).

Let me tell a fable to illustrate this.

Accountant Roger was working at firm XYZ. He'd always obeyed, and observed the Laws of Accountingdynamics, because after all, that's the way the universe worked. But one day the ledgers did not balance. They showed more money coming in then going out. Eureka!!! He ran to the CFO and CEO, telling them of the amazing discovery. Smiling, they told him to prepare 1 million identical ledgers. In a short time, the bank account of XYZ hit 1 trillion dollars, and Roger retired as the richest man in the world. The end.


Um, no. Why is this necessarily a fable. Well, there is no law of Accountingdynamics, as such. People didn't just observe that ledgers are balanced. We can use 'creative' math, or plain old mistakes, to make the books unbalanced, but that is nothing more than a mathematical model disagreeing with reality. If I write you a check for $1, but don't deduct it from my ledger, I don't have that $1 even though my book says I do. The dollar is gone, transferred from me to you. If I think otherwise, I'm quickly going to learn what the penalties are for writing bad checks. It has nothing to do with small minded lawmakers insisting I adhere to the "laws of accountingdynamics", it's because X-1 can never equal X. Basic math. The laws of accounting dynamic fall out from that fact. It's not an arcane rule somebody made up that might one day be violated. If I give you a dollar, I don't have it anymore. All the math mistakes in the world will not change that.

Similarly, apple farmers obey the laws of AppleDynamics. Imagine a farmer who came up with a scheme where he puts some apples in baskets,shuffle the baskets around, and claims that he ends up with a net gain in apples.

No one would think that would be possible, ever. But wait, the farmer says, don't be so close minded. It might be possible, and wouldn't it be wonderful if I did. Now, look at my apparatus, and tell me why it won't work. See, I put apples in these bushel baskets. Now, people take these bushel baskets, and go ride on a ferris wheel with them at the fair. Now, when at the top of the wheel, they throw them to people sitting at the bottom of the wheel. Sometimes they drop some, though. So when the ride is done, there are fewer apples. Now, the person who buys my apples pays for the apples before the ferris wheel, but picks them up after the ride is over. I collect the apples that have fallen, and do the same with them. Tada! Infinite apples!!

I don't have to explain how stupid that story is. The buyer would complain that she isn't getting all she paid for, and the total # of apples stays the same. No one would make the mistake of thinking this scheme worked (except through fraud). No apples are being created. This has nothing to do with the Law of AppleDynamics, but it's a consequence of simple math. Apple farmers get tired of doing this analysis for every possible apple transaction they ever do, so the result of this math is codified as the "Laws of AppleDynamics". It's just an easier way to thinkk about it. If I encounter a system where it seems that more apples come out then go into it, I know I made a math mistake or the system is open (another farmer is putting apples into the system when my back is turned).


Physics is a lot harder, and these machines are often ingenious, so the errors are often hard to find. But it's still just apples. If I give you an apple, I no longer have that apple. If I give you energy via photons, I no longer have that energy. A closed system will always sum to zero. The laws of thermodynamics codify the result of all those sums.

So people like merko asking questions like this is a good thing. Do enough energy diagrams of closed systems, and you start to see how the math works. If you think your unbalanced wheel, or whatever, achieves overunity, you've just done some creative accounting. Every such device yields to a correct energy analysis. That's why they don't work. X-X=0. It does not give you an excess of energy, any more then creative book keeping gives me an endless supply of money.

 

[DavidS]

Here is how I see the device in a classical world. Envision a tub of water. You have a closed cylinder. On one side of the cylinder you have a door with a spring, and on the other you have a turbine. Impose a vacuum inside of it. Stick it in the vat of water. Will it run? If so, for how long?

That's how I understand the problem, except that your drawing shows an arbitrary depleted initial state rather than the proposed running status which would have deeper water on the inside of the gate+turbine cylinder.

Running, the Interior -> Turbine -> Exterior part is analogous to a water wheel in a stream. While that's a part of the system, there's no difficulty with that part; can do if the interior level is higher than the exterior.

The larger system, Exterior -> Gate -> Interior -> Turbine -> Exterior is more analogous to a water wheel in a bucket in a lake -- pretty much what you drew. Its obvious that the classical water analogy won't work. A piece of water leaves Exterior where it had a certain amount of energy, passes through the Gate -> Interior -> Turbine section that sends some energy away as useful work, leaving the piece of water with less energy than it needs to push its way back to the Exterior. The device grinds to a halt when the bucket drains or fills to lake level (plus/minus gate/turbine static resistance).

The proposal (as I understand it) is that on sufficiently tiny scale it might be possible to dodge the constraints of "classical" thermodynamics because its inherent continuum presumption doesn't apply:

The Exterior fluid is no longer a continuum, it's a rabble of more and less energetic particles.

The Gate amounts to a "You must be THIS energetic to ride the Turbine" sign keeping the fitness-challenged particles out of the Interior waiting room (at least, keeping out more of them than their hyperactive brethren).

The Interior collects only (well, mostly) the more energetic particles which bully their way through the Turbine without (a) getting too tired to rejoin their lazier siblings on the Exterior, or (b) leaning against the Gate too much to let more particles into the Interior.

The "new magic" is in the Gate. Proving the impossibility of the Gate is sufficient to prove the impossibility of the device as proposed; the Turbine needs the Interior->Exterior gradient to output work. Note, however, that demonstrating feasibility of the Gate alone is *not* sufficient to demonstrate the feasibility of the device; it's also necessary to demonstrate that the Turbine need not offset the magic Gate with magic of its own.

IMO we should be able to refute this thing without appeals to "common sense" non-ideal effects like friction, heat transfer through or insulation of the chamber wall, inelasticity, hysteresis, etc. While such considerations would definitely be important to a practical implementation, refutation on those gounds is hardly robust; it would leave open the possibility of building the device after only incremental improvements in construction. Proving the device impossible with perfect implementation in ideal materials would close the door forever.

Come on, guys... this is a perpetual motion machine of the second kind. Hunting these things without license or limit is a time-honored tradition... why is it so hard to get this one mounted over the mantle?

 

[Cuddles]

The Gate amounts to a "You must be THIS energetic to ride the Turbine" sign keeping the fitness-challenged particles out of the Interior waiting room (at least, keeping out more of them than their hyperactive brethren).

The "new magic" is in the Gate. Proving the impossibility of the Gate is sufficient to prove the impossibility of the device as proposed;

I think RecoveringYuppie has this covered. A particle that hits the gate will be reflected back in the opposite direction. Once the gate is open, the particle that opened it is no more likely than any other to pass through, so (assuming equal pressures initially) the same number of particles are likely to pass in both directions. While the gate can only be opened from the outside, it is equally easy to travel in both directions.

 

[Merko]

Replace the flappy door with a revolving door to do the same function and you'll quickly see that you have a box with a turbine on each end.

Sorry, but I think this is my fault. The turbine on the sketch is a symbolic one. It wouldn't actually work if it looked like it does on the sketch, because the pressure on the 'front' side of the arms is always equal to the one on the 'back' side of the arms. In reality, turbines are dynamic things, which aren't too easy to draw however. A standard turbine symbol (EDIT: Link to image NOT showing a standard turbine symbol removed) wouldn't tell most people anything unless they are mechanical engineers. However, we know that turbines work. We don't necessarily have to use a turbine even. The problem is to get the overpressure, if we can get an overpressure, which is macroscopic by definition, we know that it can be turned to useful energy.

So, no, these two mechanisms are very different. The turbine would be a normal-sized one, by the way.

Why this isn't possible in real life seems really hard to understand, but seems to involve the fact that a one-way valve for atoms isn't possible. This paper seems to address the problem for actual, or at least possible, devices.

Then there is something I don't understand about my device. That paper appears (I can only read the abstract) to be the usual kind of Maxwell's Demon, where there is some 'sensor' to control a mechanism that opens and closes the gate. And so we have the information storage problem. That problem doesn't seem to be the issue with my device - but again, maybe I'm just blind.

There is actually a running thread on learning QM in this forum where I did some recommendations; you might, I think, find Vincent Icke's The Force of Symmetry very interesting, and quite valuable if you know your way around Newtonian and Galilean physics pretty well.

Thanks. I'll definitely look out for that one, and I'll skim the QM thread too some day.

in an enclosed space (ie the box) with an increasing number of particles wouldn't you have more and more collisions with the limiting wall and thus lose more and more energy here through transfer?

No, that's not how heat transfer works. There will be more pressure inside the box, but it will have the same temperature. But anyway, you assume that the device already works! Again - inefficiency is not a problem here. If we have an increasing number of particles, it's because it works!

BTW How many atoms do you think would go in to the construction of the door? You say the opening on the left is big enough for one molecule. Do you think you can arrange for the door to be exactly the same size with no overlap of the adjacent walls? If you have the door overlapping the walls a bit then you've got a situation where the door can only have one molecule hitting it from the outside yet multiple molecules hitting it from the interior.

The door can be made of a single atom, in theory. Not only doesn't it have to overlap, it can actually be made smaller than the opening. The only thing that is required is that it obstructs enough of the opening that gas molecules can't pass through when it is closed.

Anyway, it doesn't matter if the door sometimes cannot open because a molecule is hitting from the inside at the same time. Assuming that the gas is thin enough, this is unlikely. Even if the probability is very high, it is lower than 100% (and for a normal gas, it would clearly be very low). So again we have only a problem of efficiency.

It was mentioned before but what "powers" the 'spring'?

We could see the spring as a chemical bond. These are flexible, you know, but they have a 'preferred' (minimum energy) orientation. We don't need to swing the hatch 180 degrees, by the way. Just a very small deflection would be enough, if that is all that's needed to make the difference from 'impassable' to 'passable, given a near-perfect hit'.

Merko, could you expand on how that turbine works a bit better?

See above. It's a normal, macro-scale turbine. We know turbines work.

Your device may work for a few nanoseconds until you reach temperature equillibrium with the outside, but after that, the door will recieve more pressure from the inside (all those molecules plus the spring force) than the outside, and remain forever shut.

Again, pressure is only a statistical property, it doesn't exist on this scale. See my reply to DanishDynamite in post 9.

But even here, won't all the work end up being done on the spring, where you will lose all the energy to heat losses?

I regain it through heat transfer, see point 3 in the OP.

On the other hand, if you are exhausting into vacuum, which you seem to be assuming, then you have introduced a temperature differential. You will get energy, but in exactly the same way you get energy from a water dam or a windmill.

No, I assume that the outside room is the same on both sides of the box, obviously (see point 4 in the OP). It contains a gas, perhaps one with large, heavy molecules, to make the construction easier.

The enclosed box and turbine obsfucates this point, because we all forgot to analyze the forces on the turbine.

We can remove the turbine for now. If we then get an overpressure in the box, we can re-add it. Because we know that we can turn overpressure into useful, macro-scale energy. That is evidently not the problem.

Show me how a molecule has the ability to move all the way from the door, through the turbine, and out back into the room, despite losing energy to the door, spring, walls, and turbine.

Sure. Our molecule will lose energy (speed) hitting the door. This energy turns to heat in the door, which is transfered to the rest of the box, and on to inside and outside molecules. Our molecule regains this energy (speed) from collisions with the inside wall. It does so, because normal laws of heat transfer say that the wall will transfer heat between the gas on the outside and the inside, until they have the same temperature. Same temperature, means that the molecules have, on average, the same energy (speed). Our molecule then again loses energy (speed) in the turbine. This energy is converted in to electricity and used to solve the world's problems. So when the molecule exits, it will indeed be cooler, have less energy, less speed. That's the point of the device. It extracts energy from heat in an even-temperatured room, which is impossible according to the second law of thermodynamics.

Here is how I see the device in a classical world.

You can't think about it like that. When we scale it up, we get macro-scale effects back again such as pressure. The device absolutely requires being nano-scale.

If you want more detail, the force required to enter the chanbe will be equal to the force being exerted from inside the chamber. Energy required to get in will equal energy required to force molecule out (ignoring any leakage such as friction heat). Ergo, no net energy produced.

This is not the explanation. Again, this is covered in point 3 in the OP. The molecules regain this energy through normal heat transfer.

The door swings open because it has had energy imparted to it by the molecule. So where exactly does this energy go? If it is absorbed by the swing or the frame in some way (usually through heating), then entropy increases and the 2LOT is not violated, if we imagine a perfectly elastic door and spring (or doorframe and spring), then once hit by a molecule, the door will continue to swing open and shut and not function as an adequate partition between the two sides.

Once again, covered in point 3 of the OP.

The design was first proposed by Maryan Smoluchowski in 1912, and is usually called "Smoluchowski's Trapdoor" rather than Maxwell's Demon in his honour. You can read a refutation of it here.

Wow, thanks! However, that doesn't seem to be a refutation at all. It is an attempt to describe the problem in more detail. The author adds a few caveats here and there to explain that he doesn't think it will work, and a few hints of why that may be the case. But he doesn't actually explain any mechanism (as far as I understand) of failure. Additionally, while it is similar, it is clearly different, because it relies on different properties of the door. In my device, the door is closed by a spring force, such as that of a chemical bond. In that device, the door is left swinging, and is assumed to close by itself sooner or later.

For one thing, any molecule from the exterior that opens the door is going to be rebounding to the left after the collision not proceeding to the right in to the chamber.

Not necessarily. If the gas molecule is heavier than the door, and/or moves faster than the door will after collision, it can continue to the right.

So the molecule that opens the door is virtually precluded from being the one that enters unless it engages in further collisions.

But it will engage in further collisions, sometimes.

It's probability of it's motion changing due to collision is now exactly the same as any other gas molecule in or out of the chamber. It has no more chance of being the one through the door than any other molecule.

The probability of changing direction is indeed the same, but it has one advantage: it is close to the door. Imagine that the gas is very, very thin. The chance of a molecule being close to the door at any time is very, very small. If a molecule hits from the outside and opens the door, it may go through directly, or bounce around a bit. But simply because it must be close to the door, there is an increased chance that it will enter, one way or the other, compared to a random molecule on the inside. (EDIT: Or a random molecule on the outside.)

Physics is a lot harder, and these machines are often ingenious, so the errors are often hard to find. But it's still just apples. If I give you an apple, I no longer have that apple. If I give you energy via photons, I no longer have that energy. A closed system will always sum to zero.

That analogy works for the first law of thermodynamics, but not for the second law. This device also 'sums to zero', it doesn't break the first law. The second law is much, much more complex to prove, and in fact there isn't a watertight proof for it that shows it can't be circumvented by nano-scale mechanisms.

Of course, very few respectable scientists actually believe it can be circumvented. I also don't believe it can be circumvented. I don't believe that this device would actually work. But it is definitely not enough to say that it can't, because so far, it has not been proven.

 

[Modified]

Note, however, that demonstrating feasibility of the Gate alone is *not* sufficient to demonstrate the feasibility of the device; it's also necessary to demonstrate that the Turbine need not offset the magic Gate with magic of its own.

As Merko pointed out, if the gates work at all, then with enough chambers in series an arbitrarily high pressure differential could be achieved. Everything could be valved off until that differential was achieved, then the valves opened up to turn an ordinary macro-scale turbine. I think the feasibility of the device is made or broken on feasibility of the gate.

Edit: Oops, Merko beat me to it.

 

[DavidS]

For one thing, any molecule from the exterior that opens the door is going to be rebounding to the left after the collision not proceeding to the right in to the chamber. <snip> Basically, after colliding with the door the gas molecule will now have some motion to the left.

For a second I thought you had it there, but...

Devil's Advocate: "aren't you implicitly presuming that the colliding molecule doesn't bear momentum very much larger than the inertia of the gate? Consider a bowling-ball molecule colliding with a bowling-pin gate and your assertion's implications for safe beer-pitcher placement. It might not cost much more to use UF6 (Unobtanium hexafattide) vapor working fluid and a gate of LiHf (Lightweightium-Haffashevi) alloy."

It has no more chance of being the one through the door than any other molecule.

I suspect this is the right track, I just don't yet see how it's spiked to the ties.

 

[RecoveringYuppy]

DavidS,

Funny, I was thinking the extreme case of Uranium Hexaflouride and lithium myself.

Yes, I am making that assumption and I plan on discusssing whether it's a realistic assumption later. Mecko mentioned in his last post that he thinks his door can be a single atom lighter than the molecules of the gas. Keeping in mind that the door is attached to or somehow also a spring I really think the scenario is stretching credulity. Can the door/spring assembly really have so little inertia and yet not become a vapor of gas itself when it gets it hits by molecules?

 

[Modified]

We should probably consider the simplest case: two chambers with a gate between them, initially containing one gas molecule each.

 

[Merko]

Funny, I was thinking the extreme case of Uranium Hexaflouride and lithium myself.

Well, I don't want to get into the implications of mass-producing Uranium Hexafluoride for energy production at this stage..

Mecko mentioned in his last post that he thinks his door can be a single atom lighter than the molecules of the gas. Keeping in mind that the door is attached to or somehow also a spring I really think the scenario is stretching credulity. Can the door/spring assembly really have so little inertia and yet not become a vapor of gas itself when it gets it hits by molecules?

I think the scenario is credible. The bond can be fairly strong, even when there is only one atom in the door. We only have to balance the heaviness/temperature of the gas towards the spring force/mass of the hatch so that it can open and the gas molecule still pass - we don't need it to smother the hatch. Surely the chemical bond can be strong enough that it can be flexed, and still not break - this is the basis for many real-world materials, after all.

 

[RecoveringYuppy]

Merko,

Back in post 47 you replied to my point about the incoming molecule rebounding off the door with:

Not necessarily. If the gas molecule is heavier than the door, and/or moves faster than the door will after collision, it can continue to the right.

Yet in post 52 you're saying that that atom will need to be bonded to something else to achieve the spring effect. For purposes of analyzing the collision everything the "door" is bonded to is also going to count one way or another.

 

[Merko]

Yet in post 52 you're saying that that atom will need to be bonded to something else to achieve the spring effect. For purposes of analyzing the collision everything the "door" is bonded to is also going to count one way or another.

Which way?

It will obviously be bonded to some part of the wall. But we don't have to move the wall. Only the door. We also need to overcome the spring force, for example that of altering the orientation of a chemical bond. But the required force to do this can be made arbitrarily weak, and the device will still work (in theory).

 

[RecoveringYuppy]

I doubt you can make that arbitrarily weak and yet still have a device that won't simply evaporate when struck by something massive enough to pass through rather than rebounding.

Also, the weaker you make the spring the longer the door is going to stay open.

When I said "one way or another" I was referring to the spring. The components of the spring (and whatever the spring is bonded to) are all going to contribute some inertia to the door. Details will depend on the exact nature of the spring.

 

[roger]

However, we know that turbines work. We don't necessarily have to use a turbine even. The problem is to get the overpressure, if we can get an overpressure, which is macroscopic by definition, we know that it can be turned to useful energy.
<snip>
See above. It's a normal, macro-scale turbine. We know turbines work.

Okay, this is where I have to say 'show your work'. Tubines work when there is more pressure on one side of them than on the other.

How do you get and maintain positive pressure across the turbine? Why does the interior of your device have more pressure than the room it is in?

Sum all of the energy in the system. For example, you talk about the molecule losing energy from the door, but then getting heated (speeded up) through thermal distribution through the walls (to quote: "The molecules regain this energy through normal heat transfer."). But then you don't account for the fact that the molecules outside the device are now cooler (slower). You just moved your apples from one basket to another, but didn't write the number in your ledger. You have to include the energy loss to the molecules outside of the device in your energy budget. When you do that, the sum is 0.

ETA: to make my turbine objection clearer. The opposite side (outside of your device) of the turbine is getting bombarded with molecules. The molecules inside your device have to fight against that pressure to exit the box. How? Where does this extra energy come from? Show the math.

 

[DavidS]

Funny, I was thinking the extreme case of Uranium Hexaflouride and lithium myself.

Heavens, no. Uranium is just plain nasty stuff burdened by all manner of governmental laws, the commies have all the flourine tied up to poison our water supply, and if things keep going as they are I'll need the lithium for myself . Unobtanium, Fattine, and Lightweightium avoid all those problems. I admit that Haffashevi was just something I made up.

Again: I hope to see this cat skinned without getting bogged down in practical construction details. It "should" be impossible to construct from anything that obeys the physical laws we agree to accept as true, without regard to whether anybody really has what we want, what we want is what we need, or what we need actually exists to be had.

 

[Schneibster]

"Hiya. I'd like two perfectly rigid metal plates fifty miles long, each perfectly flat to within less than three nano inches."

"Right. Would that be the ones with the coefficient of temperature expansion of zero?"

LOL, yeah, that was the sort of thing I had in mind. Well done.

 

[Schneibster]

Okay, this is where I have to say 'show your work'.

I think this post has part of the solution to the conundrum.

Here's a look into my current thinking: since we are looking to prove that the device is impossible, instead of concentrating on the 2LOT, let's look at the 1LOT: energy conservation. If it seems to violate both, then we have at least a view into why it is impossible other than the 2LOT; from there, we might be able to specifically pinpoint the objection. I am, as I say, still working on this. I haven't spent as much time as I might, being busy with other things.

 

[DavidS]

Suggestion: Let's use the term "engine" rather than "turbine" to drop all the implementation baggage the latter suggests. Besides, it makes us sound more like real thermodynamicists.

I think this post has part of the solution to the conundrum.

Here's a look into my current thinking: since we are looking to prove that the device is impossible, instead of concentrating on the 2LOT, let's look at the 1LOT: energy conservation. If it seems to violate both, then we have at least a view into why it is impossible other than the 2LOT; from there, we might be able to specifically pinpoint the objection. I am, as I say, still working on this. I haven't spent as much time as I might, being busy with other things.

I don't think FLT can save us here. I can't imagine any FLT objection to the device so vile that it couldn't be replaced by an FLT-compliant SLT violation. If you think you've got one, tee it up.

If we're gonna require SLT compliance, let's just declare victory and go home: The proposed device isn't SLT-compliant. QED.