Could a Bar Magnet on Feynman's Ratchet Generate AC Current?

  • Context: Graduate 
  • Thread starter Thread starter cragar
  • Start date Start date
Join the discussion
Registration is free. Ask a follow-up in this thread, or start your own.
Sign up Log in
18 replies · 3K views
cragar
Messages
2,546
Reaction score
3
Lets say we take the ratchet off and put a bar magnet on the end of the shaft. So if the paddle wheel moves back and fourth it will rotate a bar magnet inside a copper coil. Could this produce ac current and I can't seem to figure out why this violates the 2nd law of thermodynamics.
 

Discussion

Physics news on Phys.org
So you are saying that i can't power a light bulb with this setup.
 
Ok, I am not quite sure yet, why It doesn't work, You say the net current will be zero, ok i get that . But when the magnet moves one direction, will it cause electrons to move in the copper wire? Are you saying the magnet won't move at all. When the molecules hit the paddle wheel do they lose some momentum and cool down. And then they won't have enough energy to overcome the friction of the bearing .
 
Ok, something that needs to be made explicit: all the components of the system (gas, ratchet/pawl, magnet, lightbulb, etc) are at the same temperature. Otherwise, you are allowed to heat a (cold) device from a hot reservoir.

No, although the fluctuation-dissipation theorem allows for momentary voltages and currents to exist, averaged over time, you cannot extract work in this way. Put another way, since incandescent bulbs work by heating a filament, you cannot refrigerate the gas by heating the filament.
 
Ok i think i see what you are saying . Thanks for your answer by the way.
 
The whole idea of the 2nd law is that there is no tendency of the particles to move in any particular way. For every particle moving one way, there would be one moving the opposite way.

The magnet will not swing since every time it is hit by a particle from one side, it will also be hit from the other in a nanosecond, and there is no way to get anything useful from the extremely small variations (which cannot be noticed).

You can't say "I have 5 super-massive particles at high temperatures in a large box" because the 2nd law statement is strong for large systems. It's all probability.
 
I don't see it, though it doesn't look like Feynman's paddles, paw and ratchet anymore. There is a hot box at temperature T_h, and a cold box at temperature T_c<T_h. Particles in Brownian motion hit the paddles generating current in the coil via a magnet. The wires from the coil run to the hot box where a resistance heats the hot box hotter.

Why doesn't this happen?

-------------------------------------------------------------------------

On second thought, Faraday's law of induction, and the second law of thermodynamics, together, predict thermal noise current.
 
Last edited:
cragar said:
Lets say we take the ratchet off and put a bar magnet on the end of the shaft. So if the paddle wheel moves back and fourth it will rotate a bar magnet inside a copper coil. Could this produce ac current and I can't seem to figure out why this violates the 2nd law of thermodynamics.

Let's simplify it even further by making the ratchet a coil that rotates in a magnetic field.
 
I think htis is the problem:

http://en.wikipedia.org/wiki/Johnson%E2%80%93Nyquist_noise"
 
Last edited by a moderator:
And is there a way we could shape the paddles on one side so that the momentum transfer is greater on one side of the paddle than on the other, like a pelton wheel or something. Maybe have one side that is raised with an angle to it . Not sure if this will work.
 
cragar said:
And is there a way we could shape the paddles on one side so that the momentum transfer is greater on one side of the paddle than on the other, like a pelton wheel or something. Maybe have one side that is raised with an angle to it . Not sure if this will work.

If you immerse a rigid body of any shape in a fluid of isotropic and uniform pressure, the net force acting on it is always zero. I think the same holds for the fluctuations in the net forces, but I cannot prove it right now.
 
Phrak said:
I don't see it, though it doesn't look like Feynman's paddles, paw and ratchet anymore. There is a hot box at temperature T_h, and a cold box at temperature T_c<T_h. Particles in Brownian motion hit the paddles generating current in the coil via a magnet. The wires from the coil run to the hot box where a resistance heats the hot box hotter.

Why doesn't this happen?

That *can* happen, because you are transferring heat from the hot reservoir to the cold reservoir. Once the two boxes are at the same temperature, your system grinds to a halt.
 
cragar said:
And is there a way we could shape the paddles on one side so that the momentum transfer is greater on one side of the paddle than on the other, like a pelton wheel or something. Maybe have one side that is raised with an angle to it . Not sure if this will work.

No, and this is also discussed in the gear-and-pawl mechanism. Furthermore, a key concept is that the pawl does not collide elastically with the gear. If it did, the pawl would not work and the gear could freely turn in either direction.
 
The idea to take away from the ratchet and pawl, as Feynman points out, is that any mechanism sensitive enough to be affected by probabilistic fluctuations is also affected (in the way you don't want) by fluctuation on the receiving end where you are trying to collect free work. You CAN reduce the negative effects by cooling the other side (i.e. Feynman says you can cool the pawl) but then you have a temperature difference and there is no contradiction with the 2nd law.

This is actually intuitive since there is no particular bias in the random fluctuations to begin with, so no mechanism can take anything useful out of this (unless the mechanism is "intelligent" but then we have a whole other issue)
 
Andy Resnick said:
That *can* happen, because you are transferring heat from the hot reservoir to the cold reservoir. Once the two boxes are at the same temperature, your system grinds to a halt.

I think you skimmed to quickly. I've recited the scenario where the paddle wheels are in the cold box and the electric filament is in the hot box.
 
Phrak said:
I think you skimmed to quickly. I've recited the scenario where the paddle wheels are in the cold box and the electric filament is in the hot box.

That doesn't change my answer- one part of the system (the 'hot box') can transfer heat to the other subsystem (the cold object) and in so doing, extract work. When everything is at the same temperature, that is no longer possible.