Perpetual Motion machine... it doesn't work...

[zidane1strife]

But why?

Since the particles are hit continuously from all sides, they move in a random fashion. For a long time, life was good with this explanation – that is, until Richard Feynman got his hands on the problem. Feynman suggested that we could build a device that took advantage of Brownian motion.

brownian ratchet

I know why the feynman ratchet fails to work, but I've a nice modification that seemingly works in my thought experiment. Yet I know it can't possibly be right so what is the cause of its failure?

Brownian Ratchet modification Thought experiment

From what I understand the problem is the ratchet and pawl(saw tooth wheel and little lock thingie) are subject to thermal vibrations too allowing for the possibility that the ratchet, despite being designed to move in just one direction, moves in the opposite direction. This movement in the opposite direction actually matches movement in the intended direction, stopping us from extracting any work. Good.

But a small modification can be made to rectify this problem. If the ratchet(saw tooth wheel) is made from a light magnetic memory alloy, and a small magnet is placed atop the lock mechanism(pawl), the teeth should expand slightly above that location as they get closer to the magnet(they'd contract later on as they get away from the weak magnet.) making it physically impossible for the ratchet to move in the opposite direction despite thermal vibrations. This will cause a movement to occur slowly in the intended direction, allowing us to perform work.

Since that's impossible, there must be an explanation to why such a design fails.

 

[DemoCoder]

Isn't this just a variation on Maxwell's Demon and will fail for similar reasons?

 

[KimB]

I don't see how that's any different from just having slightly larger teeth.

 

[zidane1strife]

Isn't this just a variation on Maxwell's Demon and will fail for similar reasons?

Brownian ratchets actually work, no knowledge acquisition is needed. At any single moment some particles will randomly strike and cause slight movement in one direction. The problem is the ratchet cannot obtain work since vibration causes the device to move at times in the intended direction and at times in the opposite direction.

Maxwell's demon needed to gain knowledge of the molecules, the ratchet does not.

I don't see how that's any different from just having slightly larger teeth.

It is, the teeth extend, making them slightly longer and making it impossible for them to go back(if all the teeth were too long/large the ratchet would not move at all, if they're only suitably large/long in one direction it will move in one direction.). They later contract as they get away from the magnet allowing them to undergo another cycle.

 

[KimB]

Maybe I'm not visualizing your idea completely, but I'm having a hard time seeing how this system could move one way if it can't move the other.

 

[zidane1strife]

Maybe I'm not visualizing your idea completely, but I'm having a hard time seeing how this system could move one way if it can't move the other.

Maybe that could be the flaw in the system.

The idea is that there's a small magnet above the pawl/lock mechanism. It is placed at just the right point so as to cause the tooth that's just passing/going-above the pawl/lock to extend slightly making it too long to go back in the other direction.(The ratchet's material is such that it takes a while to extend, and after doing so it takes a while to return to its original shape.)

Everytime the pawl vibrates it won't be able to cause movement in the opposite direction since the tooth above is too large, but brownian motion will cause movement in the intended direction since the tooth below is small enough to pass the lock(and it takes a short period of time for the magnetic force above to cause it to extend, and make the movement irreversible.)

 

[DemoCoder]

The brownian ratchet would undergo brownian motion itself jittering back and forth, how do you propose to bias it in one direction? The "ratchet" is handwaved away without specifying how it is constructed and operated. How do you prevent brownian motion from making it slip backwards?

Here's a debunk from sci.physics

You are describing what is sometimes known as a ``Brownian Ratchet.''
The problem with this naive and often-proposed idea is that because
equilibrium statistical mechanics satisfies the ``Principle of Detailed
Balance,'' the thermally excited transition rate depends on the barrier's
_HEIGHT_, not its shape. Since the height of the barrier is the same
in both directions, the rate of "forward" and "reverse" transitions are
necessarily equal to each other, so on the average the ``Brownian Ratchet''
executes an UNBIASED random walk.

To bias the motion of the ``Brownian Ratchet,'' one must establish and
maintain a NON-thermal-equilibrium situation (e.g., by introducing an
external ``bias'' force to ``tilt the energy landscape'' of the ``ratchet'').
Since thermal equilibrium is the state of ``maximum entropy'' for a given
amount of energy, to maintain a non-equilibrium state necessarily requires
_DISSIPATION OF ENERGY INTO ENTROPY_ --- and the amount of work one can get
out of the ``Brownian Ratchet'' will always be less than the amount of
energy being dissipated to maintain the non-equilibrium state, as shown
by the Fluctuation-Dissipation theorem.

We see an example of the above in the dynamics of the protein kinesin
when it crawls along cellular microtubules. Kinesin only executes a
biased thermal motion when in the presence of the cellular energy-
transport molecule ATP, which provides the external source of energy
required to allow the kinesin molecule to ``rectify'' its thermal motion
via a non-equilibrium cycle of conformation changes. The external energy
provided by the ATP is dissipated into heat --- it cannot be recovered.

I do see similarities to MD. In MD the "ratchet" is the demon, and he biases movement of the molecules, however, since any demon has finite information storage, it must eventually erase its measurements, thus performing work, and dissipating energy.

In the brownian ratchet scenario, your ratchet is either unworkable (unbiased, and therefore, does not do what is claimed), or, it is biased in one direction. This bias mechanism either requires energy (thus performing work or being a net consumer), or perhaps, or stores energy somehow (heats up during use) for which it has a finite reservoir and will eventually fail.

Brownian ratchets don't work. Brownian motors have been built, and biological examples exist, but they do not exist in a uniform temperature bath without any external energy input, which is how the proposed ratchets are supposed to work, extracting net work from a uniform reservoir and thus violating the 2nd law.

 

[zidane1strife]

I do see similarities to MD. In MD the "ratchet" is the demon, and he biases movement of the molecules, however, since any demon has finite information storage, it must eventually erase its measurements, thus performing work, and dissipating energy.

In the brownian ratchet scenario, your ratchet is either unworkable (unbiased, and therefore, does not do what is claimed), or, it is biased in one direction. This bias mechanism either requires energy (thus performing work or being a net consumer), or perhaps, or stores energy somehow (heats up during use) for which it has a finite reservoir and will eventually fail.

Brownian ratchets don't work. Brownian motors have been built, and biological examples exist, but they do not exist in a uniform temperature bath without any external energy input, which is how the proposed ratchets are supposed to work, extracting net work from a uniform reservoir and thus violating the 2nd law.

It is possible it heats up and eventually breaks, but the ratchet material is allowed to regain its shape and relax while it moves through most of the cycle, as it is not being affected by the magnet(it would be interesting to know how long it'd last and given it must be impractical, the reasons why it is so with any possible material used.). Normally one can use a permanent magnet to lift an object, and later apply energy to remove the object stuck to the magnet(or influenced by it in the case of iron fillings within a container), and reuse the magnet. We're using the collisions of a few molecules as the applied energy to move an object, that's being stretched a bit by a magnetic field, away from this field.

The mechanism I presented biases motion by making the tooth bigger after passing the lock/pawl, due to the effects of a weak magnet being placed nearby, on the light magnetic memory alloy tooth of the ratchet. The thought experiment modification to the brownian ratchet, causes a change in the size of the tooth, that makes it physically impossible for it to move in the opposite direction in spite of brownian jitter.

 

[K.I.L.E.R]

What's the point of perpetual motion if you can just get a human to spin the wheel every so often?

 

[Mintmaster]

DC, the problem with that debunk is that it's talking about single transfers of energy in its statistics. The reason a ratchet is different is that in one direction you don't need to exceed the energy barrier all at once. Big energy jumps make the ratchet go either way, but small ones occasionally let it go only one way. Well, at least theoretically, anyway. What one needs to do is explain why it's impossible to make such a device with different transition rates in each direction, but I don't know how.

Maybe the statistics just work out in some way when you take all this into account. The WKB approximation for arbitrary barriers would suggest so. It just seems like even if you had a chattering ratchet hook due to thermal energy, its interaction with the vibrating ratchet teeth wouldn't be symmetrical.

zidanestrife, your idea is interesting. It wouldn't stop the ratchet from moving back entirely, but rather just increases the barrier height on one side and reduce that transition rate. The problem with your system is that there's some energy loss because the energy stored in the memory material's field is lost without recovery when relaxing. But what if electrostatics were used instead?

So the question now is the following: Is there any intrinsic energy required in raising and lowering a potential barrier?

 

[zidane1strife]

DC, the problem with that debunk is that it's talking about single transfers of energy in its statistics. The reason a ratchet is different is that in one direction you don't need to exceed the energy barrier all at once. Big energy jumps make the ratchet go either way, but small ones occasionally let it go only one way. Well, at least theoretically, anyway. What one needs to do is explain why it's impossible to make such a device with different transition rates in each direction, but I don't know how.

Maybe the statistics just work out in some way when you take all this into account. The WKB approximation for arbitrary barriers would suggest so. It just seems like even if you had a chattering ratchet hook due to thermal energy, its interaction with the vibrating ratchet teeth wouldn't be symmetrical.

agree.

zidanestrife, your idea is interesting. It wouldn't stop the ratchet from moving back entirely, but rather just increases the barrier height on one side and reduce that transition rate. The problem with your system is that there's some energy loss because the energy stored in the memory material's field is lost without recovery when relaxing. But what if electrostatics were used instead?

Well you can use other things too, it was just a basic thought experiment. I do think it'd stop the ratchet from moving back, though, if the magnet's strength was such it lengthened the tooth enough(we know that with long enough teeth the ratchet won't budget in either direction, cause the pawl can't physically go over a long enough tooth.), it'd simply be too long for the pawl to go over it(it'd move slightly back but the pawl would not be able to physically get over the tooth that's gotten too long, making the movement irreversible in one direction.). Well there's energy loss, like when you get iron fillings away from a magnet, or get something stuck to a magnet away, that doesn't stop you from reusing the object/fillings and the magnet over and over again, for countless years. Would the permanent magnet loose its charge first, or would the ratchet portion of the device break first? Using any type of magnet and memory alloy material or electrostatics, it must be impossible to design a device that can extract more useful energy prior to breakdown or energy loss than was put into it.

So the question now is the following: Is there any intrinsic energy required in raising and lowering a potential barrier?

Well lowering a potential barrier is easy, catalysts do it, raising it would require the catalyst's function to be deactivated by something else, which can also be arranged.

(I'm using the diagram presented in the linked page as a guide.)
If the pawl/lock is vibrating due to multiple collisions it should tend to exclude some molecules from colliding with the upper portion of the bottom tooth and the bottom portion of the upper tooth. That is in the bottom tooth more molecules collide in the bottom than in the top portion of it, which is compensated by the pawl/lock collisions above. The opposite happens in the tooth on top, since the pawl is beneath this time.

One could make the ratchet teeth be like some sort of engineered protein catalyst, aka enzyme. There'd be two binding sites in each tooth shaped protein enzyme. One on top and one below, a molecule that got into the bottom site would cause a conformational change that'd make the tooth shorter(so it'd be easier for it to cross the pawl/lock.). If the molecule got to the site on top it'd cause a conformational change that'd make the tooth slightly longer. So the teeth would be getting longer and shorter all along the ratchet but something special would happen around the pawl/lock. Since the pawl and its vibration impedes some molecule collisions on the sides of the tooth colliding with it(those right next to it, above and below.), it causes the bottom tooth to be statistically more likely to be slightly shorter while causing the upper tooth to be slightly longer(due to changes in the number of molecules binding in each of the two sites of these two tooth shaped enzymes), giving bias to the motion.

 

[KimB]

Well, the extending of the teeth would be thermodynamic behavior as well, so it certainly could go back.

And bear in mind that there is essentially no difference between the pawl being right at the peak of the tooth and about to fall, and at the bottom of the other side in terms of how long the tooth would be (because if temperature allows the pawl to get that high, it's sometimes going to bounce just as high).

 

[zidane1strife]

Well, the extending of the teeth would be thermodynamic behavior as well, so it certainly could go back.

Yes it(the elongated tooth) could go back to its former shorter shape, on the enzyme example(though it should still statistically favor one direction), but I just don't see how it could do so under the influence of a magnetic field. It's probable but just extremely unlikely, that'd be like iron fillings changing the patterns they've just acquired under the influence of a magnetic field, while still being under said influence. It probably can happen, but I've never seen it happen, so it must be extremely unlikely.

And bear in mind that there is essentially no difference between the pawl being right at the peak of the tooth and about to fall, and at the bottom of the other side in terms of how long the tooth would be (because if temperature allows the pawl to get that high, it's sometimes going to bounce just as high).

It could go high, but if the tooth's length has increased under the influence of the magnetic field(something that normally shouldn't revert as long as the tooth is being influenced by the field), it will create a physical barrier that'll stop it from reversing the movement.

 

[3dilettante]

What keeps the magnet from expanding the teeth on both sides?

In that case, the difficulty increases equally for both directions.

 

[zidane1strife]

What keeps the magnet from expanding the teeth on both sides?

In that case, the difficulty increases equally for both directions.

distance. The one below is far enough, most of the time that it can shrink back should it happen to extend while getting closer to the magnet. The one above is at all times just close enough to the magnet, to always be at least partially influenced by the magnetic field and thus not able to shrink enough, thus the physical barrier remains.

 

[OpenGL guy]

Magnets are "cheating" as you are adding a source of energy. Over time, the magnet will wear out and will need replacing. Creating a magnet is also not free, but you may be able to find enough natural magnets for a good, long time.

 

[KimB]

Magnets are "cheating" as you are adding a source of energy. Over time, the magnet will wear out and will need replacing. Creating a magnet is also not free, but you may be able to find enough natural magnets for a good, long time.

Er, I don't believe magnets are a source of entropy (which is what's needed). I mean, sure, their entropy is lower than the same material that is not magnetized, but you can't usually make direct use of that entropy (and not at all by the system described).

 

[NANOTEC]

But why?



brownian ratchet

I know why the feynman ratchet fails to work, but I've a nice modification that seemingly works in my thought experiment. Yet I know it can't possibly be right so what is the cause of its failure?

Brownian Ratchet modification Thought experiment

From what I understand the problem is the ratchet and pawl(saw tooth wheel and little lock thingie) are subject to thermal vibrations too allowing for the possibility that the ratchet, despite being designed to move in just one direction, moves in the opposite direction. This movement in the opposite direction actually matches movement in the intended direction, stopping us from extracting any work. Good.

But a small modification can be made to rectify this problem. If the ratchet(saw tooth wheel) is made from a light magnetic memory alloy, and a small magnet is placed atop the lock mechanism(pawl), the teeth should expand slightly above that location as they get closer to the magnet(they'd contract later on as they get away from the weak magnet.) making it physically impossible for the ratchet to move in the opposite direction despite thermal vibrations. This will cause a movement to occur slowly in the intended direction, allowing us to perform work.

Since that's impossible, there must be an explanation to why such a design fails.

What's the difference between a BR and an alternator?

 

[Simon F]

Isn't this just a variation on Maxwell's Demon and will fail for similar reasons?

I don't see how that's any different from just having slightly larger teeth.

Is that the demon (see Decocoder) or the ratchet?

 

[Jabbah]

I would have thought that even if you could build a perfect ratchet that would work in the desired way then perpetual motion would still not be the result. Wouldn't the system just cool down and stop?