Why wouldn't this device work?

[GotMex?]

I wasn't sure where to post my question but it seems like it would belong here. I was hoping for an anti-physics category or something hehe. I apologize in advance if I was wrong. Anyway, I've been reading about perpetual motion machines and I understand the physics behind them and why they are impossible.

I found a lot of examples of machines and enjoy figuring out why they won't work. There is just one example I found that I'm still not entirely sure why it won't work. Here's a link to an image describing it.
http://www.kilty.com/graphics/magnet.gif"

Wikipedia gave it as an example but didn't give an explanation for why it wouldn't work. The other sites I found gave confusing explanations as well. Anyone have any experience with this example and can tell me why it wouldn't work? Thanks.

 

[Danger]

It's a blatent violation of the 2nd law of thermodynamics. There is no way that the top magnet will be able to pull the ball back up.

 

[DaveC426913]

It's a particulary bad example (despite scoring points for simplicity) because the flaw is so obvious: why would the ball fall down in the first place?

 

[GotMex?]

It's a particulary bad example (despite scoring points for simplicity) because the flaw is so obvious: why would the ball fall down in the first place?

That's sort of what I was thinking. If the magnet is strong enough to pull it up the ramp, then why wouldn't it be strong enough to keep it from falling.

 

[Danger]

Bingo, buddy! And don't let anyone try to tell you that it's an electromagnet that only comes on at certain times, because it still wouldn't be practical without an external source of energy for the magnet.

 

[Tido611]

If the ball started at the bottom of the ramp, in the little holder and it began to move due to the magnetic force from the magnet then that means that the force of gravity is weaker then the magnetic force moving the ball closer and closer to the magnet, also increaseing the magnetic force so there is no chance the ball would make it down the hole in the top of the ramp.

 

[DaveC426913]

In defense of the poor PM device, I think the rationale is thus:

The ball falls down the sloped ramp. Because the ramp is extremely steep at the top, the ball is virtually in freefall, enough to overcome the magnetism.

Once at the bottom though, the ball is stopped abruptly, losing all its momentum.

Now, since the ball is supported on the diagonal ramp, it doesn't require as much effort for the magnet to lift it.


At least, that's the rationale. It's still flawed of course.

 

[sdemjanenko]

It would work except once the ball falls the lower ramp is steeper so in order to meet up with the straight ramp it must eventually have a slope less than that of the straight ramp. At the point, or even before it reaches the same slope the ball would get caught by the magnet and stuck in place.

 

[leright]

yeah, the ramp is steeper on the downward ramp than the upward ramp.

 

[sdemjanenko]

its only steeper for a portion, then it turns less steep since its a curve, not straight

 

[DaveC426913]

its only steeper for a portion, then it turns less steep since its a curve, not straight

Yes, but by that point the ball has plenty of momentum.

 

[rcgldr]

If the path was a loop, and there was no loss of energy to friction, then you wouldn't need the magnet. The only near frictionless systems are orbits of planets, moons, and stars, since space is almost a pure vacuum. The other scenario is current flowing through a loop of superconducting wire.

 

[Gokul43201]

Anyone have any experience with this example and can tell me why it wouldn't work?

No experience with this example, but I can tell you why it will not work (only to subsequently find someone else come up with a much simpler and more elegant explanation).

It will not work because you can not build the system without friction. In the absence of friction, it (or some form of it) would work just fine, and as pointed out by Jeff, wouldn't even need the magnet.

However, the magnet does not "offset" the losses due to friction.

Quick reason: the magnetic field is conservative, unlike friction.

Longer explanation:

Magnetic and gravitational fields are conservative, i.e, the work done by these fields is independent of the path. As a result of this, the work done by these fields on a particle, over the course of a loop is zero. (See this by writing the path integral from A to A as the sum of two path integrals from A to B and again from B to A; these integrals are equal in magnitude and opposite in sign, irrespective of the actual paths from A to B and back, and cancel each other off). So, after each loop, while there's a loss of energy from dissipative forces, there is no gain of energy from the magnetic field.

Still longer, and more rigorous proof:

1. Pick any point (call it A) where the velocity of the ball is v(A;n) during the n'th looping of the loop.

2. Watch the ball do a loop (assuming it can; if the ball can not do a loop there's nothing left to prove) and come back to A. During the course of this loop, the total work done by the conservative forces (gravitational + magnetic) is zero. The work done by the non-conservative forces (friction, air resistance), must hence equal the loos of KE of the ball (ref: Work-Energy Theorem). The ball thus has a smaller velocity when next it arrives at A. i.e, $v(A;n+1) \leq v(A;n)$

3. One could still argue that as the ball gets slower with each successive loop, the loss of KE also gets smaller, and the velocity asymptotically approaches a terminal value. However, since air resistance is proportional to velocity, a non-zero velocity will imply a non-zero drag and hence a non-zero loss to the KE. So, the only possible asymptotic value, if one exists, is zero. (For the specific case of the ball on the ramp/hoop, the loss of speed is faster than asymptotic. One can show that there's a non-zero lower bound on the work done by rolling friction, and hence, a finite upper limit on the number of loops before the KE reaches zero at A).

4. Now we make use of a specific flaw in this system. Since the magnetic field has a spatial variation and the gravitation field does not (or it has a much smaller spatial variation), we can safely conclude that that the net field is not zero everywhere on the loop. This in turn implies that the net potential energy is not a constant during the motion, and specifically, there must exist a pair of spots where the net PE are respectively a maximum and a minimum for the loop.

5. If the PE at A is smaller than the maximum PE by some P' > 0 (which we can ensure by our choice of A), we need a minimum KE at A given by: KE(A,min) = PE(max) - P', which then imposes a minimum speed required at point A, v(A,req). Now since v(A) at best approaches zero asymptotically (with the number of loops completed, n), we can always find an n=N when v(A;N) is smaller than any chosen number. Specifically, there exists some N, where v(A;N) < v(A,req). At this N, the ball has insufficient KE to reach the point of maximum PE and hence fails to complete the loop.

Since N is finite the motion is not perpetual.

 

[DaveC426913]

In the absence of friction, it (or some form of it) would work just fine, and as pointed out by Jeff, wouldn't even need the magnet.

Uh, can you just elaborate on this? You're saying that without friction, this device would operate just fine? How do you figure that? Gravity is still in effect.

 

[rcgldr]

Uh, can you just elaborate on this? You're saying that without friction, this device would operate just fine? How do you figure that? Gravity is still in effect.

Similar to an orbiting object, if the ball could roll in a loop without friction, and without any extrenal forces it would roll in the loop forever at a constant speed. Adding gravitational or magnetic forces would vary the speed, but without friction losses, the total energy would be conserved, and you'd just have variation between potential and kinetic energy.

I don't the device as pictured would work, even without friction, but if the ramp curves were modified to be more loop like, it could work. Instead of a hole, if there was a top to the ramp to guide the ball downwards, and at the bottom, a guide to loop the ball back upwards, then it could work, if there was no friction. If a hole at the top were to be used, then the hole would need to be far enough away from the magnet that it wouldn't stop the ball. The shape of the bottom ramp would have to loop the ball back around and upwards on the upper ramp.

 

[J77]

If the magnet were powerful enough to pull the ball back up the slope (with enough strength to easily get past the hole)...

and if the hole were big enough for the ball to fall down without skipping over the hole...

and if gravity overcame the magnetic pull and the ball could drop to the start...

Surely, the only way for this motion not to be perpetual is that the magnet would eventually lose strength!

I don't see how friction comes into the argument unless the magnet induces a field such that the ball only just makes it to the hole when the set-up is new - or, likewise, after the magnet loses strength to be at the point where the ball just makes it to the hole.

 

[bomba923]

Surely, the only way for this motion not to be perpetual is that the magnet would eventually lose strength!

I don't see how friction comes into the argument unless the magnet induces a field such that the ball only just makes it to the hole when the set-up is new - or, likewise, after the magnet loses strength to be at the point where the ball just makes it to the hole.

As explained earlier, magnetism is a conservative force and thus performs zero net work on any mass traveling in a closed path (i.e., where its net displacement is zero). To offset the negative work done unto the mass by the non-conservative force of friction, the magnet must perform positive work unto the mass. Since magnetism is a conservative force, this simply cannot occur within a closed path.

 

[J77]

Is it a closed path?

 

[bomba923]

Is it a closed path?

Isn't it obvious?
The ball starts at the bottom of the ramp. It travels up the ramp and drops through a hole, after which it rolls down a curve back to...the bottom of the ramp.

The starting and ending positions coincide, and hence you have a closed path
(as the displacement between two coinciding points is always zero).

 

[J77]

Isn't it obvious?

No, it's not obvious to me that the magnetic field acts in a continuos way along this path.

ie. once the ball drops through the hole - would it not be possible to shield it form the magnetic field before it appears at the bottom of the ramp once more?

 

[Farsight]

Has anybody mentioned Steorn? I read something in the paper about it, and the journalist reckoned it was some ad campaign spoof involving Microsoft.

http://en.wikipedia.org/wiki/Steorn

"Steorn Ltd. is a company based in Ireland that claims to have developed a "free energy technology". In August 2006 Steorn placed a full-page advertisement in The Economist, issuing a challenge to scientists to review their invention, which appears to be a perpetual motion machine..."

I'll be real interested to see how that one turns out.

 

[Danger]

I saw something about it on either Gizmag or How Stuff Works. Don't hold your breath.

 

[russ_watters]

No, it's not obvious to me that the magnetic field acts in a continuos way along this path.

That isn't what bomba923 said. He said it is a closed path (and system). All that means is that the ball passes over the same points every trip around and never leaves the path and no new ball ever gets put on the path.

The force most certainly is not the same everywhere on the path but the fact that the path is closed tells you that it cannot possibly work. The starting and ending postions are the same point and 0=0. Don't let the paths fool you - that's what fools most people who think they've discovered perpetual motion: if you know the starting and ending points, the path is pretty much irrelevant.

In fact, knowing how the laws of thermo work would solve a lot of problems for such people: If you understand ahead of time that the path is irrelevant, you won't waste your time looking for a better path.

ie. once the ball drops through the hole - would it not be possible to shield it form the magnetic field before it appears at the bottom of the ramp once more?

Not without doing work.

 

[russ_watters]

Has anybody mentioned Steorn? I read something in the paper about it, and the journalist reckoned it was some ad campaign spoof involving Microsoft.

http://en.wikipedia.org/wiki/Steorn

"Steorn Ltd. is a company based in Ireland that claims to have developed a "free energy technology". In August 2006 Steorn placed a full-page advertisement in The Economist, issuing a challenge to scientists to review their invention, which appears to be a perpetual motion machine..."

I'll be real interested to see how that one turns out.

It has been discussed in S&D. It is fraud (scientific and economic), though I doubt the owners of the company will be prosecuted for anything other than tax evasion.

 

[J77]

Not without doing work.

What form would this work take?

 

[Gokul43201]

ie. once the ball drops through the hole - would it not be possible to shield it form the magnetic field before it appears at the bottom of the ramp once more?

What do you mean by "shield it", and what kind of magnetic shield do you know makes a discontinuity in the magnetic field, and where would such a discontinuity exist?

True, it is not obvious that a magnetic field is conservative, especially considering the result of Ampere's Law or the working of a DC motor. But in both cases, any net work would be done by the current density that needs to be set up insode the loop, not by the magnetic field. But of course, work done by a current is not free - it takes work to drive the current in the first place.

 

[J77]

What do you mean by "shield it", and what kind of magnetic shield do you know makes a discontinuity in the magnetic field, and where would such a discontinuity exist?

So when it drops through the hole, the magnetic field is reduced - are there not materials which deflect the field in different ways?

I'm just using my imagination here.

You know, thinking outside the box :tongue:

 

[Gokul43201]

So when it drops through the hole, the magnetic field is reduced - are there not materials which deflect the field in different ways?

Passive shields (high permeability or Meissner) do not create a discontinuity in the field. At best they create an exponential decay of the field in a small region of space. The only way for the ball to be shielded at one spot and not at another is for it to have to traverse this space. The ball will hence, not see any discontinuity in the field.

Do you see a way that Maxwell's Laws permit a discontinuity in the field at any place where there no current density?

 

[DaveC426913]

I don't understand the issue here.

Even discounting friction, this device will not go anywhere.

If the magnet is weak enough to let the ball fall through the hole when it's closest, then there is no way it's going to be strong enough when it's farthest from the magnet to act upon it.

The ball will drop to the bottom by gravity and stay there.

 

[Gokul43201]

The ball will drop to the bottom by gravity and stay there.

Now, that would violate energy conservation wouldn't it? Where did its initial energy go?

 

[DaveC426913]

Now, that would violate energy conservation wouldn't it? Where did its initial energy go?

If I drop a rock on an ice rink, and it stays there by gravity, does that violate conservation of energy? The energy goes into heating and deforming the ball and ground when it comes to a sudden stop.

I'm trying to figure out why anyone thinks the ball would go back up the slide. It would just sit at the bottom.

 

[rcgldr]

Even discounting friction, this device will not go anywhere. If the magnet is weak enough to let the ball fall through the hole when it's closest, then there is no way it's going to be strong enough when it's farthest from the magnet to act upon it. The ball will drop to the bottom by gravity and stay there.

To make this simple, replace the mechanism with a circular loop, like a pair of circular rods that the ball rolls on.

Start off with this system free-falling in outer space, with no friction in the system. Then with any speed, the ball will travel along the path of the loop. Now add a force, like gravity, if the ball's initial position and speed aren't enough, the ball ends up rolling back and forth, but with enough initial total energy, the ball will travel along the path of the loop. Add a magnet on top, and if the magnet isn't too strong, you reduce the amount of initial total energy required for the ball to travel around the loop. If the magnet is too strong, then the initial energy required to travel the loop will be higher.

What you end up with is a ball "orbiting" within the loop, and opposing forces varying the energy between potential (position) and kinetic energy (speed), but without changing the total energy of the ball. Any path that doesn't reduce the energy of the ball with collisions or friction will work in the ideal case where there is no friction or other sources of energy loss.

 

[rcgldr]

I'm trying to figure out why anyone thinks the ball would go back up the slide. It would just sit at the bottom.

As pictured in the diagram it wouldn't go back up, but with a curved lip at the bottom right to re-direct the path of the ball back up the ramp, then it would be the balls momentum that would cause it to initially follow back up the upper ramp.

If you changed the mechanism so the bottom ramp was a half circle, and the top ramp was a horizontal line, then it might be easier to see.

 

[Gokul43201]

If I drop a rock on an ice rink, and it stays there by gravity, does that violate conservation of energy? The energy goes into heating and deforming the ball and ground when it comes to a sudden stop.

Okay, so you too are counting on dissipative forces to cause the machine to fail. I used the term 'friction' to refer to all non-conservative forces (as I explained a little further down in that post).

But there's no reason to design the loop in a way that requires such an inelastic collision, is there?

 

[GotMex?]

Hey, thanks for that explanation Gokul43201. I asked my dad a few days ago and he told me to look up magnetic forces and to take note that it is a conservative force like you say. I did some reading and I was arriving towards the conclusion you give, but you make it very clear now.

My original explanation was that the magnet would be too powerful and would overcome gravity at the top. However, I did not like this explanation for a perpetual motion device because it seems to imply that it is an engineering deficiency, not a violation of the laws of physics.

Explaining it this way shows that even if we were to find a magnet and an arrangement so that the intended cycle was possible, the prescence of friction would still prohibit this machine from working perpetually.

Thanks to all who replied.