# Wind-Powered Perpetual Motion

(NOTE: It appears that I really blew it with this one. I’m the bozo in this story. After lots of discussion, a few equations, and a bunch of time scribbling on paper, I’m convinced that I got this one wrong in a big way. No excuses; I should have done the analysis much more carefully before posting this; looking back, what I did do was pathetically shallow and, frankly, stupid. I’m sincerely sorry
for calling the guys doing the experiment bozos. I’ll follow up later this weekend with a detailed post showing my analysis, where I screwed up, and why this thing really works. In the meantime, feel free to call me an idiot in the comments; I pretty much deserve it. I’m leaving the post here, with this note, as a testament to my own stupidity and hubris in screwing this up.)

Technorati Tags: ,
,

This has been quite the day for the bad math; I’ve encountered or been sent a bunch of real mind-numbing stupidity. Unfortunately, I’m too busy with work to actually write about all of it, so as I have time, I’ll pick out the best tidbits. Today’s example is a fascinating combination of perpetual motion and wrong metrics.

Via BoingBoing comes a bunch of bozos who believe that they can create a “wind-powered” vehicle that moves faster the wind that powers it.

This is, obviously, stupid. Wind power isn’t some kind of magic: you generate power by the motion of air pushing against some surface. And there’s the catch: the power comes from moving air pushing. The air has to be moving relative to the
surface being pushed – or there’s no way for the air to push it.

The idea that somehow wind moving in the same direction as a vehicle can make the vehicle accelerate when it’s moving faster than the wind. There’s no mechanism by which the air can push. It’s actually really remarkably simple: the force of a wind pushing on a surface is (roughly) proportional to the speed of the air relative to the surface, and in the same direction as the motion of the air relative to the surface. Once you’ve got that much, then it’s just good old F=ma.

So, you start with the vehicle at a standstill, and the air moving relative to it at 7mph. So you’ve got an air speed relative to the vehicle of 7mph. This pushes the vehicle, so it accelerates. After some amount of time, it’s been accelerated up to 3.5 mph. Now the relative speed of the air relative to the vehicle is down to 3.5mph. The force of the wind against the vehicle is now half what it was when it was stationary. And so on, until friction becomes dominant, and you get a stable speed.

What would happen if you went faster than the wind? Like, say, 10mph? The speed of the air relative to the vehicle would then be -3mph. So the force from the wind would be in the opposite direction – that is, the air will now be decelerating the vehicle. There’s just no way for a wind moving more slowly than a vehicle to accelerate that vehicle in the direction of the wind – the wind isn’t moving from the viewpoint of the vehicle anymore.

But that’s not the end of the stupidity of this. We haven’t gotten to the
error of measuring the wrong thing.

A couple of bozos decided that they wanted to test the vehicle under controlled
laboratory conditions. They cite relativity to make the claim that a vehicle moving relative to the ground is equivalent to the ground moving relative to the vehicle – they’re just different, but equally valid frames of reference. And they’re absolutely right about that, as far as it goes.

Using that reasoning, they make a model of the vehicle, and put it on a treadmill. And, suprise! – when they do that, the propeller spins and accelerates the vehicle towards the front of the treadmill – the propeller on
the vehicle accelerates it forward, so that it’s moving faster than the treadmill.

Anyone with a clue should see the problem here.

If you’re testing a wind powered vehicle, then in a closed, windless
room, putting the vehicle on a treadmill moving at 10mph is not the same thing as putting the vehicle on a stationary surface in a 10mph wind.

You see, they didn’t actually preserve the original reference frame. In the original reference frame, you had a stationary vehicle with no motion relative to the ground, with the air moving in a forward direction relative to vehicle, and with no other forces acting on it. In the treadmill experiment, you had a vehicle that was moving forward relative to the ground beneath it, but stationary relative to the air. There is no wind relative to the vehicle when it’s put onto the treadmill. That might seem like a problem – but in
the outdoor experiment, once the vehicle accelerates to windspeed, it’s really the same situation as in stationary air with the treadmill moving under it.

But… the treadmill reproduction isn’t an accurate reproduction of
the original outdoor demonstration of the vehicle. There’s a force operating
on the vehicle on the treadmill that isn’t there in the outdoor demonstration.

If you want to accurately reproduce an experiment under controlled conditions by producing an equivalent reference frame, you actually need to perform the experiment in an equivalent reference frame.

In the real world, we obviously can’t perfectly reproduce everything. But
you can look at the experiment you’re performing, and ask what the key,
relevant factors are to accurately reproduce it.

In the case of this “wind powered vehicle”, the most crucial thing is
the forces operating on it. What the experiment is trying to do is
demonstrate that wind can exert a force on the vehicle even when the vehicle is moving faster than the wind. So it’s crucial to make sure that your controlled experiment
includes the forces from the original demonstration, and doesn’t add any
forces that weren’t present in the original.

The reason that I say this is a problem of metrics is because the bozos in question believed that they were creating an equivalent reference frame because the ground speed in the two reference frames were equivalent. Speed is one metric – that is, one measurable quantity in the experiment. But it’s not the important
one. If you can reproduce the experiment using a windspeed of 10 mph instead of 7mph, no one is going to claim that you’ve changed anything important. Velocity is an important metric for measuring the outcome of the experiment. But all that you really care about is the difference between the velocity when the experiment
started, and the velocity when it finished. On the other hand, force is
the key metric: what the experiment really wants to test is if wind can exert a force on the vehicle under the correct conditions. That’s the key.

Their results are rubbish for exactly that reason. There’s a force present
in their controlled experiment that wasn’t present in the original. And that
force is being channeled through the mechanism of the device to provide some
apparent forward acceleration.

Whenever I point out a perpetual motion machine, I get people
insisting that while the device in question might violate the laws of physics, it’s not a perpetual motion machine. So I’ll address that criticism before anyone
even raises it. What this device claims is that if you start the vehicle in a wind, once it’s started, if it’s placed it in non-moving air (that is, the situation which is equivalent to when it’s accelerated to windspeed), it will continue to move, and even to accelerate. That’s classic perpetual motion.

Finally, I’ve got to say that I’m incredibly disappointed in how utterly
clueless about things like this most people are. In the original post on
BoingBoing, the author says:

I admit that I don’t understand the physics involved, so I don’t really know whether DDFTTW is possible, but I am siding with Charles on this because I’ve never known him to be wrong when it comes to math, physics, or electricity.

Everyone should be able to understand the physics involved here. My third grade daughter can understand this. This isn’t difficult. There’s nothing
tricky or subtle about it. If you have a vehicle moving at the same velocity as
the wind, the wind cannot possibly exert any force on the vehicle. No force, no acceleration. Period. How can supposedly intelligent, educated people not know this?

## 399 thoughts on “Wind-Powered Perpetual Motion”

1. Blake Stacey

As the old joke goes:
“It’s so obvious, why, even a third-grader could understand it!”
“Uh, can somebody find me a third-grader?”
That kind of credulous treatment of perpetual motion and kooky propulsion schemes is nothing new, sadly. Sometimes, the techno-savvy bloggers and other such insiders of “New Media” seem particularly susceptible to the lure: “It would be so cool if it worked”. . . completely overriding their baloney detection faculties.

2. Justin M

Can we just go ahead and say that anytime anyone uses relativity to “explain” a phenomenon that “breaks the laws of physics” they are officially a quack.

3. Mark C. Chu-Carroll

Kuroki:
You know, there’s a very good reason why wind-propelled vehicles use sails, and not windmills.
The amount of force against a surface by a typical, moderate wind is very small. The point of a sail is to create a very large lightweight surface, so that you’re catching more force, without adding much weight.
You can’t avoid the need for a lot of surface area to gather a lot of energy from the wind. Even the smallest wind turbines have a very large surface area for the amount of energy they produce. For a a vehicle, anything less than a sail is just impractical – the size of the mechanism you’d need to capture enough wind to do anything useful with just
doesn’t make sense.

4. bill

I gather you are not a sailor.
“Some extreme design boats are capable of traveling faster than the true windspeed.” http://en.wikipedia.org/wiki/Sailing#The_physics_of_sailing
The physics still has a TANSTAAFL constraint, Thermodynamics isn’t threatened, but Momentum and Energy are the conserved quantities, not speed.
[Of course, friction or other conversion losses usually wins anyway, except in useless edge cases.]

5. William

It’s a little more complicated than just an object being pushed downwind because you have the propeller (being driven by the wheels) creating thrust also. The issue is whether the thrust from the propeller is really enough to drive the vehicle faster than the wind considering all the drag factors also at work.
I’ve seen video of one of these vehicles on it’s own and to me, it looks like most of their ‘faster than the wind’ effect is from gusts that give the vehicle enough momentum to carry on ‘faster than the wind’ for limited stints.

6. Eric

It’s not that hard. The claim is that the difference between windspeed and groundspeed creates a force that accelerates the object from a stop. Once they’re equal, there’s no force on the object from the difference. Therefore, no acceleration. Therefore, drag takes over, you decelerate below windspeed.
Not to mention there’s a sign change in there too. If the thing has a force on it in the forward direction when the windspeed is 2mph less than the groundspeed, then the shouldn’t the force be the other direction when the windspeed is 2mph greater than the groundspeed? The wind is the only place any energy enters the system, so once that energy is going the other direction, you’re screwed.
It’s telling that 90% of the support I’ve seen for this is analogies to sailing, not calculations of the actual device being talked about.

7. Bart

I’d like to point out that it is possible to go faster then the wind in a wind powered vehicle. And you can do it without violating any natural law.
Just not in this way.

8. Bart

Oh, forgot to mention that I didn’t mean that you can “outrun” the wind. Just that your speed can be faster than the speed of the wind.

9. Stephen Jones

I can see how these devices could potentially get above a constant wind speed by any “give” in the system being able to store energy from the wind when the device is moving slower than the wind being then released. This effect would be very short lived though. For example if the driven wheels are spinning faster than the ground speed, their momentum could carry the device over the wind speed temporarily.
An interesting question this raises though is can such a device travel into the wind faster than the windspeed. It would seem, given sufficiently low friction, that it’s quite easily possible and potentially much more surprising to see.

10. Paul W. Homer

Is it possible for the energy in the wind to indirectly go into creating some type of lift from the propeller, essentially causing it to corkscrew forward at a pace entirely independent of the actual speed of the wind? I.e the wind isn’t propelling it, it’s just fueling the forward motion from the propeller (and not causing drag)?
Paul.

11. Scott

I’m sorry, egg on the face I’m afraid, but DWFTTW is entirely possible. It’s not intuitive, but it is just an engineering problem. No perpetual motion is required.
The mental block I think is that people attempt to think of this as some sort of sail boat. A sail boat tries to have as little friction with the water as possible, whereas these little carts actually need to be tightly connected to the ground.
To put it another way, I’m sure that it is obvious that a windmill could gather enough energy in order by power a little cart to a speed faster than that of the wind. The counter intuitive part here is that a windmill can still gather energy, even if it is on the cart, as long as it has a way to exploit the difference in speed between the air and the ground.

12. Mary

Just want to add another voice saying that your argument about why a wind-powered vehicle can’t move faster than the wind is somewhat flawed. There’s an article here from physics today about the complicated and not entirely well-understood physics of sailing. What I gather is that sailing is best understood not in terms of the wind pushing the boat around, but in terms of lift, like what holds a plane up (also more complicated than what we are taught in school, having a lot more to do with angle of attack and vortices than I used to think).
Anyway, this is why it is possible to sail *into* the wind (well, not directly into it, but at a slight angle to it.)
The article actually says that “downwind is not the fastest direction for sailing.” Since you presumably move at windspeed when sailing directly downwind (barring friction) then I think this means you must be able to move faster than windspeed when sailing at the appropriate angle of attack.

13. Nick Johnson

@13: I can imagine (hypothetically) a cart like this, with high friction wheels, pointing _into_ a headwind. The headwind causes the propellor to turn, causing the wheels to roll and the machine to move into the headwind. What’s not clear to me is if this is possible or not – wouldn’t the force into the wind applied through the wheels be offset by an equal amount of wind resistance on all parts (stationary and the propellor)?
My apologies if I sound like a themodynamics-ignorant idiot. I like to think I have a fairly good grasp of thermodynamics, but a counter-intuitive problem like the one above taxes me a bit too far. đŸ™‚
Even if I’m right, though, I don’t see how that would allow something initially travelling downwind to exceed the windspeed, since the wind direction relative to the craft (and hence the direction of the force applied to the wheels) would have to reverse at the point it exceeds wind speed.

14. Mark C. Chu-Carroll

Mary (and others):
Simple “push”-based wind propulsion isn’t the best possible way of extracting energy from wind. A flat sail facing exactly to catch the wind is very inefficient.
That doesn’t change the basic underlying fact that the bozos who believe this story don’t get: you can’t extract energy from “wind” if the air isn’t moving relative to whatever device you’re using to extract energy from the wind.
Sailboats are complex – but not because there’s any mystery about whether the wind needs to be moving relative to the sail. The basic mechanics of a sailboat sail are amazing; basically, the wind is doing two things with the sail. First, it’s filling the sail, giving it its shape. Second, it’s flowing around the sail as if it were a wing. So you get a bernoulli effect around the sail -basically the same idea as lift on a wing.
The trouble comes about because you’re not just dealing with the wind. There’s a complex interface with the water; and water is (mostly) uncompressible, with very high friction in same aspects, and very low friction in others. Sailboats make this particularly complex, because they’ve got very large, shaped keels that are also using the Bernoulli principle to generate an airfoil-like effect in the water. The propulsion of the boat is a combination of the interaction between the the force of the wind on the boat via the sail, and the force of the water on the boat via the hull and keel. You can get some very surprising effects that way. If you’ve ever looked at the math of how a sailboat tacks against the wind, you can get a taste of that. You’ve got the set set at an angle which produces a thrust nearly perpendicular to the wind; the boat’s keel in the water is able to effectively turn that into motion slightly into the wind.

15. Mark C. Chu-Carroll

Re #13:
No. The speed of the wind relative to the ground doesn’t matter to a moving vehicle. If the air isn’t moving relative to the vehicle, the vehicle has no way of extracting energy from it. The speed of the wind relative to the ground can’t even be measured from a moving vehicle unless you know its speed relative to the ground.
If you’re not moving directly with the wind, then you get a host of other factors that can creating fascinating effects. You can imagine, for example, a wind turbine or a sail, which can each be used to move in a different direction than the wind – or even in the opposite direction from the wind. You can also get some elastic effects from the wheels against the ground using a wind nearly perpendicular to the vehicle’s direction of motion.
But if the wind has no velocity relative to the device on the vehicle which is supposedly gathering the energy, there’s simply no physical way to extract energy from it.

16. Nick Johnson

Mark: Are you able to comment on my hypothetical scenario? Would a device such as theirs pointed into the wind (and rigged to move into the wind if the propellor is spun in the direction the wind is blowing) move? And if so, what would its equilibrium speed be dependent on?

17. Scott

Mark,
I’ll try and build an imaginary device that solves this problem.
Imagine a windmill with robotic legs like a kangaroo, and a electrically powered cart. The windmill powers the cart via a 10 foot long wire, and the windmill has the ability to jump 20ft.
When the wind blows, the windmill powers the cart, and it starts rolling into the wind. When the cord runs out, the windmill jumps 20ft to a new position in front of the cart, where it starts gathering energy from the wind once more.
The cart, being very aerodynamically shaped, continues to accelerate, into the wind. When the cord runs out, the windmill jumps to the next position, and the cart continues to accelerate, faster and faster. What is our top speed? It’s limited only by the jumping of the windmill, and in no way tied to the wind speed.
This is of course ridiculous, but the intent is to illustrate that we can gather energy from a difference between to mediums regardless of our own velocity.

18. Mark C. Chu-Carroll

One more thing re: Mary:
It’s not correct to assume that a sailboat will be able to achieve the same speed as the wind if it’s sailing directly downwind. As I mentioned above, a sail that produces thrust by just catching wind – effectively by something like elastic collisions between the molecules of the air and the sail – is not a very efficient mechanism for extracting kinetic energy from the air. A very aerodynamic/hydrodynamic boat with huge, perfect sails will never achieve the speed of the wind purely on the thrust from the sails, because the “collision” mechanism of extracting energy from the wind is so inefficient.
The airfoil technique is much better, and much more efficient.
Another way to think of this is windmills. You can build a windmill that works on the nearly the same principle as a waterwheel: put the vanes perpendicular to the axis of the
spindle, and only expose one of them to the wind. You’ll get a force from it, from the flat collision principle. But no one builds windmills using that model. Pretty much every windmill or wind turbine uses some variation of an airfoil – because the airfoil is much more efficient.
A sailboat sail is really operating as a wing, using Bernoulli effect – and so it’s going to get its best speed running at an angle to the wind which maximizes the lift against the “wing” shape of the sail while filling the sail strongly enough to maintain its airfoil shape.

19. jivemasta

If you think about it another way, they are saying that this will move even with no wind. Once it gets up to the speed of the wind(best case scenario) there is a net force of 0. So to an experiment that actually proves something. Remove the wind factor, and push it up to 7mph in a long closed track with no wind. According to their theory, it would keep going even though it’s going faster than the wind(0mph).
This makes the cartoons with the sailboat+fan idea seem smart.

20. Mark C. Chu-Carroll

Re Scott #21:
Nope, no good. Doesn’t work.
The problem is, where does the energy for moving the windmill come from? In your scenario, during the periods when the windmill is stationary, you can provide power to the vehicle to make it accelerate. But the real vehicle in your scenario isn’t just the little cart with wheels: it’s the entire system of the windmill with the cart. The “jumping forward” part represents a huge hole in the energy equation. Where does that energy come from?
In fact, you can entirely eliminate the cart from your scenario, and just have the jumping windmill. And your jumping windmill will be able to go faster than the wind for short periods of time – because it’s not moving at all at others. During its non-moving times, it’s storing energy from the wind, and then expending it in one
large burst during the hops. But it’s never going to achieve better constant speed
than the wind.

21. Scott

“If the air isn’t moving relative to the vehicle, the vehicle has no way of extracting energy from it.”
This is your error! You don’t take energy from the wind; you take it from the difference between the wind and the ground.
You are in a cart, with a sail, traveling at wind speed. Stick a wheel with a generator to the ground, and gather some of that “wind” energy in a battery. Then, take down the sail and power yourself into the wind with the gathered energy.
Can the gathered energy be enough to offset the cost of gathering it? Yes, because you can get back to wind-speed for “free” before you turn on the motor, and you can coast DWFTTW after the battery runs out.

22. Anonymous

“During its non-moving times, it’s storing energy from the wind, and then expending it in one large burst during the hops. But it’s never going to achieve better constant speed than the wind.”
Energy is expended to move hop forward, but it can also be recovered in the landing. The landing energy can be recovered via regenerative breaking. A hybrid jumping windmill. Energy is being gather but not lost. Except for friction, air resistance, etc.
Well, ok, the point is lost before the improbability of the device. It’s probably better to focus on the cart example from my last posting.

23. Scott

“During its non-moving times, it’s storing energy from the wind, and then expending it in one large burst during the hops. But it’s never going to achieve better constant speed than the wind.”
Energy is expended to move hop forward, but it can also be recovered in the landing. The landing energy can be recovered via regenerative breaking. A hybrid jumping windmill. Energy is being gather but not lost. Except for friction, air resistance, etc.
Well, ok, the point is lost before the improbability of the device. It’s probably better to focus on the cart example from my last posting.

24. Mark C. Chu-Carroll

Anonymous:
How about this? Just say that the cost of moving a vehicle forward faster than the wind is zero. Then the whole thing is moot.
Even in the perfect, ideal situation for your hopping windmill, the process has to start with the windmill having enough energy to make the jump. So you’re starting the process with a bunch of stored energy – and what you’re doing is powering your vehicle with the energy extracted from the wind, plus the energy stored in the system before you started it. In that case, you can move faster than the wind, until you’ve used up the stored energy.
And you’re talking about an entirely different scenario than what the original experiment was purportedly testing. They’re talking about a device whose only source of energy is the wind blowing during the time that it’s moving.
Add more energy, and sure, you can do it. But in that case, what you’ve really done is said “If I’ve got an electric car with a windmill attached to it, I can make it go faster than
the wind as long as the battery holds out.” No one is surprised by that.

25. Scott

Sorry, the anon was just me.
I have a pretty hard time understanding this device myself, so I resort to these crazy examples.
You see, the spinning prop is geared to the ground. By doing that, it acts like the jumping windmill. The forward speed is neutralized by the increasing speed of the blade. The result is that the difference in relative speed can be exploited.
Did you see my post #25? It seems the better example. And I apologize for the double on 26/27.
I’m going to go away for a while and let other people post.

26. Flex

Pjb wrote, “I’d like to see what these guys have to say about the ol’ airplane-on-a-treadmill riddle.”
You may know the answer but others here may not.
It’s quite simple really, the wheels on the airplane spin faster.
To understand this, consider hypothetical worst case; the plane is being pulled back by the treadmill. At this point the possible lift generated by the wings is zero.
The thrust of the plane (propeller or jet, doesn’t matter) will move the plane forward on it’s free-spinning wheels until enough forward motion is reached to generate lift.
Regardless of the speed of the treadmill, and assuming nothing interferes with the system like the plane pulled into the treadmill mechanism, there isn’t a problem.
It may take a little longer to reach the point of flight, but the only additional time needed is to get to the point where the plane is stationary relative to the airspeed. At this point the wheels are moving at the same speed as the treadmill.
To look at it another way, it’s identical to taking off with a tailwind. When taking off in calm air, the tires are motionless with respect to the ground, and so are the wings. With a tailwind, to get to the point where the wings are motionless to the air, the tires need to be moving relative to the ground.
And, of course, with a headwind the tires don’t need to move as fast relative to the ground as in a calm.
I think where people get confused is the missunderstanding that when a plane in of the ground the tires are driving the plane. Even though we know this isn’t so, most of our everyday experiances with vehicles involve the driving force coming from the interface between the wheels and the ground. We rarely encounter vehicles powered any other way.

27. Mu

In the original video, they are using a huge wooden propeller which essentially serves as a flywheel to power the device through low wind conditions, giving the impression of “faster than wind” by the position of the wind indicator. In the treadmill experiment, they are transferring a small percentage of the energy provide by the treadmill motion into acceleration of the vehicle using the thrust of the geared propeller. Neat device, but, as MCC pointed out, not a wind powered machine at all. You notice how they always place the vehicle on the running treadmill, thereby spinning up the propeller before releasing the vehicle, providing a certain amount of stored energy. If they’d followed their own claims it should work with a stopped start. In which case their device would accelerate backwards at high speed đŸ˜‰

28. peter

point to someone above, you can outrun the wind, and by a fair margin in a low drag boat.
by the time a sailboat on a broad reach, (~120 to 140 deg away from the true wind direction (which is the fastest point of sail for any boat using lift based sails.(as opposed to spinnakers or square sails.))) reaches it’s top speed, you are actually sailing upwind against the apparent wind. at that point you are close hauled and doing about 1.5 times the true windspeed. the drag is that you can only really do this in one direction.
windsurfers and catamarans (and iceboats) are famous for this. someone better equipped can work out the vector math.

29. KJC

If we are agreed that forward speed on a borad reach can be faster than the wind speed due to the sail acting as a wing, is it somehow impossible that when running (wind directly behind) a propeller cannot act in the same way? Does the properller have to introduce a net drag? My intuition is that it does not but my aerodynamics (knowledge) is not good enough.

30. Dave

So if, instead of a normal sail, you used a solar sail, and the photons were blowing against the sail with a speed of c, then, if you could sail faster than the wind (photons), you’d have FTL travel! đŸ˜‰
Dave (tongue firmly planted in cheek)

31. peter

kjc@#33
I think if you imagine that a propeller pointed directly into the wind is actually made up of individual little sails that are angled to be at the equivalent of a broad reach you might find it easier to understand why aiming a propeller in any other direction would cause it to go slower.
a sailboat on a broad reach that is going fast enough appears to be heading upwind, and there’s a limit to how close to the wind you can point a sailboat and still go forward.
there are some windmill powered boats around, woodenboat magazine profiled one some years ago. as I recall, its upwind performance was pretty good but you had limits going downwind. the only reason that it retained any use at all going down wind was that as a displacement hull it was already limited in speed by drag from the water and the speed of the wave generated by the size of the hull. therefore, providing the wind was strong enough, the boat would be going slow enough to provide enough difference in speed to allow the blades (and eventually the propeller) to turn in some useful fashion.

32. elspi

A sailboat which is moving at an angle to the wind will be able to go faster than the wind.
However the component of velocity of the sail boat in the direction of the wind will be strictly less than the speed of the wind.
This is the point. You can go faster than the wind but not faster in the direction of the wind which is what the twits are claiming.
And yes, they have the stupid turned up to 11.

33. Mary

You can’t extract energy from “wind” if the air isn’t moving relative to whatever device you’re using to extract energy from the wind.
If we are agreed that forward speed on a borad reach can be faster than the wind speed due to the sail acting as a wing, is it somehow impossible that when running (wind directly behind) a propeller cannot act in the same way?
I think these are the best arguments against and for the idea of “outrunning the wind,” and whichever side convinces me is going to have to address one of these arguments very clearly. đŸ™‚
On the “for” side — it’s pretty clear to me that with the propeller scenario, the relevant surfaces are the propeller blades, which are not at 90 relative to the wind, so any argument about what would happen to a vehicle which is simply pushed by wind is irrelevant. There are definitely angle-of-attack issues here. Counter-intuitive things can happen in cases like this. Again, sailboats can go into the wind, can “outrun” the wind by sailing at an angle to it, planes can fly upside down, and as someone in the Boing-boing thread mentions, helicopters can “glide” to a landing using the lift generated by the rotors as they turn due to the downward motion. My physics intuition doesn’t know how to handle any of that. (And incidentally, Mark — I think you’re oversimplifying by talking about Bernoulli’s principle too much. Here are couple of links from NASA and from some guy who has lots of nice diagrams which discuss the controvery about the source of “lift” and make the case that the “Bernoulli” explanation is too simple.)
Then too, from an energy conservation point of view — well, there is a lot of energy available from wind, and if I could build a sail big enough, I imagine I could collect enough of it to move a light vehicle more or less as fast as I please. As someone else said, it’s not speed that’s conserved, it’s energy and momentum. And as we’ve all agreed, a stationary windmill that could wirelessly beam the power it produces to a vehicle could power that vehicle to an arbitrary speed as well.
On the other hand, as the case “against,” there’s the point you make above “You can’t extract energy from “wind” if the air isn’t moving relative to whatever device you’re using to extract energy from the wind.” This is a good point. If I am moving downwind at windspeed, then I am at rest relative to the wind. How do I then move faster? (I don’t buy the argument about extracting energy from the turning wheels, either. That would just slow you down.) And yet this same argument would seem to apply to a sailboat — which we have already seen *can* move faster than the wind. True, it can’t do so directly downwind, but then this thing uses a propeller, so even when this cart is travelling directly downwind, its “sails” are angled to the wind.
The truth is I’m not sure if this is possible or not (and I’m about six months from my physics PhD, albeit in optics, not aerodynamics.) In any case, it does not strike me as obvious, and I don’t think the reasoning in your post is sufficient to disprove it. Again, counter-intuitive things happen all the time with aerodynamics. It *is* true that energy must be conserved, of course. But it’s going to take a more careful analysis of the energy in this system to prove that merely travelling faster than the wind requires a violation of energy conservation. Because again, sailboats do that all the time.

34. Davis

So you get a bernoulli effect around the sail -basically the same idea as lift on a wing.

Not to get too far off-topic here, but my understanding is that lift on a wing is a bit more complicated than just the standard “Bernoulli effect” explanation.

35. hibob

Re Mark #22
“A sailboat sail is really operating as a wing, using Bernoulli effect – and so it’s going to get its best speed running at an angle to the wind which maximizes the lift against the “wing” shape of the sail while filling the sail strongly enough to maintain its airfoil shape.”
That’s the complication that has kept this argument alive on the net for years – the propellor is an airfoil running at a relatively high velocity compared to the cart, and at an angle to the wind. Thermodynamics may be absolute, but math is tricky.
Also, I don’t think the best model is of a static situation where the cart is matching the speed of the wind exactly.
Here’s the model I want to see:
Pick a wind speed, say 20 kph, and a prop size, say 2 M. Assume there is a lossless continuously variable transmission to connect the wheels to the prop, and a similarly lossless rig to change the angle of attack of the prop, so both parameters can be optimized on the fly at any speed. Wheels are similarly spec’d from Physics 101, but still somehow have perfect traction on the ground.
Now comes the tricky part. Find an aeronautical engineer with spare time on their hands (well, that part might not be tricky these days) and have them pick a prop design from the tables that should be well optimized for the conditions. Find them a workstation and software that can handle the math, and have him model the thrust generated vs drag encountered as a function of speed of the cart for 15 to 25 kph when it is moving in the same direction as that 20 kph wind, while optimizing for the best ratios for the CVT transmission, AOA for the prop, etc. Try it with a few more prop designs, see which one does best. Fiddle with the amount of drag generated by the cart and the prop as a system. Buy the engineer a lot of beer and commiserate on the lack of jobs available these days.
Then show us the plots. I want to see where the numbers go negative for the best case scenario.

36. bigTom

36 seems to have it right.
If you are traveling directly downwind in a steady wind field, you can see that as you approach the windspeed, the force on the turbine blades will approach zero, so you can’t travel directly downwind as fast as the wind. I’m not sure about tacking -can you move downwind faster than the wind speed -I doubt it as when your downwind velocity is the same as the wind speed, you only have relative windspeed at right angles to the wind, and that can’t provide horizontal thrust…
In any case they did have an engineering challenge with Ventomobiles, wind turbine powered vehicles, but race was directly into the wind. The winner was something like .6 something of the windspeed, which I thought was pretty impressive. Of course if you could store energy internally you might be able (on average) travel downwind faster than it. You would have two travel modes, one energy generating, at velocity considerably less than the windspeed, then state 2 is powered by the stored energy faster than the windspeed, and the blades are feathered to minimize drag. With sufficiently small enough friction that would work, but it is probably too hard to get the frictional loses low enough to actually do it.
Interestingly the tips of commercial wind turbine blades are typically moving at 5-6 times the wind velocity!

37. Scott

Mark,
I think you have got are misunderstanding what’s going on in the video. Grainy youtube and all…
You seem to think that the slope and the movement of the treadmill are acting against each other and the vehicle is in the middle, using the slope to capture energy from the treadmill.
In actuality, the slope is pushing the vehicle to the left. Also, the movement of the belt is pushing the vehicle to the left. The only thing pushing the vehicle to the right appears to be the propeller. How does the vehicle climb the slope while working against the treadmill? It is clearly doing just that.
The “wind”, is of course the still air in the room, which relative to the treadmill is moving. Think of the same experiment performed on a breezeless day on the deck of an aircraft carrier making a steady 10 miles per hour. The air is “still”, but the viewer perceives a breeze.
(reposted from my post on Word Munger, edited for not grammar and clarity)
Also, Mary, great posts!!!

38. Robert S.

It seems like reducing this to the forces involved would be the easiest way of looking at it.
The cart can extract energy from the differential speeds of the ground and the air (if there is no difference the cart can extract no energy). The cart, in turn, has a few forces acting on it, drag/thrust against the ground, non moving parts drag in the wind, and the harder to calculate forces on the propeller.
My math skills are rusty and my physics skills are damn near non-existent, so I cant work out the math. It’s clear that at some speed the various drag elements would balance any energy extracted from the differential in speeds.
Someone mentioned that max speeds for sailboats were in the range 150% of wind speed earlier. I’m fairly sure that the new 90 foot trimarans being prepped for the next Americas Cup Challenge are routinely hitting 2.5 times wind speed due to better construction.

39. jhnboy

The question isn’t whether you can use the wind to travel at a greater speed relative to the ground than the wind itself is traveling. Of course you can; sailors do it every day. What is interesting is whether you can get from A to B more quickly than the wind does, solely by using wind power. Temporary speed bursts and the ability to go really fast but not in a straight line don’t get you from A to B faster.
When sailing, you can tack downwind faster than the wind, but you’re traveling a greater distance when you tack. So if the wind is traveling at 10 knots, you can tack at 18 knots. (example here: http://www.physclips.unsw.edu.au/jw/sailing.html)
If you keep tacking back and forth, can you arrive at point B from point A more quickly than someone could who was traveling in a straight line, but at exactly 10 knots?
If you can, then you’ve just sailed downwind faster than the wind. Solely by using wind power, you’ve arrived at B from A more quickly than the wind did.
The proponents of the downwind faster than the wind device, in addition to claiming that their device can get from A to B faster than the wind, claim it is possible to tack with the wind, and arrive at a destination faster than the wind does. I do not know if this is correct.
They furthermore claim that if you have two boats tacking with the wind, but at opposite angles, you could connect a telescoping tube between them. A chair mounted at the center of that tube would be traveling downwind faster than the wind. I don’t see the flaw in this, apart from the fact that the distance you’d be able to travel dwfttw is limited.

40. 2-D Man

So if, instead of a normal sail, you used a solar sail, and the photons were blowing against the sail with a speed of c, then, if you could sail faster than the wind (photons), you’d have FTL travel! đŸ˜‰

It’s actually tachyon particles that hit your sails when you pass Bajor. Before you know it, you’re at warp and you arive at Cardassia Prime. Gul Dukat congratulates you.

41. cm

jhnboy said:

They furthermore claim that if you have two boats tacking with the wind, but at opposite angles, you could connect a telescoping tube between them. A chair mounted at the center of that tube would be traveling downwind faster than the wind. I don’t see the flaw in this, apart from the fact that the distance you’d be able to travel dwfttw is limited.

This is begging the question. What new information does this image provide that should convince anyone that the chair at the center would traveling faster than the wind? There is none.
I think the “cognitive trap” here is that since the sailboats at either side are traveling faster over their lane of water than the wind is traveling over its lane of water, we wrongly assume that the chair in the middle has the same speed over its lane of water as the boats do over theirs–after all, “it’s connected to them”. And then, since the chair’s lane of water is the same as the wind’s, we think the chair races ahead of the wind.
But that thinking is wrong. The boats at left and right, though moving faster on their trajectories than the wind on its trajectory, are “wasting time” in this race against the wind by tacking off at angles while the wind is taking the shortest distance between two paths at a clip such that the speedy tacking of the boats will win not the race. The chair will be seen to slowly move forward, following the wind but not beating it, even as the two boats are achieving a higher knots than the wind on their indirect angles.

42. jhnboy

cm
http://www.boatdesign.net/forums/sailboats/tacking-downwind-faster-than-wind-24761.html
http://sports.groups.yahoo.com/group/2nalsa/message/161
Demonstrate that it is possible to beat the wind from point A to point B by tacking.
The chair example is weird. But if I can beat the wind on a sailboat by tacking twice, and I can, that satisfies me that traveling downwind faster than the wind is possible.
I have no idea if the cart works. But who cares if it does? It would simply a new mechanism to accomplish what is already possible.
Here is some text from the Yahoo thread.
QUESTION:
> On a frozen lake, with a steady 15mph wind from north to south a
> start point is defined. Directly south, one mile away a finish point
> is defined. The ice boat takes a ~45degree track (say SW) and comes
> across the start point at full speed. At the moment the ice boat
> passes the start point, a neutral buoyancy balloon is released,
> floating to the south in the wind. At the ~halfway point, the ice
> boat tacks to the SW and takes aim at the finish point.
>
> Can the ice boat win the race to the finish point, or will physics
> dictate that the balloon always wins?
“Welcome John,if you measure out 2 identical tacks at 45 degrees as you
proposed you travel 1.41 miles to reach the same 1 mile mark. If it
was blowing 15mph that means to have a dead heat you would only have
to sail an average speed of 21.15 mph.
That would be pretty easy to do in even the most basic low performance
dirt or ice boat.”

43. Mary

Yeah, I’ve thought about it all afternoon (way to kill my productivity, people) and actually watched the video, and while I still think that it may be possible to go DWFTTW, *this* is not the device to do it.
When the thing is at rest relative to the ground and moving relative to the air, fine, the wind can push the propeller blades, which forces the wheel to turn, which exerts a force on the road, which causes the road to exert a reaction force pushing the vehicle forward.
But when the thing is at rest relative to the air and moving relative to the ground, the moving wheels are forcing the propeller to turn against stagnant air, and *that* takes energy away from the wheels. If you want to spin a pinwheel in still air, you have to keep pushing it, and that takes energy. This would slow the wheels down. The steady state should be (due to friction) slightly less than wind speed. Now a good gust might indeed give you momentum to carry on faster than the average wind speed for a while, until the next gust, but that’s not so special.
Maybe with flywheels and propellers facing different directions, or ratchets, or something, but not with a fixed gear propeller / wheet cart.

44. Stephen Wells

@47: I notice that the threads you posted contain a large number of assertions that you can beat the wind by tacking, but not apparently any evidence; for example the yahoo thread you quote ends with the _assertion_ that you could easily achieve the speeds described but again no evidence. Also the problem in the Yahoo thread is ill-posed; if you released a balloon from a standing start while the boat gets to cross the start line at full speed, the balloon would average less than the wind speed across the track, it would need time to accelerate, no?

45. Stephen Wells

Re. the video on the “Mark is wrong” blog post:
The demo is hilarious. It’s clear that the treadmill is running in normal treadmill direction (right to left in this video), that the treadmill spins up the wheels _while the cart is held stationary_ and that sets the propellor going, and that the rotating propellor drives the cart. If you set the cart down without holding it it’d be flung off the treadmill immediately. No energy is being extracted from the wind.
I’m pretty sure on energy conservation grounds that if you set the thing on a long enough treadmill and spun it up then let it go, it would only move a finite distance up-slope before reaching steady-state or heading downhill again. With this short treadmill the guy can keep catching it on the up-slope part of the journey, which is being powered by the energy stored when the prop was first spun up.
Always be very suspicious of demos that interrupt a process and then assert it would have continued indefinitely.

46. Stephen Wells

Also I just thought of this: there is no difference relativity-wise between the machine travelling at say 22mph downwind with a wind speed of 20 mph, and the machine travelling at 2 mph with a wind speed of 0; so the wind argument implies the machine, set down at rest in still air, will spontaneously set off! that should make it clear that it’s the ground interaction that makes the difference.

47. T_U_T

I think it is possible but not that straightforward.
Consider this. First. You start tacking. You will reach say 1.5 winds speed. Of course the component of your velocity in the direction of the wind can NOT exceed wind speed so you will move say 0.5w in wind direction and 1.4w sideways, together 1.5w. But. Then you suddenly roll up your sail and generally assume the most aerodynamic shape you can. Then you will make sharp turn windwards and because of inertia, you will find yourself moving faster than the wind at 1.5w. slowly, your speed will reduce to say 1 ws, then you will turn to side, and start tacking again in opposite direction. so the windwards component of your speed will alternate between accelerating from 0.33w to 0.5w then jumping to 1.5w and deccelerating to 1w then reducing to 0.33w and again.
So, if the decceleration phase is long enough ( > 7/15 the time you need to accelerate by tacking from 1w to 1.5w ) you will move at average higher than wind speed.

48. Scott

For the doubters, the following video:

It’s the same cart, but a video longer with less cuts. If it’s not clear from the video, the treadmill is pushing the cart to the left. Also, the treadmill is inclined, which also pushes the cart to the left. Yet the cart is climbing to the right.
Another complaint is that this doesn’t test DWFTTW. But it does — still air and a moving surface is what a vehicle observes when it reaches equilibrium with the wind. Can a vehicle at rest with the wind extract energy from the moving ground and use that energy to further change its momentum? If you have watched the video, you have seen it happen.
I think that it’s reasonable to doubt the DWFTTW premise, but if so the video needs explanation.

49. jhnboy

Stephen
So start the balloon earlier so that when it crosses the point it’s traveling at windspeed.
Unlike the cart, the issue of tacking faster than the wind in the direction of the wind is easily solved.
I’ve seen evidence that it is possible, both from measurements of boats, vector analyses, and sailors who have actually done it.
By contrast, the only evidence I’ve seen that it’s not is the unsupported assertions of people who simply assume that it’s impossible.
I’d say again: you can travel upwind using the wind as your only power source. Isn’t this much more counterintuitive than tacking downwind faster than the wind?

50. Mark C. Chu-Carroll

Scott:
The longer video shows nothing different than the short one, and it’s already been explained plenty of times.
The *claim* is that this vehicle extracts energy from the wind, allowing it to go faster than the wind.
The demonstration is one where it’s extracting energy from the *treadmill*.
Can a vehicle at rest relative to the wind extract more energy from the wind? No.
Can a vehicle at rest relative to the wind extract some energy *from some other source*, and use it to accelerate? Of course!
But the claim isn’t “By using both windpower and some other energy source, we can go directly downwind faster than the wind?”. The claim is “By using nothing but windpower, can we go downwind faster than the wind?”.
When the mechanism allegedly extracting energy from the wind is stationary relative to the wind, it cannot extract any more energy from the motion of the wind. That’s simple, trivial, elementary physics.
It can get energy from other things: from mechanical energy from a moving belt; from an energy storage mechanism, like a spring or a battery; from a heat engine; from solar power; from a fuel source. But you can’t extract energy from the motion of the wind if you’re not moving relative to the wind.

51. jhnboy

Mark,
How would you construct a test that allows the vehicle to remain stationary while exactly modeling the situation of traveling downwind? You need the opposite of a wind tunnel. In other words, a treadmill.

52. jhnboy

See “The Practical Encyclopedia of Boating” 297-98 for more information on how it is possible to tack downwind and arrive at a destination faster than windspeed would allow. (available on Google Books)
This is not different in principle that traveling directly downwind faster than the wind.
Mark is wrong.

53. Scott

Are you in the google NY office? ’cause if so, I could come down town and we could settle this geek style — with a whiteboard.

54. Mu

jhnboy, there are plenty of ways of traveling downwards of wind direction faster than the wind, no one denies that. The problem is – you cannot do it in a straight line with the direction of motion equal to the wind direction. And that is what the whole claim is about.

55. jhnboy

People have claimed that tacking downwind and arriving at a point faster than the wind would is impossible, and they do so quite regularly in discussions of this topic. It’s important to clear the air on this. Whether or not the cart is possible, it is in fact possible to outrace the wind, using only the wind as a power source. If you had a perfectly efficient sailboat traveling dead downwind at exactly windspeed, I could always beat it to its destination by tacking.
(The proponents of this cart claim that their design is based on the “exact” same principle as tacking *upwind*.)

56. T_U_T

problem is – you cannot do it in a straight line with the direction of motion equal to the wind direction. And that is what the whole claim is about.
You could. Just let the windmill blades carry out the same movements a sail would do during “FTW” maneuvering. Of course the mechanism would be more complicated. just a fixed pitch wind turbine simply can’t work.

57. Scott

I’m going to tell a little story which should hopefully explain why a treadmill that moves at 10mph is equivalent to a flat space with a 10mph hour wind.
Mark has been selected by google for a special project. He is put in a jet, and flown to google’s secret underground base in Ecuador. The project, Mark learns, is a time machine!
Mark is strapped in, and the time machine is turned on hurling Mark into the distant future.
When mark exits the capsule, the lab is empty and the lights are off. Mark can hear a howling wind outside, so he knows that he can use the lab’s windmill system to power up the time machine and get home.
As he prepares to activate the windmill, the stations weather sensors reveal that the wind is blowing at an astonishing 1000mph from the — and that is the rotational speed of the earth!
It’s night, and Mark doesn’t have a compass, so Mark can’t tell east from west. If this wind is coming from the west, that means that the atmosphere of the earth is moving at 2000mph, so there is a great deal of wind energy. But if this “wind” is coming from the east, that means that the air is standing still, and the planet is moving — so there is no wind energy at all! Will the windmill work?
Of course it will. In our real world treadmill example, Mark has asserted that it doesn’t count because the device is slowing the treadmill and speeding the air. But that is exactly what any “wind powered” device does. A windmill catching a breeze from the east slows the air a bit, but also conveys a push to the earth, slowing the earth’s rotation a bit. The energy is derived not “from the wind”, but from the capturing a part of the ground/wind momentum difference.

58. Jan Tolenaar

Is it the fault of global warming that nobody has mentioned speed skating? A skater can go faster than his legs move for the same reason a tacking sailboat can go faster than the wind speed. A skater has a low friction in the direction his skates are pointing and a high friction at right angles. A good skater pushes sideways while keeping his skates almost but not quite parallel. This small angle (large cotangent) multiplies the small movement in the high friction direction to a large movement in the low friction direction. Skates are sharpened to get the friction right. I suppose sailboats are sharpened as well.

59. Jan Tolenaar

One more remark: every skater knows, if you push in the low friction direction, you are not going anywhere!

60. othermark

The problem isn’t just that Mark is wrong, it’s that he’s so arrogant and insulting in his ignorance.
It should be clear that he understands nothing about the complexities of wind-powered travel, or he’d never write something like:
“Via BoingBoing comes a bunch of bozos who believe that they can create a “wind-powered” vehicle that moves faster the wind that powers it.”
Um, Mark, you mean like a sailboat?
He writes,
“What would happen if you went faster than the wind? Like, say, 10mph? The speed of the air relative to the vehicle would then be -3mph. So the force from the wind would be in the opposite direction – that is, the air will now be decelerating the vehicle. There’s just no way for a wind moving more slowly than a vehicle to accelerate that vehicle in the direction of the wind – the wind isn’t moving from the viewpoint of the vehicle anymore.”
Again, wrong. Sailboats do this. THEY TRAVEL __AGAINST__ THE WIND.
Aerodynamics are complicated. Not intuitively grasping them is nothing to be ashamed of. But calling people who have actual science and evidence on their side “bozos” while just quoting basic physics that everyone agrees with as if that settles the matter is arrogant beyond belief.
Mark, I’m sure you’re used to being right. This one is going to be a bitter pill to swallow.

61. SteveM

If the windmill is moving down wind at the same speed as the wind, then what difference does it make what angle the prop blades are at. The wind speed relative to the windmill is zero. Take a stationary windmill in still air, is there some angle you can adjust the blades to make it start spinning? No, there isn’t.
As for the treadmill experiment. I can imagine being able to design a vehicle such that in a steady wind, the treadmill has to keep accelerating to keep the vehicle motionless in the wind tunnel. But that represents a wind speed that is also accelerating relative to the ground. What you would need to do is to reduce the wind speed equal to the speed of the treadmill. It then becomes obvious that when the treadmill is at the original wind speed, the actual wind speed will be zero and the car will no longer go forward. Equilibrium will have to be at a speed less than the wind speed.
As for all the sailing arguments, the claim is “Downwind faster than the wind”, there is no sailboat that can do that.

62. jhnboy

“As for all the sailing arguments, the claim is “Downwind faster than the wind”, there is no sailboat that can do that.”
You should research this more. The only meaningful metric is being able to arrive at a given destination that is dead downwind faster than the wind can, powered only by the wind. You can. Sailors do this all the time. Your downwind “velocity made good” can be faster than windspeed.
The differences is that they zigzag to do it. But how does that change anything? They still get from A to B faster than the wind did.
The only question is whether what is already possible through tacking is possible without tacking.

63. SteveM

The only question is whether what is already possible through tacking is possible without tacking.
That is what I meant, straight downwind cannot go faster than the wind.

64. Stephen Wells

If you’re travelling downwind faster than the wind, which let’s say has speed U, then you’re travelling at a ground speed V>U and you’re travelling _into a headwind_ whose magnitude is V-U. Any form of argument suggesting that the wind is pushing the entire vehicle in its direction of motion is now clearly wrong. Any argument that the wheels turning are driving the propellor and keeping up the vehicle’s speed is a perpetual motion machine.
The argument then becomes that this headwind, speed V-U against the vehicle’s direction of motion, can drive the propellor and have it deliver power to the wheels sufficient to maintain the vehicle’s motion. Plausible? I am dubious.

65. Stephen Wells

Re. the above: Yes, I’m aware that you can make a turbine-powered vehicle move directly upwind, I think we had an example posted earlier. In that case we achieve ground speeds that don’t exceed the wind speed. This makes it implausible to me that you can use a smaller headwind to achieve a much greater ground speed, which is the claim here.

66. Mark C. Chu-Carroll

Re #66:
Sailboats can sail against the wind, but that’s absolutely *irrelevant* to this.
We’re talking here about a device that is moving in exactly the same direction as the wind, which is purportedly being moved by that wind, and nothing else.
A sailboat works by creating an airfoil at an angle to the wind, and using that in combination with the hydrodynamic effects of its keel and hull in water.
Put a sailboat in still water, give it a flat sail, face it dead forward into the wind, and try to make it go into the wind. Nothing is going to happen.
I can’t understand what makes this so hard for people. The fundamental issue is incredibly simple: can a vehicle extract energy from a moving wind, when that wind is not moving relative to the vehicle. And the answer is no.
Sailboats don’t suffer from this problem – because the whole point of tacking is that you’re not going in exactly the same direction as the wind. You’re creating an airfoil effect by moving at an angle to the wind. The air is always moving relative to the sail.
Nothing can extract energy from the motion of air if the air isn’t moving relative to it.

67. Dave Munger

Mark: This comment (by the creator of the craft) gives a great explanation of why faster-than-wind downwind travel is possible.
Quick summary: sailboats can and do travel faster than the wind if they tack (zig-zag back and forth). Their downwind vector is faster than the wind. A fast sailboat can beat a hot-air-balloon in a steady wind. This is established fact. Attach two boats together with a telescoping rod and sit in the middle and you’re moving due downwind, faster than the wind.

68. jhn

Mark,
The problem is that #46 is wrong. He assumes it is impossible for a sailboat to have a velocity made good faster than the wind. It isn’t.
Since it is possible for a boat to have a velocity made good downwind faster than the wind, it is possible have two boats take off at opposite tacks with a long piece of rope or telescoping pole connecting them. They each reach their destination before the wind would. And some point at the center of the rope will have been traveling directly downwind, faster than the wind.
This is a mechanism totally different than the cart. I don’t know if the cart design could work. But the idea of “directly downwind faster than the wind, powered only by the wind” is possible.
Ice boats can have a VMG much in excess of the windspeed when traveling downwind.

69. T_U_T

Nothing can extract energy from the motion of air if the air isn’t moving relative to it.
well, it can.
Imagine a variable pitch wind turbine that can slide along its axis, and has a mechanism to convert its rotation to movement along this axis.
Now. The wind is blowing in the direction of the axis.
Let the turbine spin up while it moves slower than the wind along the axis.
Now. Set its blades straight with wind direction and use the mechanism that convert its rotation to movement along the axis. You will find it thrusted along the axis to faster than the wind velocity.
Now. The turbine will slowly bleed speed because of air friction. Once you slow down to wind speed you reverse the mechanism, causing the turbine to spin up while slowing your movement further below wind speed. Set the blades again to working position.
Now. Again wait till the turbine spins up, and repeat the whole thing. and you will find that its average velocity along its axis is faster than the wind speed via ratchet like action of the turbine blades.
So. You can attach it to a vehicle via springs to convert its jerky movement to smooth .
And voila. you have a vehicle that travels straight downwind faster than the wind itself.
Of course it is not perpetuum mobile because it leaves pockets of slower moving air in its wake and thus always pays its tribute to both energy and momentum conservation.

70. Mu

Actually you could engineer a vehicle that most likely could perform the trick in real life, so not under controlled conditions ;).
You can construct a vehicle that moves into a headwind by having a propeller being driven by the wind and drive wheels that pull the device into the wind. You can spin up with brakes tight, and then accelerate.
To use that to move faster than downwind, your system will need to change pitch at the moment of equal force – and you must catch the right moment where the wind drops so you can get from push into pull mode without stalling at zero relative speed between wind and vehicle.
Which is why it would not work with a fixed wind speed under controlled conditions, e. g. a wind tunnel, since you would never get past the dead point.
Just remember, like taking with a sailboat, you’re just engineering around the the barrier, not actually breaking it.

71. Scott

Mark,
I if I was the sort to go in for conspiracy theories, I’d say that you are link-baiting us.
C’mon already, you are trying to prove a negative! No vehicle that you can think of can draw energy from still air and a moving surface, therefore it is not possible?
Clearly the cart appears to refute your hypothesis — isn’t it time to adopt a neutral stance?

72. Mark C. Chu-Carroll

Scott:
I’m not saying that there’s no way to draw energy from still air and a moving surface. In fact, I’ve repeatedly explained how the vehicle in the video is drawing energy from the moving treadmill.
But the basic claim isn’t “You can draw energy from a moving surface in still air”. The claim is “You can draw energy from the motion of the wind, even when the wind isn’t moving relative to you”.
And proving a negative really isn’t a problem. Scientists are used to it. You can’t move faster than the speed of light. You can’t
create energy out of nothing. You can’t create a system where momentum isn’t conserved. You can’t create an object that accelerates without a force acting on it.

73. jhn

Mark,
A cart going windspeed is exactly the same as a cart that is surrounded by still air with a surface moving beneath it. They have only ever claimed to be drawing the extra bit of energy out of the moving ground beneath them. Maybe they’ve phrased it poorly
Unlike a car, the cart is not powered by its wheels, but primarily by the wind. Drawing energy from the moving ground is not like a car getting extra energy for moving from moving, which is what some people seem to assume. That would be a perpetual motion machine. But like an airplane, the primary purpose of the wheels is to reduce friction, not provide any power. There is no problem with taking a tiny bit of energy from a moving belt beneath you and using it to go slightly faster than the wind is pushing you.

74. Mark C. Chu-Carroll

Dave:
As I’ve said repeatedly: sailboats are working by a combination of sail effects and keel/hull effects. By working at an angle to the wind, the wind is always moving relative to the sail. The keel or hull of the boat is in the water, and it’s shaped, hydrodynamically, so that in one direction, the boat can move freely, but at a right angle it it, it’s almost impossible to move. By setting at an angle to the wind, the sail can
draw energy from the moving wind; by having something to push against, you can shift the direction of motion so that the way the sail is pushing isn’t exactly the same as the direction the boat is moving.
But the key is, again: the air is moving relative to the sail. The air must be moving relative to the sail, or it can’t provide any force.
The situation is similar with ice-boats; but instead of the keel underwater providing something to rush against, the skate edge provides that. So again, you work the sail at an angle to the wind – which means that the wind is always moving relative to the sail, and you use the rigid skate edge as something to push against to shift the actual angle of motion.
But again, you’ve created a system where the wind is always moving relative to the sail.
The cart in this supposed experiment is not a sailboat. It’s not tacking against the wind. It’s not operating at an angle relative to the wind. It’s operating in a mode where the air is completely stationary relative to the sail. There’s no way for it to get energy from that.
You can’t draw energy from wind if the wind isn’t moving relative to you.

75. The Science Pundit

I’ve been reading all the comments on both this post and Dave’s and been giving this problem a bit of thought. I’m going to agree with Dave and disagree with Mark. The kernel of Mark’s objection is that when the vehicle is moving at the same velocity as the wind, there is no relative wind from which to extract energy. This sounds good but it overlooks a basic fact: The airfoil (propellers, in this case) is mechanically linked to the bottom of the wheels. While the cart may feel no relative wind, the bottom of the wheels (which are mechanically linked to the propellers) are moving at the same speed as the ground. In other words, unless there is no wind and the cart isn’t moving, there will be a relative velocity > 0 between the wind and/or the ground. If the wind is moving with respect to the ground, and since a part of he cart which is mechanically connected to the sail is in the ground’s reference frame, then the cart can extract energy from that wind. And that is what we are seeing in the treadmill demonstration.

76. jhn

Mark:
“But the key is, again: the air is moving relative to the sail. The air must be moving relative to the sail, or it can’t provide any force.”
But the wind is not moving relative to the two-boat contraption which is the point of Dave’s post, only to parts of it.
Before I can accept your argument that tacking doesn’t count, you need to explain why the telescoping pole argument fails. It seems to give you a dwfttw vehicle, albeit a weird accordion-like one.

77. Scott

Mark,
moving surface in still air == moving air across a still surface
The observer can not tell weather he is on a giant treadmill with “still” air or standing on “still” ground and moving air. There is no test an observer can perform that can reveal a difference, up until you reach the end of the treadmill, of course.
The cart’s prop is geared to the ground in a way that makes the blade surfaces act as if they were stationary relative to the ground; thus they can still act as sails.

78. KJC

The trick it seems to me is the extract more energy from the wind than is used purely to power the vehicle, and this doesn’t seem that difficult. The wind will push the vehicle along at x mph regardless of whether the wheels spin freely or if they are attached to a shaft driving something. If we make the wheels drive something on the vehicle we can extract energy. A simple (theoretical) way to do this is to put up a sail, and from the spinning wheels charge a battery. Once the battery is full drop sail and accelerate away. Explain to me why I cannot accelerate now?
I think the claim is I can draw MORE energy from the wind than is needed to move me along with the wind.

79. SteveM

Unlike a car, the cart is not powered by its wheels, but primarily by the wind. Drawing energy from the moving ground is not like a car getting extra energy for moving from moving, which is what some people seem to assume.
No, the wind is not pushing the cart forward and using the wheels to generate more energy. The wind is driving the propeller which is geared to the wheels at a ratio that should drive the cart faster than the wind. But as soon as the cart reaches windspeed, there is no wind to drive the propeller which stops driving the wheels and instead, the wheels will start driving the propeller which would act as a brake since the propeller would be trying to drive air forward.
Maybe the dual iceboats would work, but that is not what this cart is doing.

80. Anonymous

Re #79:
There is a problem with that if you’re claiming that all of the energy is coming from the wind.
What you’re claiming is, basically: the wind accelerates you. The wheels under you are being spun by friction against the ground. The turning of the wheels turns a propellor, which
accelerates you.
Sounds great: you’ve got two sources of acceleration: the propellor, turned by the wheels, and the wind. The wheels are turned by friction with the ground.
But where’s the friction coming from? It’s coming from the motion of the vehicle. Where’s the motion of the vehicle coming from? From the wind. So what you’ve done is, in some sense, “stored” some energy from the wind in the spinning of the wheels. But how much energy is in the spinning of the wheels? Exactly as much as the wind put into it. And what happens when the wind is no longer putting energy into the spin of the wheels, because it’s not moving relative to the vehicle? Turning the propellor extracts energy from the spin of the wheels – which slows down the spin of the wheels.
Unless you’ve got something adding energy by making the wheels spin – something other than the wind – the moment you’re moving as fast as the wind, you lose your source of acceleration. You can extract some energy from the spinning wheels, but only at the cost of slowing the wheels down.
You’re basically trying to double-dip in terms of energy; you want the wheels to be free-spinning, but you also want to be getting energy from their spin. It doesn’t work.

81. Dave Munger

you work the sail at an angle to the wind – which means that the wind is always moving relative to the sail,
And so it is with the the propeller. Assume the prop is a simple one made of flaps angled at 45 degrees. As long as the rotational velocity is greater than the relative wind speed, the wind will be “pushing” against the prop.
So assume a 10 MPH wind. The force of the wind can move the cart 9 MPH without the assistance of the prop. Now assume at this speed the wheels can drive the prop at an average rotational velocity of 5 MPH. There is an effective additional wind force on the prop of 2.5 MPH. This makes the cart go 2.25 MPH faster. We’re now going 11.25 MPH, in a 10 MPH wind!
Obviously it will take things a while to equilibrate, and the final velocity computation is complicated, but the net result is a car moving downwind faster than the wind.

82. SteveM

So assume a 10 MPH wind. The force of the wind can move the cart 9 MPH without the assistance of the prop …
Watch the video again, there final design is basicly a coathanger with a prop and wheels; a drag coefficient of nearly zero. Without the prop, the wind isn’t going to push that thing anywhere.

83. Dave Munger

Without the prop, the wind isn’t going to push that thing anywhere.
I mean, without the force provided by the rotation of the prop. When the vehicle starts, the prop acts as a sail. It’s only the last little bit of energy to get the vehicle moving faster than the wind where we need to consider the rotation of the prop.

84. SteveM

So assume a 10 MPH wind. The force of the wind can move the cart 9 MPH without the assistance of the prop.
Okay, even assuming that. When the car is moving at 9mph in a 10mph wind there is a net 1mph wind applying pressure to the sail. Now we have the wheels connected to a generator that can power an electric fan. Say we have it balanced so that if we spin the wheels at 9 mph it can generate a thrust that could push the car at 5 mph.
So now, the 10mph wind is pushing it at 9mph, we start powering up the propeller, as the car starts going faster than 9mph, the pressure on the sail starts dropping which makes the wheels slow down which reduces the speed of your prop, etc. At 10mph, the wind is no longer contributing any push and you are just converting your momentum into energy for the propeller and so you slow down.

85. SteveM

I mean, without the force provided by the rotation of the prop. When the vehicle starts, the prop acts as a sail. It’s only the last little bit of energy to get the vehicle moving faster than the wind where we need to consider the rotation of the prop.
You are asking the prop to do two opposing things at the same time. If the forward motion of the wheels is to drive the propeller to provide forward thrust, it has to push the air backwards. But you also want a tail wind to drive prop to drive the wheels forward.

86. Dave Munger

as the car starts going faster than 9mph, the pressure on the sail starts dropping which makes the wheels slow down which reduces the speed of your prop
No. The fan is the sail. It’s the same thing. If the fan is pushing against the wind fast enough to increase the car’s speed by 5 MPH, then the car can go 5 + 10 MPH minus the inefficiency in converting the wind power to fan rotation. The bigger your fan/sail, the more power reaches the wheels, and the more energy you can generate to move your car faster than 10 MPH.
Now I’ve got to stop writing in this thread and create today’s CogDaily post. See you in a few hours.

87. Scott

Steve,
A lot of people look at this and think that the wind is driving the prop and the prop is driving the wheels. That’s not it — the wind is pushing the cart as a whole (via the prop, even when the prop is standing still).
The moving car turns the wheels, and the wheels turn the prop. The motion of the prop cancels out the movement across the ground which allows the prop to act as a stationary sail even though it is moving across the ground.

88. KJC

We know that the cart will continue to accelerate until the force exerted by the wind equals the frictional force of the wheels.
Let us start with a simple cart with no propeller. We can (theoretically) increase the effective sail size to overcome as much friction as we like until the speed equals 10mph.
Now add the propellor, the frictional force will increase as the wheels need to spin the prop. so we need to increase the effective sail size to overcome that force (or our original sail had to be oversized).
It would appear from experiment that the sail effect of the propeller captures enough energy to overcome the force of the wheels both with the ground and driving the properller, and that the driven properller can convert that extra force into forward motion more effiecintley than the drag it creates (which is what a properller does).

89. Mu

I think the big issue here is – this is a math blog. If you model the vehicle as purely wind driven, then at some point your force F goes to zero at Vwind = Vvehicle, so acceleration a goes to zero, so you can never go faster than the wind. Now ,on a Mythbuster type blog, where you can engineer around the model, you can find multiple ways to get around F = 0 by either storing some energy to push the vehicle through the dead spot or use variations in the wind speed to never stall your progress. In which case you get pretty videos, but the effect you claim (wind driven downwind faster than the wind) is not what is really driving your vehicle. It’s a bit like anti-gravity. Yes, the item is floating, but that doesn’t prove it’s anti-gravity.

90. Scott

Mu,
It think it should be clear that myself and the other in-favor commentators are not saying that this device wins by a technicality. It does not store energy or rely on gusts.
Rather, we are saying that it continues to gather energy from the difference in momentum between air and ground even after it has reached airspeed and beyond

91. rob

the origial video is not on youtube anymore, but i watched the one linked to in the comments.
there is nothing amazing about this vehicle. nothing. anyone who thinks it is amazing and moving faster that air does not have a good understanding of kinematics.
first off, there is no “wind” in this video. not the kind of wind you think of that moves windmill vanes and sailboats. that sort of wind has a continuous velocity profile that exists independent of the object you put into it. if you take a sailboat out of the water, there is still wind blowing around at sea. if you take the vehicle off the treadmill, you will find no wind, since the vehicles fan is not there to create it.
in this video, the “wind” is actually thrust produced by the vehicles’ fan. this vehicle is acting like a rocket. but instead of shooting hot gas out the back, it is shooting air. the reaction force is the force that keeps the vehicle moving on an incline relative to the treadmill belt.
also, the power source for this fan driven vehicle is the motor on the treadmill. there is not other energy input into the system that is being converted into kinetic energy.

92. Mark C. Chu-Carroll

No, I don’t intend to overturn Galileo. I just claim that that the treadmill is not a valid reproduction.
Physics shows that you can always change your reference frame, and everything should
stay the same, despite the change in point of view. A vehicle moving on the ground
at 10 mph, versus a vehicle stationary on a treadmill moving 10mph is really the same thing – provided the setup of the treadmill is really identical to the ground with respect to all other forces.
The thing is, it’s not.
Like I keep saying, the treadmill setup is not a valid equivalent system. The treadmill is sloped, and the rotation of the treadmill is used to spin up the wheels of the cart; and the cart is only being allowed to move a very short distance without being tapped, pushed, or otherwise manipulated.
And as I keep saying: the ultimate question here is: where is the energy coming from? If the vehicle isn’t moving relative to the air, then how is the “wind” accelerating it?
This really isn’t that complex. Where’s the energy coming from? If it’s truly a wind-powered vehicle, then instead of arguing about the treadmill, explain *some* mechanism which allows the wind to transfer energy to the cart, when the air and the cart are stationary relative to one another.

93. rob

also, i would like to add that this system of treamill and vehicle is not equivalent to the same vehicle sitting out on a flat surface outside in a breeze.

94. Scott

You are hung up on the angle of the treadmill. That’s only present to slow the machine down. It works better without the angle.
At 0:27 in the video below, the bubble level shows a flat surface, and the result is that the cart climbs even faster.

Better?

95. KJC

Mark
Increase force required on the “sail” by attaching the wheels to a generator, now the wind has to push the cart and generate electricity. That is allowed right?
Use electricity to spurt forward; wind force = 0, battery becomes depleted but we have move ahead of a balloon in the wind. Battery dies and we slow back to the speed of the wind but we remain ahead of the balloon?
This would not work Theoretically?

96. T_U_T

I think i understand it now. It really can work. Even on perfectly flat treadmill and potentially even make the vehicle faster than the treadmill itself. (which would be euqivalent to vehicle moving faster in the direction of the wind than the wind itself.
So, perfectly flat treadmill ( no slant no slope ) IS equivalent with steady wind blowing at constant speed on a flat plain.
Now have a vehicle that stands still in respect to air, make its propeller spin and make the ground under it to move against the direction of thrust of the propeller. the friction force in the propeller and in wheels, gearbox, etc. would transfer to the wheels as force trying to to drag the vehicle with the moving ground. The ground would feel the exact opposite force because of newtons laws, and we know that P=v*F so the energy would be coming from the moving ground causing it to slow down infinitesimally

97. The Science Pundit

Mu:

If you model the vehicle as purely wind driven, then at some point your force F goes to zero at Vwind = Vvehicle, so acceleration a goes to zero, so you can never go faster than the wind.

The problem above is with your model. Your model is ignoring the ground. When Vwind=Vcart, then the cart is moving with respect to the ground at the same velocity with which the wind is moving with respect to the ground. It is from here that the cart gets the energy to power its propellers and outrun the wind.
But where does that energy come from?
A)The wind? I claim yes!
B)The ground? If we assume that in the case of the treadmill experiment, it accurately simulates motion at Vwind=Vvehicle, then that’s just silly.
C)Stored energy? Where is that energy stored? Why isn’t that energy depleted once the cart gets to its equilibrium speed with the wind? If the cart can accelerate faster than wind speed (which we have empirical evidence that it does) because of stored energy, then it should begin to deplete that energy as soon as it hits wind speed. If that were the case, then the cart would hit a maximum speed, then slow back down to wind speed. During this period it would twice pass through speed Vm (where Vm = (Vmax-Vwind)/2 for Vmax>Vwind), once where it’s accelerating because it has stored energy, and again as it’s slowing down because that energy is depleted. So how does it know which Vm it’s in? Where is this stored energy source?
Clearly the cart is not accelerating due to stored energy (unless you can identify the energy storage source that satisfies the above example of Vm). It obviously isn’t getting the energy from the ground (in the road experiment–and the treadmill experiment is a valid laboratory simulation of that). The only possibilities left are that the video is fake (which I think everyone here agrees is not the case), or the energy came from the wind.
It seems pretty simple when you put it in those terms.

98. SteveM

Put an ammeter on that treadmill, if it not supplying any energy to the cart then its current draw should stay the same or even go down since the cart’s forward motion would be helping the belt spin.

99. John

Yay! I love the treadmill trick! I had a science teacher who showed the class a tape of some TV show he’d spotted showing this exact same gimmick, declared that it was, indeed, possible; without strings or support for the vehicle. He gave us the task of sorting out how. We learned a lot about physics that day, the students leading the discussion for once, often posing difficult questions which our teacher helped us research. No matter how deeply we looked, we students never did figure it out. Our science teacher was overjoyed at the chorus of groans from the students when he revealed the method as class ended…….
Can’t see it in the video.

100. Scott

“If the vehicle isn’t moving relative to the air, then how is the ‘wind’ accelerating it?”
Parts of the vehicle (the prop) are always moving relative to the air because they are mechanically coupled to the ground. No energy needs to be expended in order to couple the prop to the ground other than the costs of mechanical friction. The results is that the prop acts like a sail when it apparently should not.

101. The Science Pundit

The cart is allowed to accelerate on the open road under the wind power of a 10mph wind until it reaches 10mph. How is the road moving under the cart at this point any different from the treadmill simulation? As far as I can see it’s NOT.

102. KJC

Science Pundit, in the treadmill case the energy is coming from the ground (belt) which is of course ultimately coming from the electricity. The change of reference means that the large energy source moves from being the wind to being the road (belt, hence electricity)
SteveM The ammeter will go up (slightly) the energy required from the treadmill is now the energy to move the belt and move the cart forward, this is to analogous to the increased number amount of air that must be “slowed” (versus the amount just to go at wind speed) by the prop in the moving on the road case

103. SteveM

This is where equivalence breaks down. The speed of the treadmill is not set by the cart, it is set to maintain a fixed velocity regardless of the load on it (within reasonable limits), that makes it possible to extract energy from it that would not be available “in the wild” on the open road in a breeze.

104. KJC

SteveM. Sure there is plenty of extra energy in the wild. The wind has much more energy than is needed to move the cart.

105. SteveM

John,
no the treadmill is not going backwards. you can see which way the cart’s wheels spin when it is first placed on the treadmill and they appear to be going the correct direction.

106. SteveM

KJC, but it is not getting it from the wind. It has to get it from the ground. That is what the treadmill shows. The treadmill drives the wheels that drive the propeller to push the car forward faster than the treadmill goes backwards. There is no wind. Only the thrust of the propeller.
The wind in this scenario is just as much a red herring as the treadmill was in the airplane takeoff myth. In that myth the only thing that mattered was the airplane velocity relative to the air, not the ground. Here it is just the opposite, what matters is the vehicles ground speed not its airspeed. It is the relative motion of the cart to the ground that drives the propeller.

107. KJC

The energy is in the relative movement of air and ground. Which is colloqiually called wind in the road sense and the movement of the belt in the treadmill sense.
What both are showing is that in a steady state of 10 mph relative motion between air and ground you can move a cart faster than 10mph relative to the ground or 0 mph relative to the wind

108. John

John,
no the treadmill is not going backwards. you can see which way the cart’s wheels spin when it is first placed on the treadmill and they appear to be going the correct direction.

In my science class, it was! Doesn’t invalidate the story, or the present application of the basic childhood lesson learned. It’s a simple thing to point a treadmill downhill, even simpler to shim a spirit level.
Snip the gadget’s propeller linkage, and the little cart still rolls merrily along on the treadmill without all the propeller action. Just gotta get the angle and speed coordinated to match the drag presented by the cart, and *poof*, perpetual motion machine! Kind of…

109. Scott

The energy source is the difference between the airspeed and the ground speed. It doesn’t matter if the “ground” is a treadmill.
This device brings the momentum of the air and the “ground” closer to one another, and pockets the difference.
In the treadmill example, the air ends up acquiring some of the speed of the treadmill. If you let the treadmill experiment run long enough, the air in the room might start traveling in a loop such that the air touching the surface of the treadmill was moving at the same speed/direction as the treadmill itself — at which point the cart would no longer be able to recover energy because there would be no speed difference to exploit.
BTW Mark, I was serious about that whiteboard in NY offer, if you are in NY.

110. Eric

#105:

The problem above is with your model. Your model is ignoring the ground. When Vwind=Vcart, then the cart is moving with respect to the ground at the same velocity with which the wind is moving with respect to the ground. It is from here that the cart gets the energy to power its propellers and outrun the wind.

Wait, what? A vehicle gets energy from the ground when it’s rolling? Are you serious?
It’s true that the vehicle already has some kinetic energy which can be transferred to other forms – for example, by connecting the wheels to gears which are connected to a propeller, or by connecting the wheels (or brakes if you have them) to a generator, like many hybrid cars do. But the ground is not an energy input.
If the energy comes from the fact that the cart is moving with respect to the ground, then it is the cart’s kinetic energy. So if the cart is using up its kinetic energy to make the wheels/gears/propeller turn which propels the cart, that energy is immediately converted back into the cart’s kinetic energy, with a slight loss from inefficiency (friction and such).
So tell me again where this magical energy source is? It’s obvious with the treadmill that yes, it is providing energy to the wheels because it (attempts to) runs at a constant speed regardless of load, as pointed out above. But the ground does no such thing. So what’s the energy source?
#103:
The generator would cause drag on the cart which would slow it down more than you could gain back by using up the energy in the battery. Otherwise you’d have a perpetual motion machine – take the propeller off and it becomes obvious.

111. Mu

SciencePundit, you still haven’t grasp the problem. The question is not if you can accelerate into a headwind (you can), the question is how do you get from a tail wind condition to the headwind condition without stalling. The reason you need stored energy or variable wind conditions is that you have to break through the Veff = 0 condition. Once you are there, you have wind and can extract energy again in one form or another. So you can deplete all your stored energy for that one moment, and never need to get any again unless you drop back to a tail wind.
The treadmill experiment btw is a simulation of the head wind part of the problem, and avoids the transition phase.

112. KJC

Eric? How would the generator cause drag? The force of the wind can be as high as we need to be we just make the sail bigger to overcome any extra resistance force caused while generating.
This gains more energy form the wind than just allowing the cart to move along with the wind. There is much more energy in the wind than is required to push the cart along.
If you are not happy with me moving can I at least use that extra enery to power a light on the cart? Now we have cart moving at windspeed and a light using enrgy. I didn’t have to magically create any energy there just extract more from the wind.

113. SteveM

The design of the mechanism extracts energy from the difference between the vehicle and the ground, not the difference between the air and the ground.
The way it is set up for the treadmill, when the wheels spin forward, the propeller thrusts backward (driving the cart forward). Now hold the cart up in a tail wind, the propeller will spin backwards driving the wheels backwards. That is why it needs to be in still air; either on a treadmill or moving forward at wind speed. But the thrust is coming from the wheels, not the wind.

114. KJC

No SteveM in a tail wind the middle of the prop acts like a sail propelling the wheels forward, the wheels turn the prop, just like on the treadmill and that cause the extra thrust. The question is is the “sail” capturing enough wind to power both cart forward and prop via the wheels.

115. rob

people seem to think the treadmill’s moving belt is necessary to simulate the effect of the car moving a long distance in a confined space.
do a thought experiment:
get a 100 meter long ramp, at the same angle as the treadmill. push the car up the ramp a bit to get the propeller moving so it can provides some thrust for the car (like in the video). then let go.
what will happen?
also, i repeat: there is no wind.
when you run on a treadmill in an enclosed space, do you feel a wind blowing? in the absence of some fan to cool you off, you don’t feel any wind. wind is not a source of power for the car in the video.
the source of power is the treadmill motor. the belt causes the wheels to turn. the wheels turn the propeller. the propeller pushes the air.
the car is actually *supplying* power to the air to *create* the wind. it is not receiving any power from the air.

116. scott

SteveM,
Indeed, if the cart was held off the ground in a breeze, the wheels would turn backwards. So don’t do that!
Put it on the ground, apply a breeze, and the wind pushes the device forward and the gears drive the fan against the wind. The prop is acting as sail, not a turbine.

117. John

So what happens when power is cut to the treadmill? Does the contraption shoot off the end to the right? Probably…
How about if I were to drop a marble on the treadmill during the experiment? Would it roll happily alongside the gadget, despite lacking any belts or propellers? – Or when the treadmill is brought to a halt, does a marble placed on the stationary surface roll to the right?
This isn’t hard. It’s a silly trick people have been doing for decades!
Kids do it all the time: Just point Mom’s treadmill downhill, and your Hot Wheels roll magically along without going anywhere!

118. The Science Pundit

Eric@#118
Please read my entire comment before you accuse me of saying the exact opposite of what I actually said.
Mu@#119
I don’t think you understand. There is no magic moment you speak of. If you want to know where the cart gets its energy when Vwind=Vcart, it’s the same place it gets it when Vcart>Vwind: the wind is moving faster than the ground, and the cart’s propeller is (indirectly) touching the ground and therefore moving in relation to the wind. There is no magic threshold at Vwind=Vcart.
Also, in the treadmill simulation, (a) there’s no wind in the room, and (b) the cart is placed on the treadmill so that wheel speed matches belt speed. It is in fact a simulation of this transition moment you speak of.

119. jhn

Another way to move downwind faster than the wind, using only the wind as your power source, is to use electricity.
Imagine a sail-powered railcar, and a track that is lined up with a constant wind source. It is very efficient and friction-free, so if the wind is 10 knots it travels at 10 knots also. From the perspective of the railcar, there is no wind.
Next to the rail is a series of windmills. Each generates electricity and powers the rail. The railcar draws power from the rail and adds to its speed. It travels above 10 knots, relative to the rail. The only power source is the wind. We’re just harnessing that same wind in a few ways at the same time.
It is very easy to get power from wind into a vehicle that is not moving relative to the wind as long as, like Dave Munger says on his blog, something somewhere is moving relative to the wind.

120. hibob

hey folks:
The important velocities are those of the prop blade w/r/t the air it is encountering. Since the blades are spinning quite nicely at Vwind = Vcart, the magnitude of those vectors range from zero at the center of the prop to several times Vwind at the tips.
So let’s just drop the “there’s no way for the air to push anything” arguments. Doesn’t matter if it’s in a windtunnel or on a treadmill, the thing to remember is that the parts that do the work are moving w/r/t to the air even if the cart itself isn’t.
I really don’t see this getting answered with non-mathematical thought experiments. Either someone sticks one of these in a wind tunnel, on a really big treadmill, or in a proper computer model.

121. hibob

Rob #123:
yes, there is a wind: you’ve accelerated the cart so it now has a velocity with respect to the air, i.e., it’s experiencing a headwind. For any of these models to work, there needs to be wind w/r/t to the ground or belt, not w/r/t to the body of the cart.

122. rob

the first thing i would like to point out is that the video evidence only shows that the carts are able to move outdoors using wind power harnessed by a propeller. nowhere do we see any evidence that the carts are going faster than the wind. there are no measurements of wind speed and no measurements of the carts speed. we are simply told that the carts are moving faster than the wind. until we are given measurable and reproducible data, we can only conclude that wind makes the carts move. we cannot reasonably conclude anything about the cart’s velocity, either the magnitude or direction, relative to the wind.
second. the treadmill experiment is not equivalent to the outdoor experiment. in the outdoor experiment the cart is getting power from the wind and moving on what appears to be level ground. however, in the treadmill experiment it is on an inclined plane and the treadmill is the source of power. the moving treadmill belt turns the wheels which turns the prop which then blows air, like a pusher prop on an airplane. the treadmill/cart combo is a rube goldberg fan.
third, people are incorrectly interpreting the role of the wind in the treadmill experiment because they are not rigorously defining the system. define the system as the cart itself. next, identify all the forces and decide whether or not they cause work to be done by or work to be done on the system. in the outdoor case, the wind is an external force that does work on the system, causing it to move. in the treadmill case the wind is caused by the system doing work on the surrounding air, causing the air to move. since this wind is caused by the system doing work on the environment, you cannot also identify it as a source of work done on the cart. this defies energy conservation.
fourth. there is a lot of confusion of basic physics in the comments. the terms reference frame, momentum, energy and velocity are all used haphazardly to concoct alternate explanations. proposing these explanations without backing the assertions with calculations and data is no more compelling than the authors poor video evidence.

123. not dave

Rob,
I just posted something similar on Dave’s blog.
A treadmill is how you simulate moving downwind at windspeed: it is how you simulate the situation at issue, where the claim is you can go from no windspeed relative to the cart, to an actual slight headwind.
If you want to simulate racing along with the wind, the treadmill is how to do it. You’d need a wind tunnel for anything else.
The criticisms of the treadmill and the false claims that “you can’t move downwind faster than the wind, powered only by the wind, EVER” are the hardest parts to swallow about the denier’s arguments.
The rest of your points have been addressed, allegedly with links to calculations, in the BoingBoing thread by the builders of the cart.

124. Dan L.

I wish I could have made it through the thread, but there’s too much idiocy.
Let’s make this simple, people: consider that momentum is always conserved. How does a wind-powered vehicle accelerate? It changes the momentum of the wind particles. By doing this, it changes its own motion. A change of momentum is a force (F=dp/dt).
Now suppose you have a windmill or airfoil moving at velocity V in air moving velocity V. Well, there’s no way to transfer momentum from the air to the airfoil in this condition. Try sailing (even at a broad reach!) on a windless day. It doesn’t work. Try powering a wind turbine on a windless day. Doesn’t work.
Now, consider the “ground” scenario where you place this device on the ground in a 10kph wind. Fine, the airfoil can redirect the wind, changing its momentum and gaining some for itself. But here, the reference point is at rest with respect to the ground, at at t=0, to the cart as well.
In the case of the treadmill experiment, the reference frame moves WITH the treadmill, but the cart does not. Due to the friction between the two, this will spin the wheels. As soon as momentum is transferred from the treadmill to the wheels, this version becomes different from the ground experiment. There is a change in momentum that is not due to the wind. These scenarios are NOT identical, and the treadmill experiment is no good. Now, if the cart started at the same speed as the treadmill, we would be fine — but it doesn’t. It starts at some other speed and then gains momentum from the treadmill. Experiment fail.
Some other people had some other clever ideas, such as a turbine propelling itself back and forth in the direction of the wind. You are forgetting that in the case of, say, a wind turbine, there is a net force on the turbine’s mooring, not just the turbine itself. Essentially, the wind hits the blade at an angle and a force is applied normal to the angle. As a result, there is a slowing-down in the windward direction and a speeding-up in a perpendicular direction (that is, the wind changes direction). These slowing-downs and speeding-ups are, by conservation of momentum, opposed by the perpendicular force on the turbine (which drives it in the first place) and a net leeward force on the turbine, which is slowing down the wind in the windward direction.
This isn’t difficult people. You can’t get something for nothing. Mass times velocity in any given direction is always constant except under the influence of an outside force.

125. Mary

I posted this in response to Dave Munger on his blog, but I figured I’d posted it too because, by damn, I put this much thought into it, I might as well post it everywhere. Can’t let someone be wrong on the internet, after all.
The parts in italics are from Dave’s post.
So assume a 10 MPH wind. The force of the wind can move the cart 10 MPH without the assistance of the prop.
Great, so now we are in the situation where the air is not moving relative to the center of the propeller, right?
Now assume at this speed the wheels can drive the prop at an average rotational velocity of 5 MPH.
So the wheels are forcing the propeller to turn though air which is motionless relative to its center of mass. This is like powering a fan. The air is speeding up after it hits the propeller (in the cart’s frame of reference.) The air is gaining energy from the prop, not giving energy to the prop, in this scenario. So where is that energy coming from? Right — the wheels. Which are now slowing down as they give up rotational energy.
There is an effective additional wind force on the prop of 2.5 MPH. This makes the cart go 2.5 MPH faster. We’re now going 12.5 MPH, in a 10 MPH wind!
There’s an effective force on the car from the prop, but it’s a drag force. Just because the propeller is moving doesn’t mean it’s providing thrust. Analyze this in the frame of the prop’s center of mass. Does the air have more energy or less after interacting with the prop? If the answer is more, then the prop is a drag on your wheels, not a power source for it.
Now assume at this speed the wheels can drive the prop at an average rotational velocity of 5 MPH.
If the wheels are driving the prop, then the prop isn’t driving the wheels.
If you could use the “headwind” created by your own motion to power you — well, the faster you go the more “headwind” you create, so the more power is available to you, right? What’s the limiting velocity? How fast can you go? And from the perspective of someone on the ground, where did you get the energy to do it?
Maybe you can go DWFTTW. Maybe sailboats connected by a telescoping chair could do it. Maybe a rotating propeller could do it. I don’t know. I’m still agnostic on that. But you can’t do it with just a propeller geared to some wheels. If you could do that, propeller planes wouldn’t need engines. Shoot them out of a sling shot and then their own apparent headwinds would power them, right? But it doesn’t work that way. The propeller on a plane imparts momentum to the air molecules it collides with, and is pushed forward by the reaction force. But it takes an engine, and fuel, to allow it to keep doing that.

126. Scott

Mary,
Could you please read my previous posts? The cart is pushed by the prop acting as a sail. The forward momentum of the cart turns the wheels. The wheels turn the prop. If the prop were a sail, it would stop pushing the vehicle when it matched the wind speed, but because it is geared to the ground, it acts like a stationary sail even when it is moving as a whole because the surfaces of the blade are still moving relative to the wind.
The spinning prop is made of parts which are tacking!!! The prop is the two sailboats condensed into a simpler form.
At this point, all of you doubters are just refusing to accept something which has been explained ad nauseam and proven experimentally by two different teams, both of which have posted videos. It’s hard, so you are retreating into denial.
It’s complicated. But it works. There is no free energy involved; just an application of the principles of sailing which defy intuition.

127. jhn

Mary,
I personally only got involved in this argument when people were claiming that DDWFTTW itself is simply ruled out by the laws of physics. This is false: as David points out, there are a lot of ways to exploit the motion of the wind relative to some part of a system and transfer it to a part that moves DDWFTTW. The telescoping chair is one way around it. (More practically, if a vehicles’ VMG is greater than wind speed I hardly see why it matters if you travel in a straight line.)
That said, I don’t trust my physics intuition. I think balls should travel in a spiral after leaving a spiral tube. So I’m an agnostic about the cart (though the criticisms of the treadmill model are wrong).
So, I appreciate your acknowledgement that DDWTTTW is at least maybe possible. The last we heard from Mark, he did not address the issue that the wind may be moving relative to one part of a system, but not to another, and so the whole system might travel DDWFTTW.

128. Spork

WOW – talk about a bozo! Mark C. works about 1/2 mile from me, but would rather ridicule me and call me a bozo than to ask for a first hand demonstration.
I have described exactly how this works an untold number of times. I’ve posted the parts list and build notes, and have offered to demonstrate it to anyone that would like. I’ve even agreed to build 12 of them for people that have requested them.
It’s not perpetual motion. It’s simple physics, easily explained to anyone with an open mind. I challenge Mark C. to have a look for himself.

129. spork

Well, now I’ve actually read this brain-dead blog. I can’t believe what a deep stupid-hole Mark C. has dug for himself – and then voluntarily jumped in.
“What this device claims is that if you start the vehicle in a wind, once it’s started, if it’s placed it in non-moving air (that is, the situation which is equivalent to when it’s accelerated to windspeed), it will continue to move, and even to accelerate.”
This is MARK C’s claim – not ours – and it’s WRONG. This vehicle is wind powered. No wind – no go.
Mark C also offers this gem:
“Everyone should be able to understand the physics involved here. My third grade daughter can understand this”
To which I say – yes, she probably can. Perhaps you should have her explain it to you Mark.
In the end I have to suggest that if you want a DWFTTW cart, you should ask a 3rd Grade girl, or someone with an M.S. in Aero – but definitely not a some hard of thinking geek with a PhD in Computer Science.

130. zafner

Oh, man.
“Bozos”.
You are going to look so ridiculous when you figure out you’re wrong. I’m not sure whether you should be putting the salt and pepper on your hat, or your foot, or some crow, but you’re gonna have to improve the flavor of whatever it is you’re about to put in your mouth.

131. cm

spork,
This is MARK C’s claim – not ours – and it’s WRONG. This vehicle is wind powered. No wind – no go.
So you think your cart experiences a wind when its at windspeed?

132. Mender

Since this is “Good Math, Bad Math”, I felt I should offer this:
The source of the energy needed to propel the cart to a specific speed is the momentum of the air mass in relation to the ground that the cart is rolling on. One thing to remember, the propeller is always trying to slow down the speed of the air mass around it in order to harness that energy. A wind turbine slows the air to generate electricity and the amount of power from that momentum change is well documented. The cart also harnesses that energy but in a slightly different way.
Here’s a link to the article published about a DDWFTTW prop cart that was tested outdoors and on a treadmill. In the article, data is provided that can be analyzed.
http://www.ayrs.org/DWFTTW_from_Catalyst_N23_Jan_2006.pdf
The method used for measuring the thrust and drag is listed in the article.
The amount of forward thrust was taken directly from the scale which was attached behind the cart to restrain the cart from moving forward off the treadmill. That showed the force directly at various speeds. The readings were taken at 1 mph intervals between 4 and 10 mph.
The next phase of the test was conducted with two changes: the belt was repositioned so that the propeller turned the opposite direction from the wheels, and the scale and tether repositioned to restrain the cart from moving backwards off the treadmill. The reading from the first test was subtracted from the readings from the second test to remove the imbalance, then that result was divided by two to show the total drag of the cart during operation.
By adding the thrust measured at each point to the drag measured and calculated at each point, the difference between the drag that the cart produced was compared against the thrust for each interval.
The reading at 4 mph was zero and the drag part of the test showed 185 grams. Since the drag part showed the sum of the drag in two directions, the drag in one direction with all systems functioning and accounted for was 92 grams. A thrust of 92 grams balances the drag at that point for a lift to drag ratio of 1:1.
At 10 mph, the measured imbalance was 150 grams and the calculated drag was 402 grams, giving a thrust of 552 grams or a L/D ratio of 1.37:1.
I checked a wind energy calculator to see if the energy needed to provide the thrust measured could be captured from a wind moving at 4 mph given the size of the prop on the cart (40″ in diameter). After doing the conversions and using Betz’ law as a guideline for the max efficiency, the amount of power available to a wind turbine of the same diameter as the prop is 1.63 watts. That translates to 93.37 grams of thrust at 1.78 m/s, quite close to the test figure of 92.5 grams. For 10 mph, the numbers work out to 25 watts or 585 grams thrust at 4.44 m/s, again within experimental error of 552 grams as indicated by the treadmill test.
The small cart in spork and JB’s video tends to back those numbers up, with the 169 gram cart “climbing” a 4.4 degree incline at 10 mph, indicating an imbalance of 13 grams force in the forward direction. That would give it an acceleration rate of 2.47 ft/sec2 on level ground in a 10 mph wind when the cart is at 10 mph. The large cart would have an acceleration rate of 2.12 ft/sec2 based on a weight of about 2300 grams (five pounds). The break even speed on the little cart is 2.7 mph vs the large cart’s 4 mph, so a little better performance could be expected from the small cart.
So it appears that the energy available from the different wind speeds correlate quite well to the treadmill test. There is no mystery force needed to accomplish this, no violation of any physics, in fact pretty normal physics, just applied in a unique way.

133. Anonymous

>> So you think your cart experiences a wind when its at windspeed?
The blades of the prop certainly do – just as the sail of an ice boat does even when it’s on a 45 degree downwind tack with a downwind velocity component exactly equal to the true wind speed.
You may be aware that ice boats can in fact maintain a steady-state course with a downwind velocity component of 3X to 4X the true wind speed.

134. HUDMAN

What’s up with the bozo calling people bozos? So, let me get this straight… You don’t understand it, so the people in the video must be bozos? Am I getting warm here? If you did understand it, would they be heroes?
I’m guessing that if you did understand it, you’d try to be the hero by explaining it. Perhaps there’s a black-hole-like void in your life that can only be filled with name-calling the people that created something that you obviously are too simple-minded to understand.
If you were watching the Wright Brothers attempt take-off at Kitty Hawk, you’d probably say, “Are you nuts? Man can’t fly… Only birds can fly.” The question though is what would you say after you saw it fly? I’m guessing you’d either say it was faked or you’d bow-out and continue onward with your meaningless life.
So… What would anyone gain by proving it to you? ABSOLUTELY NOTHING! But, I’ll make an attempt anyway. Just do me one favor… When you do finally understand, (if that’s at all possible) publically offer an apology to those “bozos” and publically admit that you were wrong. Fair enough?
Now, place the cart on the grtound in a tailwind. The tailwind flows through the prop, and one might think this would cause the prop to turn, geared to the wheels, which make the wheels turn, pulling the cart forward.
But, this is not true because of gearing & the advance ratio. If the wind turned the prop, the cart would actually go in reverse, sailing directly into the wind. But, because of the gearing and advance ratio, the path of least resistance is for the cart to start moving forward, downwind, which turns the wheels, powering the prop, generating thrust. Now, once thrust is generated, the air passes through the prop from the front to the back. The cart does not have to wait until it outruns the tailwind in order to do this.
The propeller, like any thrust producing propeller, now has a low pressure in front of it, sucking it forward and a high pressure behind it, pushing it forward. Once it matches the wind’s speed, and this is the critical part, one might think that there’s no way it can go any faster, because if it does, it would be outrunning its power source.
But, again this isn’t true. Why? Because the whole cart is immersed in a “medium” of air movement, (tailwind) and now there is no relative wind on the cart, in front of the cart, or behind the cart. To the cart, it’s in still air, and just drifting along at wind speed like a neutral buoyancy helium balloon. (Also, just like the cart on the treadmill) But the prop is still generating thrust in this medium, much the same way a plane still generates thrust in a tailwind. For the plane, it’s ground speed = plane speed (air speed) + tailwind speed. For the cart, it’s the cart speed + tailwind speed – drag.
The drag is represented by minimal losses due to friction, and major losses from the wheels because they have to power the prop. The trick is to make the cart as efficient as possible by reducing friction and optimizing the advance ratio through gearing and prop pitch. So even though these guys (bozos) make it look easy, it’s actually a very hard thing to do. But not only is it possible, they (and a few others) have achieved this.
Now, once the cart moves faster than the wind, the relative headwind is now being accelerated rearward by the prop, but again, the prop is doing no magic trick here. It’s a prop, and it does what a prop does, it generates thrust by creating a low pressure in front of the prop and a high pressure behind the prop, regardless of a headwind or a tailwind, or even no apparent wind.
There’s no perpetual motion going on here. The wind is the power source. Remove the wind and it stops. Add a tailwind, and since the propeller is constrained to the wheels, as long as the tailwind is present, the wheels continue to roll, powering the prop, and the prop continues to spin, generating thrust.
I know what you’re thinking, “But if it outruns the wind, it has outrun its power source, and that is just not possible.” Now, keep in mind that the cart is immersed in a “medium” of air, which is the tailwind. So in a sense, once the cart matches the windspeed, the tailwind is not pushing on anything, as the cart “feels” no relative wind at this point. But the prop, using thrust, is pushing on the still air behind the cart. Now, with that said, it’s not entirely accurate to say that the tailwind is pushing on anything, since there is no relative wind to the cart. But if you must think of it this way, then think about this;
Once it outruns the wind, the only thing that has changed is that the tailwind no longer “pushes” on the frame of the cart, it is now “pushing” on the propwash or thrust from the propeller. The thrust is the sail, and since the thrust is accelerated rearward, it is obviously going slower than the cart, and obviously not outrunning its power source. The thrust is colliding with the tailwind, propelling the cart forward, and making the cart outrun the wind. While its doing this, the propeller is spinning faster, providing more thrust, and staying ahead of the tailwind’s speed.
If you’re still having problems wrapping your arms around this idea, consider this. If you have an electric powered prop cart that goes 5 MPH in still air, and you place it in a 10 MPH tailwind, it will go almost 15 MPH. I say almost because of minimal frictional losses. I think we all know why… It is immersed in the tailwind, and always adds almost 5 MPH to the tailwind’s speed using thrust. The electric powered prop generates 5 MPH of thrust rearward which “collides” with the tailwind, giving us our overall speed.
The DDWFTTW prop cart does the same thing, only the prop is powered by the constraint of the wheels instead of an electric motor. I used the electric motor example because most people understand that, but can’t quite apply the same logic to the DDWFTTW cart.
By the way, what’s that all over your face? Oh, it looks like egg… Now how did that happen?

135. Subductionzone

Well it looks like I was a bit late for the party and the doubters when given good arguments have slinked away in shame for the most part. On the original youtube video of Jack’s outdoor cart I was one of the original posters to say what a hoax it was, it seemed obviously towed to me since he never had a downwind shot. Oh well, it is not so bad to admit that you were wrong once you start to see the light, and spork is a real gentleman, he is not an “I told you so sort”, though he does react rather strongly if you accuse him of being a hoaxer.

136. cm

HUDMAN,
So, to summarize your point, on an open road with a wind, with the cart at windspeed:
– the cart experiences no wind (I agree).
– the wheels turn the prop. (I agree).
– the prop pushes off the still air (I agree).
– This speeds the car further (I disagree).
The wheels are powering the prop to do work, and in so doing there is a loss of energy: the wheels must slow down as they impart their energy to the prop. You can not then say that the prop then drives the wheels back, and in fact faster than before–that’s energy from nowhere. (If that worked, every car should have a big propeller on the back to boost the speed of the car for the same fuel)

137. spork

“that’s energy from nowhere”
It’s not energy from nowhere. Let’s consider the energy balance in the ideal case (no friction or other losses). Assuming a 10 mph tailwind, the wheels are rolling on the road at 10 mph, and feel a force of X lbs. This gives them an energy input of 10X. The prop must produce a thrust of X lbs to maintain this condition. However, the prop is not working on a 10 mph wind. When we’re at wind speed, the prop is working on a 0 mph wind. So the ideal prop (infinitely large and slow turning) would actually require 0 energy input to achieve the required X lbs of thrust.
In the real world there will clearly be losses in the transmission, rolling friction, aerodynamic drag, etc. And of course the prop will be of finite size. However, there is plenty of extra energy available to overcome these innefficiencies (10X – 0X in this case).

138. jhn

cm:
Ideally, the cart could still move forward even if the wheels where stationary. The wheels reduce friction and power the prop– that’s it.
Wheels that don’t provide the motive force for the vehicle can slow down without slowing down the vehicle. They skid a little.
Then the turbine does not power the *wheels*. It powers the entire cart, to a degree that exceeds the friction caused by the slightly-slower wheels.

139. cm

spork,
However, the prop is not working on a 10 mph wind. When we’re at wind speed, the prop is working on a 0 mph wind. So the ideal prop (infinitely large and slow turning) would actually require 0 energy input to achieve the required X lbs of thrust.
Yes, at windspeed, it is equivalent to parked in no wind. Right. But that certainly does not mean that any prop requires 0 energy input to create X lbs of thrust, ideal or not.
Imagine a cart sitting there, there in no wind, and you supply 0 energy input to the prop–and yet given what you said you’d expect it to generate thrust? Thrust is doing work, moving the cart, and work requires energy. It requires energy whether you are flying at 500mph or parked or flying backwards at 500mph. Work requires energy. Isn’t that the whole point of this controversy?

140. Chris

An excellent point was made over at BoingBoing: “Amazingly, so far as I can see, no one has addressed the fundamental problem that if the cart transitions from moving slower than the wind to faster than the wind, the reversal of air flow will try to turn the propeller backward, thus tending to stop the cart. It bothers me that so many people are conned by this idea (or con themselves).”

141. Subductionzone

Chris if you go to youtube you can see quite a few of spork’s videos dealing with his carts (the one shown at the top of this page was a crude early version, though it obviously works), look under the user name of spork33. In one you can see an outdoor test where the cart starts from a dead stop in the outdoor air. Now there might be a little truth in you claim because in the video I am thinking of under an even wind it starts and moves forward with no reversal needed. But he restarted the cart in the video and when a strong gust occurred the propeller at first did counterrotate, but it straightened itself out on its own. So a strong gust might overcome the sail/propeller nature of the cart for a short time and try to make it run like a turbine, but since the wheels are also rotating backwards it quickly reverts to sail/propeller mode.

142. idontcomment

cm: In your last example (0 mph) the force on the wheels (X lbs) is 0, so yes, trivially the prop is providing that force.
It helps me to think of the force on the prop in two parts – there is the base part it would get if it wasn’t even spinning by the effect of the wind – this acts to keep it going at almost windspeed. Then there is the additional backwards force it applies by pushing that air (almost still in its reference frame) backwards. The energy to do this, of course, comes from the wheels. The trick that makes this not a perpetual motion problem is that its wrong to equate speed lost due to power generation at the wheels to speed gained through prop rotation. This is because speed lost is taking energy from the ground/wind difference, and speed gained is simply in speeding up the cart (mv^2, friction, that kind of thing). There’s no reason these should be zero sum – for example, you can double your prop size and immediately collect more wind energy. Another way to think about this is that the energy you can collect from the wheels is from your total speed (passive blowing plus active paddling) but you only have to put it into your paddling.
It’s very hard to deny brian’s skateboard example – plant a wind tubine on the ground on an extension cord, use its power to drive your skateboard forward, then uproot the turbine and replant it. bigger more efficient turbine, go faster. clearly energy harvested from the turbine depends on one set of factors, energy used moving it forward on another.
Finally I’ll link to the example I’ve had the most luck with when i try to explain IRL:
http://www.boingboing.net/2008/12/03/directly-downwind-fa.html#comment-345825
hello spork, jb, mender!

143. idontcomment

Chris – check out hudman #145 about the gear ratio for why it doesn’t reverse direction. also my attempt:
It depend on the gear ratio of prop to wheels:
E.g, consider a wheeled cart with a tread mounted on the “roof” (like a wind paddlewheel, in a way!) geared such that forward motion of the top of the tread (Relative to the cart) causes the wheels to propel the car backwards.
For a low gear ratio (lots of tread motion causes small wheel motion), running your hand along the top in a forward direction will cause the cart to travel backwards.
BUT for a higher gear ratio (small tread motion causes lots of wheel motion), running your hand along the top in a forward direction will cause the cart to travel forwards, faster even than your hand (your hand will eventually slide off the back if you stay holding the same part of the tread). I think the breakeven ratio is 1:1 (in terms of tread motion to wheel surface motion) – at 1:1 pushing forward on the tread will not move the vehicle either way (until the wheels slip!!)
Of course I’m assuming a prop/wheel directionality where rolling the cart forwards causes the prop to push air backwards. It is harder to calculate the “gear ratio” with a prop, but the breakeven 1:1 point is where the desire to rotate the blade at the cost of driving the cart into the wind is perfectly balanced with the desire to push the cart downwind at the cost of counter-rotating the blade.

144. T_U_T

I think I found a perfect way to understand the DDFTTW phenomenon by finding a simpler device that does the same seemingly impossible feat. thing A pushing and accelerating thing B at faster speeds than the thing A itself moves.
take two sheets of paper (thick paper makes it a lot more reliable than thin ) and one small coin. Glue sheet A of paper to the desk so that its edge is at 60 deg angle. Now place the other sheet B above the lower edge it so that its upper edge is at 30 deg angle and crosses the lower edge of sheet A somewhere in its lower third. Now, you have a 30 deg angled wedge of empty space between them. Place a coin C in it and start moving the paper sheet B forward. You will see that the empty part recedes quickly and the coin having nowhere to go also moves, and moves in forward direction FASTER than the paper sheet that pushes it.
Now imagine the coin replaced by prop blade, the sheet B replaced by mass of moving air ( the pitch of the prop acting exactly as the edge of sheet B ) and the sheet A with cogs and wheels that cause prop rotation be constrained to forward movement, so acting like the edge of the paper A.
And there you are. You have the DDFTTW device.

145. idontcomment

nice T_U_T! also good is the “paper faster than the paper”

(I’m just plagiarizing the boing boing thread now, thanks for the link futurenerd)

146. Stephen Wells

Any claim that this device can be set going in steady state _at wind speed_ and do anything but slow down is clearly wrong on energy conservation grounds.
Any claims about the wheels powering the turning of the prop and then the prop accelerating the cart is also wrong on energy conservation grounds.
Flywheel effects might give you a transient apparent violation of the above, but not a steady-state.
The only tenable claim is that the device can be set going in steady state, heading downwind at more than wind speed; it then has a large ground speed and meets a small headwind; and that it can extract enough power from this headwind to drive the wheels and continue to move in steady-state. This claim may not be prima facie absurd (it does seem that a vehicle with a wind turbine mounted on it can advance into the wind) but I still want to see a serious analysis before I find it plausible.
Somebody build the two iceboats/captain’s chair arrangement. We all want to see what happens to that one.

147. T_U_T

Any claim that this device can be set going in steady state _at wind speed_ and do anything but slow down is clearly wrong on energy conservation grounds.
Do my faster than paper speed experiment or the faster than paper speed from the link above and you will see with your own eyes that you are wrong.

148. Stephen Wells

T_U_T: I didn’t say anything about your paper-pushing experiments (I think the yo-yo one is good, by the way), and those experiments say nothing about this cart. It’s simply the case that, if the vehicle is travelling at exactly the speed of the wind and it’s in steady-state, the propellor is just a rotor turning in still air, it must be powered by the wheels, and it’s impossible to produce more energy in propulsion from the rotor than you extracted from the wheels to drive it.
If the device is to have any hope it must be run-up to greater than wind speed. I could be convinced in that regime. But at exactly wind speed, no. And people who want to convince us that it’ll work in the above-windspeed regime are doing themselves no favours by insisting it’ll work in the impossible regime.

149. T_U_T

It is not powered by the wheels ! That is the counter intuitive thing ! The wheels merely provide constraint that constrains the rotational movement of the blades to their translational forward movement together with the whole cart.
The wind provides power ! It transfers power to the prop even if it is moving at the same speed as the wind does. Even if it overtakes the wind ! That is the other highly counter intuitive part. ( note that because of the rotation the blades keep moving against the wind even if the center of the prop is not )

150. T_U_T

Other way to imagine it. Take the yo-yo thing that is pushed at say 2cm/s by a paper strip moving 1cm/s, that is the video.
Now, make it inefficient by letting the yo-yo slip a little thus reducing its speed to say 1.5cm/s Now. make it to slip more, till it moves at 1cm/s along with the strip and this is exactly what you wanted. the strip transfers energy to it while it is moving at the same speed as the strip does

151. Stephen Wells

T_U_T: you have just claimed that the wind transfers power to the prop _even if the cart is travelling at the same speed as the wind_. You want to try that again, with fewer exclamation marks, and think about what you’re saying? You’re saying that _still air makes a prop go round and drives wheels_.
Facepalm.

152. T_U_T

Essentially yes. The wind exerts nonzero forward force against the prop even at zero relative velocity to its center ( it can not be at standstill against the prop blades because they are rotating )

153. T_U_T

( translating to the yo-yo. you can have the paper strip transferring power to the yo-yo even if the yo-yo center is at the same speed as the paper strip does. Counterintuitive ? Absolutely. Real ? Absolutely . Perpetuum mobile ? definitely NOT. )

154. ThinAirDesigns

All I can say is there’s going to be a whole lotta of third grade crow eaten by whole lotta phds when MythBusters puts a DDWFTTW on air.
(Oh, btw — spork and I are the “bozos” who built and demonstrated the device. We can walk from where we build these to your work at lunch. Let us bring one over and chat with us … you might find we’re not as stupid as you think)
JB

155. Stephen Wells

T_U_T: I’m not claiming that a rotating prop won’t exert force on still air. I’m pointing out that if that prop is exerting force on the air it must be being driven by the wheels. You appear to be claiming that it’s being driven by the air. You need to rethink or rephrase that.
This is independent of any questions about the working of the device when travelling at anything other than the windspeed. I’m just trying to point out that you cannot coherently claim that still air is driving a prop, when you mean that the prop is driving still air.

156. T_U_T

I think you simply cant get rid of the misleading what powers what simplification.
The wheels dont power the prop neither the prop powers the wheels. the wheels act merely as a energy transformation device which transforms some of the forward force exerted by the wind on the prop to sideways force making the prop spin up.
if the prop exerts force on air still relative to it then by newtons laws the air necessarily exerts force on the prop. The prop is pushed forwards. And also the blades are pushed by air friction against the direction of the rotation, so if there were not the gearbox and wheels the prop would stop rotating. But some of the forward force is transformed to force trying to spin up the prop thus making up for the loses and keep the prop accelerating till it moves faster than the air.

157. Mu

I’m perfectly willing to eat the crow, even without ketchup, if someone puts a DWFTTW vehicle on mythbusters, or provides any other even semi-believable documentation.
But it has to be exactly that, a vehicle that’s wind powered in a steady wind moving in the direction of the wind; not one of the multitude of engineered-around versions using wind speed variations or angle of attack.

158. The Science Pundit

Stephen,
You’re missing much of the picture by focusing on the reference frame of the cart. As has been pointed out before, not all of the cart is moving at the same velocity as the wind (eg. the prop blades, the bottoms of the wheels). Let’s grant you that from the reference frame of the cart the power for the prop comes from the wheels. Where is the power for the wheels coming from? It is coming from the wind (REMEMBER that not all of the cart is steady wrt the wind). By focusing on your reference frame, you get the illusion that there is a moment where cart speed equals wind speed and the wind exerts no force. This is just not so.
Let me demonstrate the problem with focusing on the cart reference frame the way you have by using another example. You are in a car when the driver slams on the brakes. From the reference frame of the car, there is a mysterious force that throws your body forward towards the dashboard. Where did that force come from? This can be explained in that reference frame (as can the cart example if you had been paying attention), but it gets complicated and the illusion is hard to break. However, if you use a more useful reference frame, the problem becomes trivial. What you felt was inertia and the force came from the car’s engine. NOTE: I am not saying that it’s the same principle as the cart! I’m merely pointing out the difficulties when choosing the wrong reference frame to work with.

159. davem

This is all a problem of gearing. The prop is spinning at say 30-40? mph at speed. Being a prop, it’s doing what sailboats do when they go upwind. The wheels are spinning relatively slowly, gearing up the prop.
Where’s the energy coming from? Umpteen million toms of air at 10mph are pushing a few kgs at 15mph. Where’s the energy conservation problem? It’s non-existent.

160. Anonymous

Mu:
>I’m perfectly willing to eat the crow, even
>without ketchup, if someone puts a DWFTTW
>vehicle on mythbusters
Excellent, then I already know how this goes for you ultimately … crow *without* ketchup.
There will also be a second course of crow regarding the validity of the treamill test. You can have ketchup with that round if you wish.
I have one distict advantage on most poster here — I know how the story ends.
JB

161. T_U_T

You’re missing much of the picture by focusing on the reference frame of the cart.
Exactly.
If someone chooses to see the process from the frame of reference bound to the ground he will see the wind to be source of power by pushing the cart.
But if someone chooses to see the process from the frame of reference bound to the chart the opposite happens. He will see the ground moving and trying to drag wheels to opposite direction to be the source of the power.

162. Anonymous

“And people who want to convince us that it’ll work in the above-windspeed regime are doing themselves no favours by insisting it’ll work in the impossible regime.”
Well, I guess you’ve got us there. Clearly nothing will work in the “impossible regime”. On the other hand, you simply made up the concept of an “impossible regime” with respect to this cart passing through wind speed. I assure you it will move at wind speed just fine – trying to accelerate for as long as you hold it back – or will accelerate to greater than wind speed if you simply let it.

163. SteveM

For a low gear ratio (lots of tread motion causes small wheel motion), running your hand along the top in a forward direction will cause the cart to travel backwards.
BUT for a higher gear ratio (small tread motion causes lots of wheel motion), running your hand along the top in a forward direction will cause the cart to travel forwards,

wrong. It is a ratio and signs matter. reversing direction is a sign change and you can’t get that by just increasing the ratio.

164. Stephen Wells

None of the above comments about frames of reference resolve the issue when the cart is running at exactly the wind speed. In that case you have a prop rotating in still air and power for that can only be coming from the wheels turning the prop. Conservation of energy means you can’t possibly get more propulsive power from the prop than you extracted from the wheels. The issue of whether the vehicle can extract power from a headwind (i.e. when travelling DWFTTW) is entirely independent of whether it can extract power from no wind at all, which is what I’m challenging here. Changing from one frame of reference to another does not give you energy for free.
If the air is not moving relative to the cart than any prop rotation involves the prop doing work on the air rather than vice versa, i.e. energy is going from the cart into the motion of the air. If you are extracting this energy from the motion of the wheels over the ground then you’re exerting a braking force on the cart.

165. T_U_T

I hate to repeat myself.
But if someone chooses to see the process from the frame of reference bound to the chart the opposite happens. He will see the ground moving and trying to drag wheels to opposite direction to be the source of the power

166. SteveM

It is not powered by the wheels ! That is the counter intuitive thing ! The wheels merely provide constraint that constrains the rotational movement of the blades to their translational forward movement together with the whole cart.
The wind provides power !

What is going on in the treadmill video. There is no wind. When the cart is placed on the belt, what is driving it forward? The spinning prop is pushing air backwards to drive the cart forward. What is driving the prop? The wheels. What is driving the wheels? the treadmill (ground). The whole premise of the treadmill experiment was that it is a Galilean analog of the cart moving at windspeed. Even if you claim that the wind got it up to windspeed, once it is there the wind cannot provide any energy to the system. There is no force on the prop from the wind. At this point it is entirely the spinning prop exerting a force on the (still) air that is driving the cart forward.
What I would like to see is the cart in a wind tunnel on a treadmill that is controlled to maintain the cart in fixed location (rather than running at a fixed speed). That would answer my contention that the cart is drawing extra power from the treadmill that it would not get on the road.

167. Scott

Stephen says:
“T_U_T: you have just claimed that the wind transfers power to the prop _even if the cart is travelling at the same speed as the wind_.”
Stephen, no one seems to doubt that one can extract energy when at rest to the ground with moving wind. But the opposite; at rest to the air with access to moving ground — impossible!
Ok, imagine a car. On roof of the car, install a treadmill. Connect the treadmill via gears to the wheels, so that the treadmill rolls backwards at the speed of the ground. The surface of the treadmill now is equivalent to the ground. What energy costs have we expended? Only friction, which can be effectively zero.
Now, put some windmills on the treadmill. The windmills are always stationary relative to the ground! They can gather energy from the wind even when the car as a whole could not.
The cart is simplified version of this design. The vanes of the prop are geared to the ground so as to cancel the forward momentum. The prop acts as a sail even when the car as a whole is moving at wind speed.

168. Scott

SteveM:
“What is going on in the treadmill video. There is no wind.”
Steve, imagine you are an ant standing on the treadmill. What do you feel? WIND.
Now if you as an ant could build a little ant sized DWFTTW cart you would be able to accelerate up to the speed of the treadmill and continue off to the right. Like the cart does, in the video.

169. Stephen Wells

“Access to moving ground” isn’t going to solve this problem; if you extract energy from your motion relative to the ground, as in T_U_T’s “ground trying to drag the wheels”, you must be exerting a braking force against the ground. Guess what happens to vehicles that exert a braking force against the ground. There’s also an important asymmetry in that I can slow down _some of_ the air but I can’t slow down _some of_ the ground!
Look, I can see that a vehicle travelling DDWFTW is travelling into a small headwind. If you want to claim that enough power can be extracted from this headwind by the rotor to supply power to the wheels, _show your work_; the claim is not intrinsically absurd but I have tried twice to go over the forces and powers, starting with Betz’ law, and so far I’m not seeing it. If you want to claim that you can take power from the spinning of the wheels and drive the prop for power, show your work; that’s a PMM on its face.
But if you want to claim that a vehicle moving relative to the ground and stationary relative to the air can extract energy from the air- sorry, not buying it. It can do work on the air, but not more work than it extracts from the wheels!
The treadmill videos aren’t helping. For one thing I don’t believe a standard exercise treadmill runs at a sufficiently uniform speed to give you a real steady state. If you’re pressing on the mill to spin up the cart’s wheels and then let go, the treadmill will first run fast for a moment (as you’ve reduced the load when you let go of the cart) and then slow down, probably overshooting; that plus momentum/flywheel effects with the rotor could be giving the behaviour in the video.
Please bear in mind that you could be right about the things’ behaviour in a free run downwind and wrong about the treadmill example; they are different cases; and still air doing work on a propellor would be kind of novel, to say the least.

170. spork

“That would answer my contention that the cart is drawing extra power from the treadmill that it would not get on the road.”
But Galileo, Newton, and Einstein have already answered your contention that you can use any form of instrumentation to distinguish between two inertial reference frames. If they fail to bring the point across I hold little hope for any of us on this blog.

171. Subductionzone

Well you can always tell that someone does not understand the concept of equivalent frames of reference when they want to put the treadmill in a wind tunnel. To go back to the idea of a boat and sailing, most people already acknowledge that a boat can go faster than the wind on a broad reach. Some will even acknowledge that a boat can have a velocity component that is faster than the wind in the winds direction if the boat is tacking, as ice boat racers and high performance sail boats can. Now try doing that without a keel, or in the ice boats case put the runners, skates whatever on ball bearings so that they pivot hopelessly (I know only an idiot would do this), now try to sail faster than the wind, it cannot be done. The wheels of the cart act in a very similar manner to the keel or the skates of an ice boat. They allow the cart to extract the energy caused by the wind blowing over the land. Everyone will acknowledge that there is potential energy in a moving wind, even if you are moving at the same speed as the wind the wind is still there and so is the energy, there is no over unity, this is just a very clever way to extract some of that energy.

172. Mender

Stephen, I provided the analysis of the energy needed to power the actual cart in #143. The next step is to understand how the cart can harness that energy while moving.
The ground is the key to this mystery. In a 10 mph wind, a very large mass of air is moving over that ground. Anything that is in contact with that air mass has the opportunity to harness energy by changing the momentum of the air, just like a wind turbine. F=MA works for the cart, not against it. The only tricky part of this is to understand how the cart changes the momentum of the air as it is moving through it.
Most people don’t think about how the momentum of the air is changed when an airplane flies through the air mass. They know the prop acts on the air but don’t stop and figure out exactly how much.
When an airplane is starting the take off roll on the ground, the amount or mass of air passing through the prop is low and the speed change that the prop inflicts on that air is fairly high. The force exerted on the air mass is high, the mass of the air is low, so the speed change is high.
When the airplane is cruising at 85 mph, the force exerted on the air is less and the amount or mass of air traveling through the propeller is high, so the speed change of the air is low. If the airplane has low drag and a big (comparatively) prop, the speed change of the air affected by the prop will be low; I’ll say that this is 5 mph for the sake of an example. If the drag of the air craft is high and the propeller is small, the amount of air that the prop affects will be smaller while the energy that is needed to overcome the drag of the airplane is higher (let’s say 10 mph), so the speed change of the air is higher than the previous example.
All this to say that an airplane moves much faster than the air around it but has no problem transferring energy to that air. To give an idea of how much the speed of the air is affected by that energy transfer, a simple calculation can be made. The amount of force (F) exerted by the airplane can be calculated by measuring the amount of air affected (M) and the rate of change of the speed of that air (A). F=MA. Good Math, Good Physics.
Let’s see what over-unity for the cart would actually be since a lot of people claim that it somehow is capable of that. If the cart pushed the air back with respect to the ground at a higher speed than the wind is moving forward, more energy would be required than is available from the movement of the air over the ground (note: the ground reference is always the most important reference in this system).
An airplane can do that because it has an onboard power source, the engine; the cart can’t do that because it uses an external power source, the movement of air across the ground.
Since the cart only pushes the air back at a fraction of the ground speed as indicated by the advance ratio, the cart isn’t trying to use more energy than is available from the wind. The highest amount of energy the cart can remove from the air is when the air affected by the prop is brought to a stop with respect to the ground.
It is not an over-unity device. It doesn’t need to be to work as described. It isn’t a particularly difficult concept to understand unless a conclusion is made before the facts are examined.

173. Stephen Wells

The cart on the treadmill is not equivalent to the cart running downwind faster than the wind. It is equivalent to the cart running downwind exactly at windspeed. No amount of handwaving will allow you to _extract_ energy from _zero_ wind. Nor will it let you do more work on the air than you extract from the wheels. Nor will it let you extract energy from the wheels without braking the cart.

174. Mender

Stephen, if you’re having trouble with the calcs for the wind power, show me what you’re doing and I or someone else will help.
The small cart moves steadily against an incline of 4.4 degrees and has a weight of 169 grams. The force needed for that is 13 grams. The treadmill won’t even notice that. In one video the cart runs steadily for almost two minutes with no change in speed or momentum.
There is no flywheel effect.

175. Stephen Wells

Post 143 includes the claim that a 4mph wind gives enough power to propel the cart. Fine. How much power does a 0mph wind provide?
I already said that I can see that a turbine can extract power from a wind blowing into it. But how it extracts power from a wind that is _not_ blowing into it remains a conundrum!

176. Scott

Stephen,
You are being given the explanation, but you’re not accepting it. If you don’t like the “handwaving”, explain the video yourself. Because there are lots of videos, and they show the cart driving uphill against the treadmill all freak’in day. Clearly the cart does extract energy from moving ground and still air.

177. Mender

Stephen, your last post shows that you have decided that the cart doesn’t work. That unfortunately is wrong. You won’t make much headway with this until you decide to learn HOW it works.

178. Mender

My apologies, Stephen, we’re posting past each other. Good, you are trying to understand.

179. Subductionzone

Stephen, you almost have it! Yes the cart running on the treadmill illustrates the cart only running at the speed of the wind. But when it accelerates down the treadmill it is moving faster than the treadmill or faster than the wind. This was not spork’s and J. B.’s best vehicle, later ones were much more responsive. Check out spork’s youtube video’s at spork33.

180. Mender

Now that you’ve read #184, remember that the cart is always in a 10 mph wind, not a 0 mph wind. There is plenty of energy available.
When the cart is moving at the same speed as the wind, the wind hasn’t stopped moving over the ground. All the cart has to do is change the speed of the air around it and it can harness that energy. If the wind stops, the cart has no power source and will stop also.

181. Mender

A quick note, understand what the cart is doing in the wind first. Once you see how that is working, the treadmill test is much easier to understand. Do that after.

182. Subductionzone

Stephen, how does an ice boat sail on a broad reach with the downwind vector of its velocity faster than the wind it is still sailing in? That is a bit of a conundrum also and though I have seen it properly explained I cannot do it myself, but it is a common everyday event and no one is complaining about “over unity” or perpetual motion when it comes to these boats. The propellers are acting as the sail of a boat does on a broad reach, and the wheels are the equivalent of the keel, or the skates in the ice boats case.

183. Dave Munger

Just as a reminder to those late to the conversation, I’m consolidating the objections on my blog and offering responses to them. If you’re a skeptic, it helps to see if your argument has already been debunked before repeating it.

This has been hashed and rehashed. The people that think this is impossible are confused about what energy (or work or power) really is. If you push against a wall with a 100 pound force for five minutes, how much work have you done? None. Likewise, if a very large propeller pushes against still air with a 100 pound force for five minutes, how much work has it done? If you make the propeller large enough (and turn it slowly enough), the work done can be made arbitrarily close to ZERO.
Work is force times distance; power is force times velocity; and you can’t talk about distance or velocity without defining a frame of reference. Until you understand that, you will not be able to figure out why DWFTTW is possible.

185. Stephen Wells

@Mender: Let’s say the velocity of the ground relative to the cart is -v (the ground is moving left relative to the cart if I sketch this in the cart-stationary frame) and the wind has a velocity u which could be positive, zero or negative. The air that travels through the rotor exits with a new velocity w. Optimum power from a turbine is obtained with w=u/3, from Betz’ law. Air has a density r and the rotor has area A. The air as it passes through the rotor has a velocity y which Betz’ analysis indicates can be taken as (u+w)/2.
u positive means the cart has a tailwind. The force of the wind on the rotor (direct “sail” force on the cart) is F_rotor = r A y (u-w) which at optimum is (4/9) r A u^2. The rotor extracts power P= (8/27) r A u^3 at optimum. This power can go to provide a motive force F_motive where conservation of energy means (F_motive v) cannot exceed P. The cart experiences a frictional force which depends on V and which we’ll call q(v). Here both F_rotor and F_motive act to accelerate the craft and, if q is small enough, can get the craft close to windspeed. However as we approach windspeed u tends to 0 and so both forces tend to zero; the craft must therefore steady-state at some positive u.
If u=0 then the turbine extracts no power, however much people want to protest, and if we extract some power from the wheels we must be exerting a force F_brake such that F_brake v = P, power extracted. The rotor can push on the air but cannot do more work than extracted from the wheel. Notice that if the rotor does push on the air, continuity requires that u become negative (we must be pulling air into the rotor). Which brings us to the third case.
If the craft is beating the wind then it’s moving into a headwind given by a negative u, say u = -k where k is positive. Again we have F_rotor = r A y (u-w) which at optimum is (4/9) r A k^2. The rotor extracts power P= (8/27) r A k^3 which limits our motive force(F_motive v).
Now, if the vehicle is to not slow down, the retarding forces F_rotor and friction q must balance against F_motive. The best you can do is near-zero friction so let’s set F_rotor against F_motive. In this case our F_motive, which is (8/27) r A k^2 (k/v), must equal our F_rotor drag force which is (4/9) r A k^2, leading to a requirement that k=3/2v, or in other words, the cart must be getting a headwind more than 3/2 of its ground speed. Sadly this contradicts the set up of the problem, in which k is the small amount by which the vehicle’s speed exceeds a wind speed which is (v-k) relative to the ground. Essentially the relatively large v kills the maximum F_motive you can exert without violating the energy-conservation requirement on the power.
Let’s not while we’re at it that there would be no contradiction in a vehicle using a turbine to power its motion into a headwind at less than windspeed, that’s just reversing the sign of v here, and we conclude that the best you could possibly do with a turbine-powered upwind beater would be to reach a v of 2/3 the wind speed. Earlier in the thread I believe there were comments that such vehicles have attained up to 0.6 windspeed in practice, so the result is clearly rational.
If anyone else wants to check my working then Betz’ law, there’s a clear derivation in the wikipedia entry if you want to check it, gives us a power extracted of (1/4) r A (u+w)(u^2 – w^2) and a force on the rotor of (1/2) r A (u+w)(u-w). My u and w are equivalent to the v2 and v1 of the wiki entry. I have not checked if the power/force requirements are better for us at other values of w, though I am investigating that at the moment.
So at the moment I can’t see any way the device can be doing what you think it’s doing, and we’ve yet to see a sufficiently well-controlled experimental demonstration, hence my scepticism. Substantive responses to the calcs above would be welcome.

“Not only was it authentic frontier gibberish, but it expressed a courage that is little seen in this day and age.”
— Olson Johnson (David Huddleson) Blazing Saddles

That whole analysis was pretty impressive, but unfortunately it was based on a misunderstanding of the device. The wind pushes the vehicle, which turns the wheels, which turns the propeller, creating thrust. You seem to have it the other way around.

187. Scott

Stephen, that’s a lot of math! Way more than I know. But math skills do not compensate for wrong assumptions.
You haven’t responded to my treadmill on a car. How do you feel about the two ice-boats and the chair on the telescoping spring?
I believe that the mistake you are making is that you are forgetting that on a vehicle which is at rest with the wind, some parts of that vehicle may still achieve movement relative to the wind and therefore gather energy from the wind.
I think it’s important for you to say yeah or neah on the two iceboats and the chair. To re-itterate it, an ice boat heading at a 45 degree angle to downwind can make 3x the wind speed. By tacking back and forth, it will easily beat a balloon in the wind to a given destination. If one were to attach two such boats via telescoping poles to a chair between the boats, one could ride in the chair DWFTTW. Can you, Stephen, find a hole in that logic?

188. SteveM

“What is going on in the treadmill video. There is no wind.”
Steve, imagine you are an ant standing on the treadmill. What do you feel? WIND.

You miss my point. I understand that in the belt’s reference frame there is a wind. But we are not talking about the belt’s reference frame but the cart’s. I agree that the belt moving at some velocity in still air is equivalent to the air moving at the same velocity over the ground. When the cart is on the treadmill in still air, it is equivalent to moving at the same speed as the wind. But disconnect the prop from the wheels and hold the cart in place on the treadmill, will the prop turn? No, the prop sees no wind to make it turn. Is there any blade angle or shape that will make it turn? of course not. The prop is being driven by the cart’s motion relative to the ground. It has to because its motion relative to the air is zero. It is only when the prop receives power from the wheels that it imparts a force against the air to move the cart forward.
Just to be clear I am no longer trying to argue whether this thing “works” or not. I am just trying to get all the forces and reactions clear. I am also focusing only on the treadmill experiment as that is clearly the most controlled. All the videos shot outside in the wind on the road leave too many variables uncontrolled.
Also, I see that my previous windtunnel proposal is not equivalent. What we need is a long windtunnel to completely eliminate the treadmill.

189. The Science Pundit

Stephen,
Please see the above posts about the pratfalls of using the cart as the reference frame.
From the cart reference frame it appears that there is no wind and that the ground is moving (much like the treadmill experiment), and so the cart takes its energy from the moving ground. To then claim that this does not constitute wind power is fallacious because the cat’s movement relative to the ground is caused by the wind and its motion relative to the ground.
Also as explained above, your calculations conclude that the cart can’t accelerate to FTWS, however the empirical evidence says otherwise.
You also seem to think that the wind is powering the turbine (prop) by passing through it. This is not the case! The wind is pushing on the prop, which pushes the cart along the ground, which turns the wheels, powering the prop, which pushes back against the wind. You need to adjust your calculations accordingly

190. ThinAirDesigns

Stephen @197:
>Substantive responses to the calcs above
>would be welcome.
The only substantive response that can be correctly given is that you are providing calcs for a premise different than on display by the cart.
All the good math in the world won’t help you if are applying it somewhere it doesn’t apply.
See #198
JB

191. Dave Munger

This is mostly cross-posted from my blog, but I think it’s important to get it down here too:
Stephen,
You’re neglecting the fact that the rotor is acting both as a sail and a propeller. In its “sail” mode, it pushes the cart along at wind speed. The larger the sail, the more power is applied to the wheels, and the more energy is available to turn the rotor. If the sail is twice as big as it needs to be to propel the cart forward at wind speed, the rest of the energy can be devoted to turning the rotor. This in turn propels the cart faster than wind speed. There will be some equilibrating, but the net result is a vehicle powered by the wind, moving faster than the wind.
…And now I need to get back to work on today’s Cognitive Daily post. See you in a few hours.

192. Scott

SteveM says:
“But disconnect the prop from the wheels and hold the cart in place on the treadmill, will the prop turn?”
Umm no, so don’t do that! The cart doesn’t work if it isn’t connected to two surfaces (the air and the ground) which are moving at different speeds.
When the cart is in zero wind and the treadmill is moving, we have simulated Down Wind At The Wind Speed. It is the same as a cart outside which felt wind when it started, and the accelerated to the wind speed. No can it go faster across the ground than the wind?
If the cart can make progress to the right on the treadmill, this is DWFTTW.

193. Mary

The wind pushes the vehicle, which turns the wheels, which turns the propeller, creating thrust. You seem to have it the other way around.
The wind pushes the vehicle, which turns the wheels, which turns the propeller, creating thrust… which turns the wheels, which turns the propeller, creating more thrust… which turns the wheels, which turns the propeller, creating more thrust… which turns the wheels, which turns the propeller, creating more thrust… which turns the wheels, which turns the propeller, creating more thrust
And zoom, off it goes, at the speed of light!

194. SteveM

You’re neglecting the fact that the rotor is acting both as a sail and a propeller.
Okay, how does this work exactly? How does it do both at the same time? Forces add, and only the net force can be transmitted to the cart. When the prop is in still air on the treadmill or moving at windspeed on the road, it can only be in “propeller mode”. The prop is spinning and pushing the air. The air is not pushing on the prop (other than in the 3rd Law sense).
When the cart is at rest in a tailwind, two things occur, the prop acts as a drag just by being an object in the wind, this will exert a net force onto the cart. We can measure that by disconnecting the prop from the wheels and locking it in place. Second, the prop acts as a sail perpendicular to the wind, making the prop spin. Unfortunately, if the prop is attached to the wheels , it would make it go backwards. So, in order to work, the force of the drag would have to be greater than the torque transmitted to the wheels. Which may be possible. So it may be possible to get it up to windspeed.
But once it is at windspeed, it is purely in propeller mode extracting energy from the vehicles forward motion over the ground. This may give it a momentary burst of speed but has to eventually slow it down below windspeed where it can once again act in sail mode to push it back up. So it would seem that this thing could oscillate around windspeed but an average only be as fast as the wind. And probably only at a particular wind speed since almost none of the forces involved are linear so there is probably a single optimum [yes, I know that is like saying a “single unique” value]

195. ThinAirDesigns

Mary @206:
>And zoom, off it goes, at the
>speed of light!
Perhaps yours, but ours obeys the laws of physics and stops well short of that.
There ends up being two limiting factors with our cart design:
A: Depending on the ‘advance ratio’ gearing (prop pitch to ground travel), there are limits to even the theoritical (dragless) speed of the cart in any single configuration.
B: Of course the real world isn’t ‘dragless’ and therefore the ultimate speed of the craft in any single advance ratio configuration ends up being slower than the theoretical.
However, as it turns out — both the theoretical and the real world upper limit speed of the craft are substantially above TWS.
JB

196. Scott

Mary:
“And zoom, off it goes, at the speed of light!”
Friction is the limiting factor. Probably it could do 1.5 or so times wind speed. Again, if you don’t like our explanations, the videos show that the device works and you are invited to create your own.
SteveM:
You haven’t responded to the two iceboat example.
If the ice boat example works, it should be possible to simplify the design, no? The spinning prop which is geared to the ground and acts like the two iceboats, even tracing a similar path through space. As such, it is always able to act like a sail and the force it gathers and transmits forward is greater than the force needed to spin the blade.
I don’t have the math to explain it beyond an intuitive level. If you see that the ice boats work, perhaps you can work from there to find a proof.

197. Dave Munger

But once it is at windspeed, it is purely in propeller mode
No, it is not. It is still providing the force equal to windspeed to the wheels from the tailwind. Then add to this the propulsion force from the prop’s rotation.

198. Subductionzone

Steve, if you watch enough of spork’s videos you will see at least one where the treadmill is at an incline and it maintains it speed for an extended period of time. No spurts in speed could explain this, it is equivalent to going downwind at the speed of the wind and using the energy of the wind to climb a hill. Again this would seem impossible, but as has been pointed out countless times, the source of energy is the difference between the speed of the wind and the ground.
Mary, do you deny that an iceboat can go faster than the wind in a downwind direction? In other words if you analyzed the velocity vector of the boat the velocity vector in the direction of the wind has been observed to outrace the wind by up to three times. Isn’t it “outrunning the wind”? I will answer, not really since it is still “in” the wind. So is the cart, it does not magically zoom ahead into a vacuum, it is still “within” the wind even when going faster than it.

199. The Science Pundit

Mary@#206

And zoom, off it goes, at the speed of light!

Please! Who here is claiming that there are no resistance forces such as friction and drag? The fact remains (and it is a fact as it has been repeatedly demonstrated experimentally) that until the cart reaches its equilibrium speed (which is FTTW), it is accelerating under the power of the wind. The mechanisms for this have also been repeated here in the comments ad nauseum, so please spare us the snarks.
SteveM@#207

But once it is at windspeed, it is purely in propeller mode extracting energy from the vehicles forward motion over the ground. This may give it a momentary burst of speed but has to eventually slow it down below windspeed where it can once again act in sail mode to push it back up. So it would seem that this thing could oscillate around windspeed but an average only be as fast as the wind.

Let me repeat my question from comment #105

Where is that energy stored? Why isn’t that energy depleted once the cart gets to its equilibrium speed with the wind? If the cart can accelerate faster than wind speed (which we have empirical evidence that it does) because of stored energy, then it should begin to deplete that energy as soon as it hits wind speed. If that were the case, then the cart would hit a maximum speed, then slow back down to wind speed. During this period it would twice pass through speed Vm (where Vm = (Vmax-Vwind)/2 for Vmax>Vwind), once where it’s accelerating because it has stored energy, and again as it’s slowing down because that energy is depleted. So how does it know which Vm it’s in? Where is this stored energy source?
Clearly the cart is not accelerating due to stored energy (unless you can identify the energy storage source that satisfies the above example of Vm).

Do you see the problem with your logic?

200. SteveM

Umm no, so don’t do that!
Would you please stop doing that. Yes, disconnecting stuff will make it “not work”. But it is a valid technique to understand what the forces and mechanisms are doing to the system.
You said, “It is not powered by the wheels ! … The wind provides power !
I said the prop sees no wind on the treadmill.
Your reply was “an ant standing on the treadmill will feel a wind”. That does not answer my point which I tried to illustrate with the “disconnect the prop” example.
We agree that the prop in still air on the treadmill is equivalent to the prop moving at windspeed on the road. How is it that we cannot agree that under either condition the prop is not extracting energy from the air but is instead pushing against it driven by the wheels and not the wind.
lets put it another way. The cart moving at windspeed, could just as well be a hot air balloon moving at windspeed. Hold up a propeller and nothing happens since you are not moving relative to the wind. But you are moving relative to the ground, so drop a wheel down to the ground geared to a propeller and you can make the propeller spin. No problem there, I think everyone can agree on this scenario. The problem is determining if you can use this sytem to make yourself go faster.

201. TFox

SteveM:

But once it is at windspeed, it is purely in propeller mode extracting energy from the vehicles forward motion over the ground.

This is half-right: at windspeed, working in a frame of reference comoving with the cart, the propeller acts as a propeller, providing thrust. However, the energy required to do this does not come from “the vehicles forward motion”. In that reference frame, the cart has no kinetic energy, so there’s none to extract. The ground does, however, and by putting roller on the ground (let’s call it a “wheel”) you can extract some, and provide the force necessary for further acceleration.

The point is, if you want to make energy arguments, you can’t mix reference frames. Energy is not invariant under Galilean transformations. “Where does the energy come from” is a question which has different answers in the ground frame and the air frame.

202. Anonymous

SteveM@#124

lets put it another way. The cart moving at windspeed, could just as well be a hot air balloon moving at windspeed. Hold up a propeller and nothing happens since you are not moving relative to the wind. But you are moving relative to the ground, so drop a wheel down to the ground geared to a propeller and you can make the propeller spin. No problem there, I think everyone can agree on this scenario. The problem is determining if you can use this sytem to make yourself go faster.

AAARRRGGGHHH!!!!
Of course the hot air balloon won’t work because as has been pointed out MANY TIMES before, the balloon is NOT in contact with the ground. The cart is contact with two surfaces moving at different speeds and is able to extract energy from them because it is a machine (a term I learned in middle school science class).

203. SteveM

Clearly the cart is not accelerating due to stored energy (unless you can identify the energy storage source that satisfies the above example of Vm).
Do you see the problem with your logic?

Yes, I do and left it unsaid so as to stay in the mode of just analyzing what we see in the treadmill video. The only source of stored energy is the kinetic energy of the vehicle itself. And that is what’s being extracted once the vehicle reaches windspeed. I think it is possible for it to momentarily accelerate as long as it “pays it back” by slowing below windspeed to get energy from the wind.

204. Subductionzone

When people talk about taking the cart apart to see if it works or where the energy comes from it is equivalent in many ways to saying take the sail or the keel off of a sailboat and see if it can still tack. The energy does not come from the wheels or the propeller alone, it comes from the system working together as a whole as does the sail and keel of a sailboat. It is non-intuitive but it isn’t rocket science, and if you cannot see that the treadmill is an excellent frame of reference equivalent to a tailwind at the speed of the treadmill it is time for you to retake your highschool physics class.

205. Scott

SteveM,
The cognitive trap here is the idea that “still air” has “no energy”. “Wind energy” is just a way of saying that the one surface is moving in one direction and another is moving in a different direction. If we can get our hooks into both surfaces we can make them work against each other and extract energy.
Do you know what a differential is?
http://en.wikipedia.org/wiki/Differential_(mechanical_device)
Executive summary; It’s a gearing system that takes as input rotation from two wheels and outputs the difference in motion to a third. So, apply 50MPH and 60MPH and it will output 10MPH rotation on the third wheel. It’s how your car can make turns without the two wheels skidding in spite of the fact that the inner wheel is tracing a shorter path than the outer.
Well, if you have two surfaces, one of which is moving at 10MPH and the other is moving at 0MPH, the diff will output 10MPH The same is true if the zero and 10mph surfaces are reversed, or if one surface is going at 20 million MPH and the other at 20 million MPH +10.
Finally, here is the video where the cart is tested without the prop, so you can see for yourself what effect that has.

206. Subductionzone

Oh Steve, c’mon now, storing energy?? Now you are violating Occam-s razor, not a real law but when you do it is time to think again. What is the mechanism for storing energy? How does the cart manage to climb the treadmill in a steady state when the treadmill is angled up? (and if you don’t think that you expend more energy running up a treadmill than on a level surface I think you haven’t done any treadmill training lately) It is alright to admit you do not understand the mechanics of it, but you have received enough proof that it is possible, and in fact has been accomplished.

207. Scott

I think this is pretty much won. The doubters rarely give a mea culpa but they do stop posting, which I think it is fair to interpret as them slinking away in defeat.
Are there any converted doubters present?

208. SteveM

When people talk about taking the cart apart to see if it works or where the energy comes from it is equivalent in many ways to saying take the sail or the keel off of a sailboat and see if it can still tack.

That is not what you do when you take things apart. You can indeed measure how a sailboat tacks by taking it apart and measuring the forces on each seperately. You can drag a keel through the water at different angles and measure the resulting forces. You can put the sail in a wind tunnel at different angles and measure the forces. ou can then add these results to get the system performance.

209. SteveM

The cognitive trap here is the idea that “still air” has “no energy”. “Wind energy” is just a way of saying that the one surface is moving in one direction and another is moving in a different direction. If we can get our hooks into both surfaces we can make them work against each other and extract energy.
Do you know what a differential is?

Now you are just being condescending and there is no reason for it. If you actually read what I’ve written you’d see that I understand what a differential is. You keep missing the point that when the vehicle is moving at windspeed, the only differential available is between the vehicle and the ground. That is all I’m saying, I don’t care how it got into that state whether it is moving over the ground in a breeze or standing on a treadmill in still air. The only differential is between the vehicle and the ground.
I am not trying to prove this thing is impossible. I am only trying to understand all the forces and energy available.

210. Subductionzone

Scot I was a doubter at the start of the year on the Mythbusters’ forum. I posted negative comments on Jack’s original youtube video also.
SteveM that is not what people are proposing to do when they take apart the cart. They say silly things like “take the propeller off and see if it runs off of the ramp”. If you know that it runs you might want to try to identify individual forces on different aspects of the cart, but there is no point in doing this unless you acknowledge that it works.

211. The Science Pundit

SteveM@#224

You keep missing the point that when the vehicle is moving at windspeed, the only differential available is between the vehicle and the ground. That is all I’m saying, I don’t care how it got into that state whether it is moving over the ground in a breeze or standing on a treadmill in still air. The only differential is between the vehicle and the ground.

No Steve, you’re missing the point. The differential is still between the air and the ground and the cart is acting as a machine to extract energy from that difference. It would be no different if the cart was stationary in a 10mph wind. The differential would be between the air and the ground. The cart is the machine that exploits the differential to begin moving.
Scott@#222

I think this is pretty much won. The doubters rarely give a mea culpa but they do stop posting, which I think it is fair to interpret as them slinking away in defeat.

I, for one, am expecting MarkCC to eventually post a Mea Culpa.

212. SteveM

Subductionzone@221:
Yes, kinetic energy is a form of stored energy, that is what regenerative braking does.
I am not trying to argue anything about whether that makes this device possible, or a trick, or impossible. Just stating that when the vehicle is at windspeed, the only source of energy available is the vehicle’s own kinetic energy which is the difference between its velocity and the ground. Whether it is moving or the ground is. This is a very simple mechanism it should be possible to analyze it.
But if you really want to take about stored energy, there are at least two springs in that little cart. I don’t think they are really contributing much, but they are there. The design of the wheels could store torsion, as well as that long propeller shaft. I don’t think they actually do store anything, but they could.

213. davem

Think of a Pitot tube. Hold it into wind, and what happens? The wind flows through the narrower section of the tube faster than the wind. Just using wind power, nothing else. So the ‘impossible’ is easily demonstrated as possible.
@Steve #207: “But once it is at windspeed, it is purely in propeller mode extracting energy from the vehicles forward motion over the ground.”
No, the propeller has created a low pressure zone in front of it. This will cause an apparent increase in the wind at the prop, driving the whole machine faster than the wind. You’re not just using the power of the wind immediately behind the cart to push it, you’re using the surrounding air, too. You’ve got the entire atmosphere to borrow energy from.

214. Scott

SteveM,
“when the vehicle is at windspeed the only source of energy available is the vehicle’s own kinetic energy which is the difference between its velocity and the ground.”
Steve, this is false. The cart has access to the “still” air and it has access to the moving ground. To get a grip on both and continue to extract energy is an engineering problem which this device solves.

215. JR

davem @ #228: That is just pure genius. Nothing could be simpler or more certainly answer all of the arguments from ‘impossibility’. Thank you.

216. SteveM

@Steve #207: “But once it is at windspeed, it is purely in propeller mode extracting energy from the vehicles forward motion over the ground.”
No, the propeller has created a low pressure zone in front of it. This will cause an apparent increase in the wind at the prop, driving the whole machine faster than the wind.

Yes, that is what “propeller mode” is. It takes air from in front of it and pushes it behind creating a pressure differential that drives the propeller forward.
My point is that to do that it needs to draw power from somewhere, it can’t get it from the air that it is acting on, it has to get it from its motion over the ground through the wheels.

217. Scott

“My point is that to do that it needs to draw power from somewhere, it can’t get it from the air that it is acting on, it has to get it from its motion over the ground through the wheels.”
SteveM, the device never got power “from the air”. It gets it from the air/ground speed difference. That difference remains constant regardless of the speed of the vehicle.
A “wheel” is being applied to both the air and the ground, and power is being extracted from the difference.

218. SteveM

A “wheel” is being applied to both the air and the ground, and power is being extracted from the difference.
Okay, I can see that. So, lets extend the “wheel to both” metaphor to actual wheels. Replace the prop with a wheel to the ground with the vehicle on the treadmill. Should be able to make that go faster than the treadmill also.

219. Mary

I think this is pretty much won. The doubters rarely give a mea culpa but they do stop posting, which I think it is fair to interpret as them slinking away in defeat.
I’m gonna stop posting now, ’cause I have other problems to think about. (What *is* the dispersion at the lasing frequency in an inhomgenously broadened laser gain medium, and how does it depend on the unsaturated lineshape of that gain?)
But please don’t think it’s because I’m not enjoying the argument, or even that I have run out of ways to say “If the wheels are driving the prop, then the prop’s not driving the wheels.” I love a good physics argument, and could read and reply to this all day, if I didn’t feel guilty about the time wasted.
Are there any converted doubters present?
I converted the other way. From thinking it plausible, before I watched the video and saw what the cart was actually supposed to be, to thinking Mark had it right all along.

220. Scott

Yes, I’m sure that that would work. Yeah!
Does this mean it is time for Mark to appear and say that we are not bozos?

221. Anonymous

Steve @224:
>The only differential is between the
>vehicle and the ground.
Scott @232:
>SteveM, the device never got power “from
>the air”. It gets it from the air/ground
>speed difference. That difference remains
>constant regardless of the speed of the
>vehicle.
There literally are no two statements made on this entire thread that better sum up both the misconception and the way to clarity.
Steve claims the the “only differential it between the vehicle and the ground”, completely ignoring that the differential that really matters (the one and only power producing differential) doesn’t change with the speed of the vehicle.
Even with the vehicle at TWS, the ground/air differential remains and as long as we can keep a proper hook into both mediums, the power source remains available.
This has been explained and demonstrated so many times now that it is no longer a physics experiment, but rather a sociological one.
JB

222. The Science Pundit

Okay, I can see that. So, lets extend the “wheel to both” metaphor to actual wheels. Replace the prop with a wheel to the ground with the vehicle on the treadmill. Should be able to make that go faster than the treadmill also.

Yes! I can’t draw it in the comments, but imagine that through a shaft or gearbox, the wheel on the treadmill is powering the wheel on the ground, which is pushing the wheel on the treadmill to the right. Are we finally winning you over?

223. cm

Mary, I’ve very much enjoyed and agree with your comments, thanks for playing. I will waste some time in your honor.

224. TFox

Mary:
Thanks for dropping by. When you have time, perhaps after a few Friday afternoon post-research beers, come back and have another look. It’s a great problem, and I suspect it’ll be appearing on a fair number of classical mechanics exams in the near future. And I’d also be willing to bet that a fair number of the answer keys will get it wrong! It’s tricky.
Things to review, before you reach a conclusion:
o the spool and thread problem (yo-yo), where the spool travels faster than the thread pulling it
o iceboats which regularly travel downwind faster than wind by tacking downwind
o the nonequivalence of the direction of energy flow in different inertial frames, which underlies a lot of the confusion about whether the wheels drive the prop, or vice versa

225. cm

Could Scott, JB, davem, Subductionzone, or The Science Pundit–but especially Scott–please actually explain *how* this thing “extracts power” from the “ground/air interface”? If it has been already explained above, just refernce a comment #.
A proper explanation will contain the following words: “EXERTS A FORCE ON” at each point in the story when something causes something else to do work. No “extraction” or “exploitation”, please. Use “exerts a force on”.
I think you are all just saying that phrase or some variant of it and have no sense of the energy bookkeeping here. This is handwaving.

226. Subductionzone

I am rather disappointed that Mary is leaving, it sounds like she should have enough physics to understand an inertial frame of reference equivalence. Too bad she did not check out some more of sprok’s videos of his improved cart that was more efficient, that little sucker kept trying to outrun the treadmill, maybe some of the doubters here could give the cart a physics lessen. Just because you do not understand how something works does not make it impossible. If it keeps working in defiance to your understanding of the laws of physics maybe your understanding is not quite up to the task.

227. Mark C. Chu-Carroll

Don’t assume that people are “slinking away” rather than admit that they’re wrong.
(A) I have an actual job that I need to do; I can’t spend all day on my blog. I’ve already
spent more time on this than I really should.
(B) I have gotten incredibly bored with saying the same thing over and over again. I’m
still convinced that the energy equation here just doesn’t work, and I’ve explained why
over and over; I can’t think of another way to say it.

228. ThinAirDesigns

MarkC @242:
>I can’t think of another way to say it.
I sure hope you can think of a way to apologize to the “bozos” when this is all over. I’m betting a third grader is better at that as well.
JB

229. Subductionzone

cm, I cannot explain the mechanics of it, I am not an aeronautical engineer or a sailor, but I am starting to pick up an intuitive sense of how it works. All I can do is handwave at this point. Most of my handwaving goes towards the videos and says “hey look it works, don’t ask me for details ask them”. I can see how the difference in speed between the ground and the air can be a source of energy. If you think that the treadmill is not a perfect representation of a tailwind at the speed of the treadmill I would like to see some evidence of that, all I can say right now is that if you claim that you are wrong. One thing I do understand is frame of reference. Getting back to the videos if they are a hoax, how did they accomplish this? You can see in their videos the carts accelerating down the tread, working its way uphill etc.. I have not seen spork or thinair to be wrong in any of their sailing claims, but I have seen others, I have not seen anyone take back their claim that a sailboat cannot sail downwind faster than the wind, just not DIRECTLY downwind. Once you realize that ice boats can go several times the speed of the wind in the downwind direction on a broad reach the accomplishments of this little cart do not seem so impossible. It is just wildly counterintuitive to most people, me included.

230. Dave Munger

Based on his comments on Twitter and here, it sounds like Mark still isn’t convinced that the treadmill demo is equivalent of real-world faster-than-wind travel. Anyone have any ideas on how to convince him?

231. KJC

Ok so here is why the frame of reference is equivalent. If you wake one day to find yourself sitting on a cart, in still air while moving along at a constant speed of 10mph, you have no way of knowing whether the road is being driven realtive to the air or if a 10mph wind is blowing. You could be on a huge treadmill or you could be in the wind. Does it seem like there is no available source of energy? Your cart is moving relative to the road so you can certainly put down a wheel and run a little gerator. Now when you do this you intersetingly find that you do not slow down. This is because the wind will always blow at 10 mph (or the road will always be driven at 10mph. Ok so now what can we do with this extra energy that we have extracted from the road well sailing down wind can get a little stuffy so I might power a fan just to get a little breeze going, and don’t you know I just happen to face it such that it creates some thrust. But you say if this is the case I can increase my speed to inifity and that is not possible because you have created a perpetaul motion machine. In this case however if we say that the air and road will *always* move at 10mph that is an infite energy source. In real life there is plenty enough energy in the wind to drive the cart a little faster. The model of the wind always blowing at 10mph is like finding some device that moves quite slowly but always at a constant speed regardless of the load. That is an infite energy source (and hence does not exist)

232. SteveM

Are we finally winning you over?
Not a question of “winning me over” yet. As I’ve said several times, I am not trying to prove or disprove this things impossibity, I’m just trying to account for what’s going on and understand just what is happening in the treadmill video. Scott’s metaphor of two wheels in two mediums at different velocities was a good one for my understanding. It should make it easier to model mathematically if aerodynamics can be ignored and focus on just the differential.
So imagine two wheels connected by a drive shaft.
[I’m drawing this as surface one being a ceiling moving to the right, surface zero being the ground also moving to the right. radial velocity is positive in the clockwise direction) ]
wheel 1: radius R1 is on surface 1
surface 1 is moving at a velocity v1
wheel 2: radius R2 is on surface 0
surface 0 is moving at velocity v0
the assembly is moving at a velocity vc.
the radial velocity of wheel 1 is w1 = R1*(v1-vc)
the radial velocity of wheel 0 is w0 = R0*(vc-v0)
since they are connected w1 = w0
so R1*(v1-vc) = R0*(vc-v0)
vc = ((R0/R1)v1 + v0)/(1+(R0/R1))
let v0 = 0 to be the ground and let R0/R1=k
vc = v1*(k/(1+k))
well that didn’t work, what’s my mistake?

233. Mary

Wait – one more. đŸ™‚
What if you had two treadmill belts, one underneath you going one direction, and one overhead going the opposite direction… Now you build a little cart, and instead of having a propeller, it has a little wheel, connected to a bicycle chain, on a pole that you can raise until it’s in contact with the upper belt. The bicycle chain causes the lower wheels to turn at the same angular rate as the upper wheel (and for simplicity, let us assume they are the same size.) How fast can it go? (And in what direction?)
That’s the question I would put on an analytical mechanics exam I was making up.

234. Subductionzone

I think Mark is hopeless, he seems to be of the “I don’t understand it so it must be impossible” school. Even when he sees something that is irrefutable (the equivalence of reference frames) he tries to deny it. As I said outdoor test would not work to convince these people either since they would claim something on the order of improperly measured windspeeds etc.. I learned by being shown that yes sailing vehicles can run down the wind faster than the wind, just not directly down the wind. Then I was taught how a boat on a river with no wind but a current can outsail the same boat on the same river with a tailwind that matches the speed of the river. That made me a bit more open minded. Then when I saw the treadmill test I was convinced. As I said equivalence of inertial frames is one thing I can “get”.

235. Mark C. Chu-Carroll

Re #243:
If someone manages to convince me that I’m wrong, I’ll publicly apologize. I’ve done it before, and I’m sure I’ll do it again.
But I want *math*. I want to know what the *forces* are in this thing. As I see it, the original claim is that using the force of the wind, you can make this cart go directly downwind faster than the wind that’s pushing it. I want to see how you’re really generating a force from the wind. For now, to me, it seems completely circular: the wind pushes the propellor; the propeller pushes the wheels; the wheels push the propeller, and the propellor pushes the vehicle. The claim seems to be that the propellor is basically both pulling energy from the wheels and pushing energy into the wheels, at the same time. That can’t work; that’s perpetual motion.

236. SteveM

my mistake:
I let w1=w0, suppose we insert an idle gear to reverse it so now :
w1 = -w0
then vc = (k/(k-1))v1
and so vc can be faster than v1.
hmmm…

237. KJC

But Mark can we extract extra energy via the wheels to say light a bulb? That is 2 otherwise idetical carts but one with one wheel attached to a gernator that lights a bulb. At least proving there is more energy in the wind than just to push it along (or does the light go out once we are not accelerating?) or can we do this but just can’t use that energy to provide thrust?

238. cm

Mark, forces-schmorces! Come on, haven’t you been paying attention? The propeller isn’t pulling energy from the wheels, but the GROUND/AIR INTERFAIT–er, INTERFACE. If you just believe in reference frames and the Great Ground/Air Interface, if you just get nice and intuitive and transcend this discussion of forces acting to do work, and wave your hands over them like the Peanuts gang over Charlie Brown’s Christmas tree, this thing goes like a bat out of hell.

239. Paul

Just a question here. If you hold the cart (the one from the treadmill video) in your hand and spin the propeller in a full circle, how many revolutions do the wheels make?

240. Mender

Mark, the cart is a gearing system between the air and the ground. The propeller is the connection to the air. The wheels are the connection to the ground.
Go find a planetary gearset. Hold the inner gear with one hand and turn the planet carrier with the other hand. The outer gear spins faster than the other two.
Which one of your hands is supplying the power? What happens when you change which hand moves?

241. ratwerks

Mark at #250
Do you believe that any sailcraft can beat the wind (where by ‘beat the wind’ I mean have the ability to reach a point downwind faster than, say, a baloon?) That they can on land, ice and water is well established and easy to confirm.
If you believe that some device (eg a sailcraft) can beat the wind must you not now abandon your impossibility argument?
Once you accept that -something- can beat the wind it should become a question of -how- this thing can do it rather than -whether- this thing can do it.
View the spork-cart from above and push it along some convenient surface. Note or trace the path followed by the propeller tips as they trace their double helix across your surface. Perhaps draw those paths as lines on the surface. You will see that each propeller tip veers back and forth across the straight line direction of travel making a nice 2 dimensional representation of said double helix. This sine wave is a function only of the gearing between the wheels and the propeller; not of the speed of the cart.

242. The Science Pundit

SteveM@#247
For simplicity, let’s assume that the drive/gearbox mechanism connecting the two wheels is infinitely rigid (this is for ease of explanation; the same principle applies if it’s not), therefore v1=v2=vc. Next, we need to arbitrarily choose an initial velocity vi for vc. For this example, I will choose vi=v0. In other words, the contraption begins at rest with respect to the wall and at -10mph wrt the treadmill. This is equivalent to spork placing the cart on the treadmill then letting go when the wheels were up to speed. In principle, the same would apply if we chose vi=vt=-10mph (vt=velocity of treadmill belt), the only difference would be that we would need a treadmill long enough that the contraption would get up to v0 before running out of belt (like one of those treadmills connecting airport terminals).
Therefore the treadmill will accelerate wheel#1 to spin clockwise at a rotational velocity of vt*r1, but is not imparting any momentum to the axle of the wheel (Are you following?). The drive shaft will cause wheel#2 to rotate counterclockwise against the ceiling with a rotational velocity equal to wheel#1 w2=vt*r1. This will cause wheel#2 to roll along the wall to the right with a velocity v2=w2/r2=vt*r1/r2. Since the shaft is rigid (v2=vc=v1), it will move wheel#1 to the right at a velocity of v2.
Naturally, the new rotational velocity of wheel#1 will be w1=(vt+v2)/r1. It will continue to accelerate until the drag and friction forces equal the forward force.
The above is a simplified (I left A LOT out) outline of how such a contraption might work.

243. TFox

Mark:

The claim seems to be that the propellor is basically both pulling energy from the wheels and pushing energy into the wheels, at the same time.

I don’t think that anyone who understands the problem would claim that. You cannot specify a direction of energy flow without first specifying which frame you are in. In the spool problem, where does the energy come from, the thread or the ground?

That can’t work; that’s perpetual motion.

Perpetual motion usually refers to first law violations. There’s an energy source here, ergo no violation.

244. StuV

ratwerks: Until such time as sailcraft can do so STRAIGHT INTO the wind, the point is completely irrelevant. Thank you for playing though.

245. SteveM

therefore v1=v2=vc.
No, that would represent the air moving at the same speed as the ground.
The “ceiling” is moving at one velocity (v1), the floor is moving at a different velocity (vc), the cart is inbetween moving at its own velocity (vc). Is there a way to connect v1 and v0 to vc such that vc>max(v0,v1)?
I think I showed that there is.

246. Mark C. Chu-Carroll

Re: Mender: I didn’t find the math in 184 particularly compelling.
On the other hand, Steve’s stuff in #251 is interesting, and it’s
got me wondering. I’m still not seeing how the force from the wind is getting transferred via the propeller – but the math in #251 seems to suggest that it’s possible. After work, when I’ve got some time, I’m going to try to sit down with Steve’s gearing based approach, check that he’s really doing it right, and if he is, see how I can translate that to the propeller.
If that works, then I’ll admit that I’m wrong, and put up a new post with Steve’s equations, my illustrations, and my apologies.

247. Scott

Mark,
You are wrong. And your insistence that the we are not explaining it right is also wrong. There are many good point that have been made against your arguments to which you have not replied.
It is not correct for you to where the mantle of Scientist and “squasher of fools” if you don’t have the time to interact with those you would label as “bozos”, etc.
Here’s one of the many excellent arguments to which you have not deigned a response; Does the two-iceboats-tacking-with-a-chair in the middle represent DWFTTW? If not why not? If so, can the design be simplified?
Remember that you started this not by making the assertion that a DWFTTW device is unlikely or not yet achieved. No, your assertion was that this device does not work, no such device is feasible, and believing otherwise was the province of flat-earthers and over-unity nut jobs. The first assertion is invalidated by expiremental evidence and the latter by the reasoned discussion which has continued for several days.
Eating your words might be easier if they were not so bitter.

248. Mary

[The solution to my exam problem: Let’s work in the frame of the lower treadmill. The upper treadmill has a relative velocity “vu”. The velocity of the cart, when neither the upper nor the lower wheels are slipping, is “vc”. The angular velocity of the upper wheel is “w”, and thanks to the bicycle chain, that is also the angular velocity of the lower wheels. Because the upper wheel is not slipping, w=(vu-vc)/ru (where “ru” is the radius of the upper wheel.) Because the lower wheels are not slipping, we also have w=vc/rl (where “rl” is the radius of the lower wheels. We don’t need to assume them equal, in spite of what I said above, it’s more fun if we don’t.) Set those two expressions for w equal and solve for vc. You get vc=((rl/ru)*vu)/(1+rl/ru) which means for vc > vu, you need rl/ru > (1+rl/ru), which is impossible. Therefor vc is always less than vu.]

249. hibob

Mark: I came up with a thought model that might convince you. If you can accept that ice boats can tack to go downwind 3x faster than the wind, I think you’re there.
Imagine your ice boat traveling at a 45 degree angle to the wind; the component of its velocity parallel to the wind is 3x Vwind. This works fine on a flat lake – or really any surface. Dirt boats do almost as well on salt flats using wheels instead of blades. So imagine a big cylinder with a wind moving down it’s length. Three little dirt boats are corkscrewing their way around the cylinder, equidistant from each other and held together (and their wheels held to the surface of the cylinder) by a ring a tad bit bigger than the cylinder. They’re moving downwind 3x faster than the wind. Nothing has really changed from racing on a salt flat.
So: you’ve just created a propellor, that through mechanical coupling to a surface generates enough lift to move faster downwind than the wind driving it. It’s the same situation as the telescoping rods between two boats, but even more relevant. Hang a gondola off the ring via some bearings (so the passengers aren’t corkscrewing as well) and you’re done.
If the iceboats can do it, so can this system.
Right?
From there, it’s easy to switch the sails for a prop, and the mechanical coupling to that of the cart on the ground.

250. SteveM

Mary@265
That is exactly what I did in my first attempt. but twist the “bicycle chain” into an “8” so that wheel 1 is driven in the opposite direction of wheel 2.

251. ratwerks

StuV at #261:
I’m sorry you see it that way, but you are mistaken. The argument from impossibility claims that downwind faster than the wind violates some natural law. The existence of sailcraft doing it demonstrates that it does not.
Now, if you would like to understand how the spork-cart propeller is -exactly like- two sailcraft on a 45 degree downwind reach, you need only read this thread with that question in mind rather than “how can i prove that this cannot be done” in mind.
Also, I suspect you mean “STRAIGHT DOWN wind” rather than “STRAIGHT INTO the wind.” Especially given that, while they are equivalent from an energy point of view, the later is easily understood and noncontroversial while the former is what is hotly debated here.

252. The Science Pundit

SteveM@#262
Sorry about that, I was using v1 as velocity of wheel#1 and v2 as velocity of wheel#2.
I was going to redo the calculation, but then I saw your post #251. đŸ™‚

253. StuV

Ah, I see. I was never debating that “downwind faster than the wind is impossible”. I was simply outraged at the notion that anyone would consider that perpetual motion.

254. PeteC

Kudos to cm@240 for the very reasonable request; I would really like to see this. A bullet-point explanation or even a PDF diagram of what is exerting force on what would really help.
I also don’t see why a consideration of two connected ice/sailboats as a single system hasn’t closed the argument yet – except for the consideration that tacking also requires energy, which would have to come entirely from onboard generators; but I guess that could be easily handled.

255. Scott

“tacking also requires energy, which would have to come entirely from onboard generators”.
PeteC, “tacking” means turning the boat to point it at an angle to the wind. It does not require power other than that provided by the wind. Christopher Columbus, etc..?

256. SC

How fun this thread is. I had to stop around #100 and skip ahead to the end for now, though I hope to have time to return to read the whole thing later. Thank you all for a lively and fascinating discussion which, amazingly, I’ve been able to follow.
*wonders if MIT has winter-session physics classes she can sneak into*
*regrets openly mentioning that*
*returns to lurking*

257. SteveM

@272:
“tacking” is also a maneuver: changing the direction of the boat from one side of a headwind to the other side. Involves a “dead zone” when your are pointed directly into the wind and the sails “luff” (hang limp). You need to have enough momentum to get you through the luffing until the sails can fill again on the other side. Many a novice sailor gets stuck pointed at the wind this way.

258. SteveM

SC,
I don’t know if MIT still does this but they used to have IAP (independant Activities Period) during January that offered a lot of mini-courses and seminars on more “fun” kinds of subjects.

259. Mark C. Chu-Carroll

Re #264:
I’m not wearing any “mantle”, and I’m not bitter; I’m just hard-headed.
But one thing I really want to stress here. I owe you nothing.
Writing this blog is my hobby, not my career. I have a family, and a job, both of which have higher priorities than this blog. I don’t owe you prompt answers to your comments. I don’t owe you answers to your comments at all.
This blog is a hobby of mine. I don’t force anyone to read it. If you don’t think I reply to your comments promptly enough, then don’t comment here. If you don’t like my attitude, don’t read my blog.
If you think that the act of your reading my blog puts me under an obligation to drop my work every time you post a comment saying I made a mistake, then you’ve got another think coming.

260. spork

“I owe you nothing”
Wrong – you owe me an apology – but I’m not holding my breath.
Why…
(and not terribly bright)

261. Mark C. Chu-Carroll

For those who’ve been accusing me of refusing to listen to reason, let me explain where I’m coming from.
I’ve been wanting a description of the math. I’m not doing that because I’m some bitter stubborn bastard, but because real physical systems that involve the play of forces are complex, and it’s very easy to make mistakes in informal language. Just wander around the net to any of the various perpetual motion schemes. Read an explanation of how burning Brown’s gas can produce more energy burning to water than it takes to spilt the water. Read one of the many explanations for supposed magnetic motors operating above the zero point.
They’re very convincing. And they’ve got explanations, and videos, and
demonstrations. They’ve got metaphors to show you why it’s not really perpetual motion, or why it still works even if it is perpetual motion. They come back with increasingly
angry responses to skeptics, just like we can see here in this thread. But the thing is,
when you do the math, it doesn’t work. And then the wonderful examples, the metaphors, the illustrations, the demonstrations, the videos, all of it falls apart – because the fundamental equation for the forces simply doesn’t work. Demonstrations,
videos, metaphors, information explanations are all wonderful – but the real proof is in the math.
The basic mantra of this blog has always been “The worst math is no math”: when you’re trying to prove something that’s really mathematical, trying to do it with informal words is a losing proposition. Without the rigorous formality of math, it’s just too easy to
screw up.
Before writing this post, I did attempt to do the math. I might have blown it, if the recently posted equations prove out. Steve is the first one that’s actually produced a nice, simple mathematical demonstration of why I might be wrong – and almost immediately on seeing it, I’ve started to admit that I might have blown it. Once I find the time to sit down with a nice pad of paper and see if I can verify his derivation, and prove to myself that it’s a valid equivalence for the wind-vehicle, I’ll happily eat my words, and put up a new post explaining how and why I’m wrong, and apologize for calling them bozos.
Steve’s math is very nice – and if it’s correct, it shows how the two mediums works. I still haven’t had time to verify it. I’m going to try to, when I find time – which will probably be sometime over the weekend. What I’m going to look for is two things:
(1) Is his equation a correct model of the geared, double-wheeled situation that he describes?
(2) Is the two-wheeled situation really equivalent to the wheels and propeller?
If so, I’ll post a cleaned up version of my drawings, along with the fleshed out details of my analysis of Steve’s proof. If not, I’ll post an explanation of what’s wrong with those equations.
In the meantime, I have no interest in repeating what I’ve already said dozens of times, nor do I particularly care to listen to anyone else repeating their arguments. If you’ve got something new to say, go right ahead. If you’re just repeating yourself or someone else, do me a favor, and don’t waste your time.

262. Dave Munger

Spork:
Actually he’s very bright. He’s just a busy man who may not have thought enough about the problem before making a dramatic declaration on his blog. I’m confident that if and when he has a chance to work out the math he’ll be able to offer an impressive analysis of the problem.

263. Anonymous

Paul @254:
>Just a question here. If you hold the cart
>(the one from the treadmill video) in your
>hand and spin the propeller in a full circle,
>how many revolutions do the wheels make?
Paul, in our videos we show two different cart designs on the treadmill.
The large video cart with the 48″ prop has a 3:1 gear ratio from wheel to prop.
The small video cart with the 15″ prop has a 1:1 ratio between wheel axle and prop axle.
The gearbox in our posted build list is 16:13 in favor of the wheels turning faster.
In the end, the actual gear ratio is not that imporant as long as you can get the proper advance ratio between the propeller pitch and ground travel. Changing gear ratios, prop pitch and wheel size all work to change the advance ratio.
We have found the best advance ratio results in the 1.2 to 1.4 range — meaning that the wheels travel farther than per prop rotation by that multiple over the pitch of the prop per rotation. On the lower side of that ratio, the cart will beat the wind at a lower speed (our best carts beats the wind when it’s only blowing 2.7mph). On the higher side of the ratio, the cart will beat the wind by a higher margin.
Yes, as odd as it seems the travel of the prop through it’s medium is less than the travel of the wheels through its medium. Some say “the how does the prop generate any thrust relative to the wheels?”. Remember the tailwind — we have to cover a lot of ground to see just a little air.
JB

264. Mark C. Chu-Carroll

Re #277:
Unless you’re one of the guys behind the original demo, then I don’t even owe you an apology. The only people that I’ve insulted are them. The rest has been a debate, and I don’t owe you an apology for not being convinced by your arguments, any more than I think you owe me an apology for not being convinced by mine.
If you’re one of the guys behind the demo, and the math works out, then I’ve already said that I’ll make a new, top-level post on my blog containing an apology, and an explanation of where I went wrong and how it really works.
Beyond that, Like I said, if you don’t like it, don’t bother reading the blog.

265. ThinAirDesigns

Mark, Spork and I are the originator of the demo.
That’s my ugly mug everyone sees every time paused at the top of this blog and I will apologize right now for whatever YouTube algorithm causes that discomfort to all.
JB

266. Mark C. Chu-Carroll

Re #282:
Ok then, if the math proves out, then you’ll get a nice prominent apology.
If it doesn’t, then I’ll keep disagreeing with you.

267. Scott

I just realized how the video is faked! The house it is filmed in is being tilted! OMG I was so wrong.
Also, there is no Math, so it doesn’t count! You saw the sun rise today? Don’t tell Mark about it if you can’t describe the rate of ascent via and change in the hue and brightness with Math. He’s busy. If you are lucky, maybe he’ll find the time during lunch to call you a bozo.

268. Subductionzone

Math aside, what possible objection could you have to the treadmill test? It is as I have said before a perfect analogy to a tailwind. By the way, I am not one of these “bozos”, so no apology to me expected or asked for, but I did disagree strongly with spork in the past. When I saw the treadmill test I knew that I was wrong.

269. Scott

Mark @281
Spork is indeed the handle of one of the guys you called a bozo.
“Unless you’re one of the guys behind the original demo, then I don’t even owe you an apology. The only people that I’ve insulted are them.”
Hmm, reading your post again, I see that you pretty much reserved the term Bozo for the inventors of the device. But you did say this:
“I’m incredibly disappointed in how utterly clueless about things like this most people are”
Oh, we’re not bozos. We’re utterly clueless. That’s completely different.

270. ratwerks

Mark @ #278
You had the time to write a couple of pages of insulting screed that these ‘bozos’ are idiots or cheaters or both. You had the time to explain why your inability to understand has to be addressed by others. You had the time to opine that you don’t owe nobody nuthin.
By most rules of polite discourse when you call somebody a bozo in public you owe them a public apology when the bozo turns out to be you. You owe it to yourself to figure out who the bozo here is, and plenty of people have volunteered their time here to help you do it. How sad for you if you “don’t have time” to clean up the mess you made (and could have avoided making if you’d done a bit more research before writing and publishing your original scathing ‘analysis’.) What price reputation?

271. Nerull

You know, there is one simple way to solve this.
Stop wanking off on a treadmill, take the damn thing outside and give it some wind.
If it works, it works. If it doesn’t, it doesn’t. But if you are afraid to actually test it in the real world, than there is no reason for anyone to listen to your cries of “It wooooorks!!!! WHY WON’T ANYONE BELIEVE US!!!”

272. Stephen Wells

Can the inventors of the device please say whether the prop is being driven by the wind and turning the wheels, or whether it is being driven by the wheels and is acting as a propellor. “Both” is not an answer. “Driven by the wheels” is not an answer as that is braking and you can’t get more propulsion than you get from braking. “Driven by the wind” is analysed in post 197 and has received no substantive response. In 197 I show that a turbine vehicle can progress into a head wind at a fraction of wind speed (actual fraction slightly different from my answer due to a small algebra error) but the slightly-outrunning-the-wind thing is not workable. “I don’t know” is an acceptable answer but won’t convince anyone. I went to all the trouble of actually doing the math and not even Mark noticed. I pout.
Incidentally I modelled the idea of DWFTTW by tacking in an iceboat; this seems workable at first glance, the captain’s chair device would work in principle but is unnecessary. Mark was wrong to argue that the whole concept of DWFTTW is wrong but right, as far as I can tell, that this prop thing running at wind speed is definitely not doing what people are claiming it does.

273. rpsms

Mark has a good reputation because he does apologize. He also makes corrections, leaving the original erroneous texts visible for the world to see.
Regarding #287, when you see someone call “most people” anything, perhaps you should exclude yourself from your consideration. You might be happier, and nobody would be the wiser for it.

274. TFox

Mary (re #265)
The math here is algebraically perfect, for people who like equations. It’s also almost entirely general. However, because we can flip signs by twisted belt, or a spool with an axis of rotation above both belts, we need to allow for negative radii. The conclusion about allowed relative velocities implicitly assumes all radii have to be positive; this is incorrect. See

for an example. Allowing for arbitrary gearing, your result gives an arbitrary ratio of velocities.

275. spork

“For those who’ve been accusing me of refusing to listen to reason, let me explain where I’m coming from.I’ve been wanting a description of the math”
Two problems with that theory… First you didn’t ask for any description of the math before posting my video in your blog and calling me names. Second, I’ve posted solutions to the math all over the internet. You could very easily have contacted me through youtube as many others have, but it was easier to call me an idiot.
I find it a little ironic that a computer scientist is calling me an idiot when A) he’s completely wrong, B) this is my field, and C) he never made the slightest attempt to understand what he’s looking at.
“Steve is the first one that’s actually produced a nice, simple mathematical demonstration of why I might be wrong”
Nonsense! He may be the first to post it on YOUR blog after you already called me an idiot. But I’ve posted so many explanations of this thing it would make your head spin.
“Spork: Actually he’s very bright. … if and when he has a chance to work out the math he’ll be able to offer an impressive analysis of the problem.”
I hope you’ll understand if I don’t take your word for it. So far he’s proven himself to be wrong, stubborn, and reactionary. I doubt very much that he’s able to understand the subtleties of this simple toy (which he claims his third grade daughter understands).
“Ok then, if the math proves out, then you’ll get a nice prominent apology. If it doesn’t, then I’ll keep disagreeing with you.”
Translation: if you’re able to understand the math you’ll offer a backhanded apology. If not you’ll continue to insult us. This ain’t my first barbeque. I know the drill.

276. TFox

Can the inventors of the device please say whether the prop is being driven by the wind and turning the wheels, or whether it is being driven by the wheels and is acting as a propellor.

It was supposedly invented in the 50’s by someone named Andrew Bauer. If he’s still alive, he’s not apparently participating in this thread. I’ll answer instead of him, if I may: your question is meaningless, because you have not specified what reference frame you wish the answer to be in. See #168 and #171 in a previous thread, http://www.boingboing.net/2008/12/03/directly-downwind-fa.html#comment-346737.

277. Stephen Wells

@297: the question is not meaningless and the answer is the same in any reference frame. Is there a braking force on the wheels and a propulsive force on the prop, or vice versa? Answer does not depend on frame.

278. ratwerks

“Everyone should be able to understand the physics involved here. My third grade daughter can understand this. This isn’t difficult. There’s nothing tricky or subtle about it. If you have a vehicle moving at the same velocity as the wind, the wind cannot possibly exert any force on the vehicle. No force, no acceleration. Period. How can supposedly intelligent, educated people not know this?”
— MarkCC
Mark, how do your reconcile this quotation with your subsequent demands for some sort of mathematical proof?
By the way, my own third grader understands this and could explain to you where your assumptions are flawed. He’s a whiz at math, but being a third grader his math skills are not up to what a serious mathematical description of this system requires. Fortunately for you, a general description is really quite enough. Feel free to drop by this weekend and he can sketch it out for you.

279. StuV

Translation: if you’re able to understand the math you’ll offer a backhanded apology. If not you’ll continue to insult us. This ain’t my first barbeque. I know the drill.
Translation: waaah.

280. David Grogono

If one replaces the propellor shaft and propellor with a threaded rod and the wind with a nut you end up with something like I have diagramed here:
http://www.grogware.com/other/DDWFTTW.png
I think most people wouldn’t have trouble believing that this cart will travel faster than the nut is being pushed. You can even introduce some inefficiencies (gearbox slippage perhaps) and it is still able travel faster than the nut. The wind driven cart works the same way. Sure the propellor isn’t 100% efficient but it’s efficient enough to get the job done. This has been demonstrated by Jack Goodman, JB, Spork and others.

281. Anonymous

“Can the inventors of the device please say whether the prop is being driven by the wind and turning the wheels, or whether it is being driven by the wheels and is acting as a propellor.”
The prop-shaft applies torque to the prop in the direction the prop turns. It always acts as a propeller.
“”Both” is not an answer. “Driven by the wheels” is not an answer…”
That the torqe is applied by the prop-shaft to turn the prop is always true. The question of what powers what is a bit more subtle, but it seems you don’t do subtle.
“I went to all the trouble of actually doing the math and not even Mark noticed. I pout.”
You should have done it right.
“Mark was wrong to argue that the whole concept of DWFTTW is wrong but right, as far as I can tell, that this prop thing running at wind speed is definitely not doing what people are claiming it does.”
And now you’re wrong.
“It was supposedly invented in the 50’s by someone named Andrew Bauer. If he’s still alive, he’s not apparently participating in this thread.”
Bauer is alive but not able to pursue this sort of thing any longer. I spoke to his wife and one of his partners recently. Both very nice folks.

282. ratwerks

Steven Wells @ #298
You are correct, the forces you ask about are the same regardless. At cart land speeds above zero there is always a breaking force on the wheels and there is always thrust from the prop. This question accurately addresses one of the most common misconceptions; that the propeller is ever driven by the wind in order to drive the wheels. The spork-cart is started in motion by the wind with the propeller (and the rest of the cart) acting as a bluff body. Once rolling the wheels drive the propeller and the propeller provides thrust at some velocity below the road speed but above the airspeed of the cart.

283. AirShowFan

I’m going to tentatively dip my toes in this conversation…
I am a huge fan of ScienceBlogs, and was shocked to see a ScienceBlogger refer to someone making a reasonable claim as a “bozo”. I guess I am a “bozo” too, since I too think that DWFTTW is possible.
To answer the very latest couple of questions: There is a braking force on the wheels, and the prop provides propulsive power. These forces are indeed the same in any reference frame. The prop pulls the cart downwind, the wheels try to hold it back upwind.
To show that this is not perpetual motion: The power into the wheels is wheel traction force (the aforementioned “braking force” on the ground) times the ground speed. The power out the prop is the prop thrust times the air speed (be it the air in the wind or the air blown by the prop; Pick either). In order for this to not be perpetual motion, the power out the prop must be less than the power in the wheels. And in order for this to move DWFTTW, the thrust from the prop must be greater than the traction in the wheels (since the thrust is also opposing drag, etc). This is doable since the airspeed past the prop is lower than then ground-speed past the wheels. If there were no wind, airspeed would be the same or grater than ground speed, and the vehicle would not work. Only when airspeed is lower than ground-speed, i.e. there is wind, can prop power (airspeed x prop thrust) be lower than wheel power (ground-speed x wheel traction force) with prop thrust still being greater than wheel traction.
And why is the treadmill test not equivalent to being outside in the wind? Sure, the machine takes energy from the treadmill, but from an equivalent frame of reference outside (from a balloon) the machine takes KE from the Earth (since that braking force slows down, by just a tiny bit, the speed at which the Earth goes by under the balloon). Where the energy comes from (the air or the Earth) does depend on your frame of reference (which is why, to say something that should make sense from any frame, we say the energy comes from the difference in velocity between the air and the Earth).
Now I might go read the rest of this thread and see if I can specifically address the questions of other skeptics. Or I might not, since I don’t appreciate being called a bozo.

284. TFox

@298
Please be careful — from your language it seems to me that you believe that your question is the same as #291, but it does not seem to be the same question to me. #291’s “driven by”, to me, implies a direction of energy flow, which does indeed depend on frame. (You did read the links before replying, yes?) Your question involves forces, which I think both you and I will agree are independent of frame. At steady state, the wheels have a backwards force from the ground, and a forward force from the propeller. The propeller has a backwards force from the wheels, and a forward force from the air. All forces balance to zero for each; as we’re at steady state. I’m not entirely sure what you mean by “braking” and “propulsive”, but hopefully this helps.

285. Mark C. Chu-Carroll

Spork:
Check out the history of this blog. When I’m wrong, I admit it. Not in backhanded fake ways, but honestly.
I spent some time on the train trying to work it out – and I’m not convinced. It’s certainly a lot more complex than I thought it was initially. I was certainly wrong about the simplicity of it.
But analyzing it as a rigid geared system, I don’t see how it works – and the problem is the same as what I’ve been trying to argue about the propellor. I’ll try to put together some diagrams later tonight, provided my kids cooperate; otherwise I’ll do it some time over the weekend. But according to my attempt to analyze it, in a rigid geared mechanism, there’s a major drag component which is a real problem. In Stephen’s model, when the cart is moving at a speed equal to the speed of the top belt, the wheel touching the top belt isn’t moving anymore – it’s dragging against the top belt. So the top belt can’t be driving it.
The situation with an air driven vehicle isn’t quite the same, because the air is fluid. So like the rigid system, there’ll be an equilibrium point where the drag equals the force, and I think that that will necessarily be slower than the wind.
This will all be clearer when I have pictures to go with it.

286. ThinAirDesigns

Nerull @289:
>But if you are afraid to actually test it
>in the real world, than there is no reason
>for anyone to listen to your cries of “It
>wooooorks!!!! WHY WON’T ANYONE BELIEVE US!!!”
What “fear”? It’s been tested in the “real world” and posted repeatedly.
JB

287. ThinAirDesigns

Stephen Wells @291:
There is *always* a braking force on the wheels which drives the prop.
This can be demonstrated on our larger cart with the long loose cogged belt. It is the side moving *down* from the prop axle to the wheel axle that remains under tension at all times.
On the human sized ride-along version I am building I plan on transfering power from axle to axle using carbon/kevlar tow (thread) and two spools. The empty wheel axle spool will pull from the filled prop axle spool. This will end all confusion regarding what is pulling on what.
JB

288. David Grogono

Mark:
>But analyzing it as a rigid geared system, I don’t see how it works
>This will all be clearer when I have pictures to go with it.
Take a look at the image I linked to in #302. This is mechanically identical to the cart that JB and Spork built but instead of using a propellor and air, it uses a threaded rod and a nut. I suspect you’ll easily see that this “cart” can move faster than the nut that is pushing it.

289. spork

“When I’m wrong, I admit it. Not in backhanded fake ways, but honestly.”
I look forward to it.
“I spent some time on the train trying to work it out – and I’m not convinced. It’s certainly a lot more complex than I thought it was initially”
I originally conceived it as a brainteaser (and later found that others had actually built them). It’s not supposed to be obvious. I would not recommend trying to understand the prop cart until you accept that wind powered DDWFTTW travel is possible. The ice-boats are an easy place to start because they can be analysed with a simple diagram like this one:
http://www.putfile.com/pic/8419299
Or you can imagine a much simpler cart such as this one:
http://www.putfile.com/pic/2794071
From there we can quickly get to the workings of the cart. But be warned, even once you’re pretty sure you understand it, about 100 more questions will pop up.

290. spork

Incidentally, you should know that some of the most compelling posts on here are from people that once swore to me that this is utter nonsense.

291. spork

Here’s a very good place to start too. It very clearly demonstrates the same principle as the cart I linked above:

292. PeteC

I mean that the boat doesn’t turn by itself; someone physically pushes the tiller or turns the wheel, and that person probably ate a good breakfast that day. In our scenario that energy would have to come from the wind and not from breakfast.
Strange that #271 gave the impression that I didn’t know what tacking is.

293. Stephen Wells

Airshowfan @305 seems to have the physics right. I was right about the power but wrong because I didn’t go on to check the forces. Neat. Thanks.

294. Michael C

It works.
Let’s look at a mechanical analogy first. I’ve made what I think is the simplest possible version of a machine that moves faster than the medium that is “pushing” it. See

The working principle is this: when a cotton reel rolls along a table, there are always points on the cotton reel moving slower than its centre of gravity. In particular, if we look at a point that is in contact with the table, its velocity is 0. If the reel is rolling with speed V, the speed of a point half-way between the centre of the reel and the point in contact with the table is V/2. If we can push (or pull) with a constant speed against this point, the cotton reel will advance faster than the thing that is pushing/pulling it. It’s easy to see in the video that this is the case, and indeed anybody can make their own version of this little device to see how it works.
What has this got to do with the Direct Down Wind cart? If we replace the strip of paper with the wind, and we can make the wind act on a part of the cart that’s going slower than the whole cart, we have a cart that will run down wind faster than the wind. But how do we make the part of the cart that the wind acts on move slower than the rest of the cart? That’s where the propeller comes in. A propeller has an effective speed through the air that is defined by its pitch and its speed of rotation. If the rotating propeller is moving through the air slower than its effective speed, it will exert a force against the air tending to make it move faster. That’s how a plane can start up, or accelerate from a given speed. Conversely, if the rotating propeller is being pushed though the air faster than its effective speed, it will exert a force in the opposite direction, tending to slow down its movement through the air.
To make the cart work correctly, we gear the propeller to the wheels, with fixed ratio gearing. We make sure that the effective speed of the propeller through the air is less than the speed of the cart along the ground, defined by the rotation of the wheels on the ground. We now have part of the cart (the propeller), effectively going slower than the rest of the cart. The essential point is that the wheels are in contact with the ground and the propeller is in contact with the air. The combination of propeller geared to wheels makes use of the relative motion between ground and air, which remains constant no matter at what speed the cart travels.
Let’s say that the speed of the cart along the ground is V. The effective speed of the propeller through the air is V*A, where A is a number between 0 and 1. If we have a theoretically “perfect” cart, where there are no losses due to unwanted friction and drag, it will reach a state of equilibrium when the real speed of the propeller through the air is equal to its effective speed. If the speed of the wind is W, the real speed of the propeller through the air is V – W (this value will of course be negative if the cart is going slower than the wind). To find the speed where the cart will reach a state of equilibrium, we only need to do a very simple calculation:
Equilibrium is attained when real speed of propeller through air = effective speed of propeller through air, that is to say when V-W= V*A.
Therefore V = W/(1 – A)
V is greater than W. By choosing a value for A close to 1, we can make V as large as we like.
In reality there are limits. There will be friction in the bearings, tending to slow down the speed of the cart on the road. There will be air drag on all parts of the cart that are not part of the propeller, tending to slow down the cart to the speed of the air around it. And no propeller is 100% efficient. However, there is no theoretical minimum value for these losses: we can make them very small indeed, since the design is so simple. To stay at constant speed, the cart only needs to develop enough force to overcome these losses.
Michael C

295. ThinAirDesigns

@313:
The problem with the freewheel that you link to is it functions just fine, but it doesn’t visually demonstrate the principle.
A: You have to *tell* someone that it’s a freewheel sprocket.
B: They have to believe you when you tell them — and as this thread demonstrates, most people only believe what they want to.
The twin spool method makes it very clear that there is no way to cheat the system — the wheels are driving the prop and it can’t be the reverse.
Additionally by placing the spools directly on the axles you eliminate the 90d gearbox or twisted belts and associated losses altogether.
It’s not a done deal, but I’m leaning toward the spool method.
JB

296. Mender

“The situation with an air driven vehicle isn’t quite the same, because the air is fluid. So like the rigid system, there’ll be an equilibrium point where the drag equals the force, and I think that that will necessarily be slower than the wind.”
The reason I provided the numbers in #184 was that I felt that at some point you’d need them to address this very issue. It helped me understand how the energy entered the system and how much was available. Have fun!

297. SteveM

re 307:
Mark, my solution works for the cart moving faster than the wind (the upper belt), the problem is that it can only go faster. If it is slower, it will not accelerate but instead be driven backwards. Thats why they always need a push start in the videos. In the “self start” video, we see the cart get pushed foraward by the fan, but there is no indication of what the airspeed is, nor that the cart could self accelerate to faster than the airspeed.

298. Mark C. Chu-Carroll

Ok, I give up. I admit it; I’m wrong. I think that I’ve got an analytical solution showing
that it works. I’m drawing pictures and writing up the explanation, as time permits. Sometime this weekend, the apology and explanation will be posted.

299. Subductionzone

Well it looks like spork and JB are going to get another apology, if Mark lives up to his word. At least he has toned down his original rant. And yes what he posted at the start definitely qualifies as a “screed”. But I have yet to see Mark try to defend his obviously wrong remark that the treadmill is not a inertial frame equivalence to a constant (and controllable) tailwind on the street. Since he was wrong about that and has not admitted his mistake then he is implying that spork and J.B. hoaxed the whole thing. You cannot have it both ways. Or their is a third possibility, he cannot detect an inertial frame of equivalence, in which case all of his other physics conclusions are highly suspect since IFE’s are in actuality fairly simple.

300. Mender

Steve, the idea that the cart needs a push to get started is wrong. The self start has already been shown. The break even speed for the small cart is 2.7 mph. Above that there is a forward force on the cart, causing it to accelerate until the energy in is balanced by the drag. This occurs at a speed higher than wind speed whenever the wind exceeds the break even point.
Mark, in #251 Steve adds one more gear to allow the correct rotation. Three gears, two inputs, one output – sounds an awful lot like a planetary gearset. Playing with one also helped me to see this. I tend to depend on physical objects rather than math symbols.

301. Mark C. Chu-Carroll

There’s now a correction at the top of the post, so that I don’t keep calling people names after I know I’m wrong; a more detailed analysis of the real (and extremely fascinating) math behind how the thing really works, and how I screwed up, will follow hopefully sometime this weekend.

302. Subductionzone

Oh my, it looks like Mark and I cross posted. Kudos to Mark for admitting he was wrong, I know, I was there myself. Well welcome aboard to the bandwagon, it is not such a bad place to be.

303. Mender

Mark, congrats on seeing this through! I await your summary!
Please include all the clues that helped you, the process of getting to the right conclusion is sometimes as devious as the device.
I’m sure you can now understand why this is such a fun brainteaser!

304. Scott

“I’m sincerely sorry for calling the guys doing the experiment bozos.”
Kudos for that! But you actually used the term a bit more freely. From post #17, four posts after I sounded the discordant note.
“That doesn’t change the basic underlying fact that the bozos who believe this story don’t get: you can’t extract energy from ‘wind’…”
Is it perhaps appropriate to issue a somewhat broader apology? People make mistakes, and they are not necessarily “fools” who must be “squashed”.

305. Mark C. Chu-Carroll

Re #322:
Give me a break. I’ve admitted that I was wrong. I’ve admitted that I blew it. What more do you want?
As far as the treadmill goes: there are plenty of ways that a treadmill can be set up so that it’s not an equivalent reference frame. A torque-adjusting treadmill (that is, a treadmill that adjusts the force on the belt so that it’s moving the same speed whether it’s under load or not); a sloped treadmill, etc. And there are plenty of people on the internet who claim to have invented free energy systems, complete with demonstrations, who don’t understand what all of the inputs to the system are.
Based on my initial calculations, I thought that the cart couldn’t work – it looked like Yet Another free energy system – something that could extract energy from still air – so I assumed that there was a fault in the treadmill setup. As I said in the updated post, I blew it. I was wrong. As far as I can tell, the thing does work, and I’ve not just admitted down here around comment #300, but right at the top of the post, where anyone who reads the blog will see it.

306. Mark C. Chu-Carroll

Re #326:
I’ll do my best. It certainly is an interesting problem. It looks very simple initially, but it turns out to be quite subtle. Hopefully I can make it clear both why it works, and why it’s tricky.

307. Mender

I can’t speak for anyone else but it may be time to lighten up and welcome Mark into the club. He goofed, he learned (considering what usually goes on around this device, this is a huge accomplishment in itself) and he admitted his mistake.

308. spork

I have to say I WAS WRONG. I wasn’t willing to believe Mark would make this blog right.
Thanks Mark for proving me wrong – seriously.

309. Chris

Great to see that Mark could make his “head around this”. And kudo’s to him to admit the wrongdoing so quickly.
Let’s hope that the whole affair will teach the people a lesson or two. What i have seen on the net about this topic pretty much looks like what (we) skeptics accuse the “true believers” of: confirmation bias (here, just in reverse), lack of willingness to check facts, etc.
I can perfectly understand that science guys are annoyed by the sheer amount of free-energy scams around. So many crackpots stealing so much time from them in which they could do better things. However, this affair shows that things should not be dismissed just because it seems wrong, until they are definitely proven wrong. People should be more cautious before making a judgement.
In the end, science knows a hell lot of stuff, but then there is a lot of stuff out there that we just don’t know yet. If we would, there would be no need anymore for scientists to explore stuff. Also, as shown by this, even if things are supposed to be known, actually they are not, at least to many people.
And who knows, maybe someone, some day will create a machine that initially looks as if it is a perpetuum mobile, but in the end turns out to “tap” into some source by some way not yet known or well understood. Although i don’t hold my breath for that to happen anytime soon, of course. What i want to say is that even if it looks impossible first, there _might_ a grain of truth in it that would be overlooked if it is dismissed directly. Everything should be checked and tested, until a definitive conclusion emerges, even if that means to waste some time onto something that is doomed to fail.

310. Clive

Well, I don’t believe the device, from a standing start, will ever move directly downwind faster than the wind. Mark is (or was?) correct!
In the treadmill scenario, the rotating components connected to the propeller (including the propeller itself) are storing energy. The person placing the device on the treadmill enables that energy to be stored by holding it there the with wheels in contact until everything is up to speed. Then when it is released the stored energy is partially converted to thrust through the propeller which initially accelerates the machine. With the treadmill version of the demo the gearing to the propeller can be effectively be made as high as desired, thus also generating as much thrust as desired, and with enough mass in the rotating components it can also store considerable energy. However, if the treadmill had enough clear space ahead (and the steering was also straight and true!) we’d eventually see the stored energy dropping to a point where the speed decreased back to less than that of the treadmill and it would eventually fell off the wrong end!
From a standing start on perfectly flat ground, etc., the machine will never reach the wind speed using only energy extracted from the wind. In fact I think you’ll find the best case would be to have the “propeller” totally disconnected from the wheels altogether, and the vanes at right angles to the wind so you just have “classic sails” (in effect). In that case, friction would stop it from ever reaching more than some fraction (less than 1!) of the wind speed. If the propeller is connected to the wheels, then the effect of the wind passing over it initially (when the speed is zero) is to oppose the wheels turning forward.
If there ever was any forward movement (and also no slippage between the wheels and the ground), then the opposing force from the driving wheels against the ground must be equal to any “thrust” being generated by the propeller. You can’t get useful forward thrust from a propeller connected directly to wheels on the same vehicle!

311. spork

“Well, I don’t believe the device, from a standing start, will ever move directly downwind faster than the wind. Mark is (or was?) correct!”
Sorry, but it really does.

Proof that DWFTTW is possible:
Let:
Â   Â  Vc:g = velocity of cart w.r.t. ground
Â   Â  Vw:g = velocity of wind w.r.t. ground
Â   Â  Fp = force of propeller on wind (thrust)
Â   Â  Fw = force of wheel on road (resistance)
Â   Â  Rw = radius of wheel
Assume an initial state where:
Â   Â  Vc:g = Vw:g + Delta,
Â   Â  where Delta = some small positive differential velocity, sufficient to demonstrate DWFTTW. Now, if the device is accelerating in this state, then DWFTTW is demonstrated. We must prove that there exists some configuration of the device such that, in this initial state, propeller thrust exceeds road resistance:
Â   Â  Fp > Fw
Assume that Fp < Fw. Modify the system by doubling both the wind speed (or treadmill speed) and the size of the driving wheel:
Â   Â  Vw:g := Vw:g * 2
Â   Â  Rw := Rw * 2
Â   Â  Vc:g := Vc:g * 2
The doubling of the cart speed (Vc:g) resets the experiment. These modifications will have the effect of maintaining the same rotational velocity of the wheel, while cutting the road resistance in half (a bigger wheel has more leverage than a small one, so less force is needed to produce the same torque). Note that the propeller will rotate at the same speed as before, and thus Fp will not change. Repeat the above step, until Fw has been cut in half enough times to achieve positive acceleration, thus demonstrating DWFTTW. Q.E.D.
Here’s the pseudocode, for any fellow programmers out there:
while(Fp < Fw)
{
Â Â  Vw:g := Vw:g * 2
Â Â  Rw := Rw * 2
Â Â  Vc:g := Vc:g * 2
}
What this proves is that with enough wind and the right drive ratio, the cart can do what countless videos have shown it doing. I chose to hold the propeller thrust (and rpm) as the invariant, to eliminate any possible non-linear effects. I tweaked the wheel size (instead of the gear ratio or propeller pitch), to make it easier to visualize. I’m sure one could formalize this proof to satisfy the most demanding mathematician (sorry, I’m an ME with a CS Ph.D.), or devise a similar proof that held wind speed invariant; I will leave that to others.
Now, who is still a skeptic?

313. AirShowFan

Thumbs up for Mark CC, for the polite apology and for not editing the original blog post to make himself seem any less rude than he had been earlier.
And hey, I understand the rudeness. I too am a skeptic, and Perpetual Motion proponents (like psychics who speak with the dead, homeopaths, faith healers, etc) deserve all the rudeness we can throw at them.

314. ThinAirDesigns

I too must admit that I was wrong. I never imagined that Mark would provide such a real, no excuses apology. Kudos to you Mark — Props and much respect. Thanks
JB

315. TFox

@#334
Hi Clive! Thanks for stopping by, this is a great problem, and you’re going to love it. To understand what’s going on, play with a yo-yo first. Watch

if you want, but it’s much better to do the experiment with your own hands. Once you understand how a yo-yo can travel faster than the string pulling it, then work on the propeller cart.

316. KJC

Good job Mark. Pointing out that SWIWOTI is easy admiting the IAWONTI is much rare and more difficult you did a really good job. Also kudos to spork for admitting he was wrong that the appolgy would not be forthcoming. Debate is fantastic and this has been one of the most enjoyable discussions I have read. I am sure many of us have been wrong and many of us have learnt from this many lessons for which we should also be greatful. Now if only I can convince my PhD physics colleagues that this works……

317. Clive

@#339
I don’t see the cotton reel trick as being relevant at all.
Imagine the cart sitting motionless on the ground and also that there is no wind at all. Assume also the cart has perfectly frictionless bearings and gears. Clearly it doesn’t move at all to start with. Now, we’ll simulate the force of the wind by pushing the cart gently by hand – enough to get it started and then moving at some slowish but constant velocity. Does the cart accelerate away because of the thrust being generated by the now rotating propeller. Of course not – we are still pushing and need to always be! And it’s not because of the small amount of air friction caused by the forward motion. (We can even assume that is also zero due to our incredibly streamlined design if you like.) It’s mainly because the force we apply is balanced by the wheel pushing against the direction of travel, and friction of the propeller against the air it is having to move once it starts rotating.
For this cart to end up going faster than the wind, there would have to be some constant speed at which we can push it by hand (constant force) where it would still (somehow!) generate enough thrust to accelerate away from our hand. Where will this magical transition point happen? At any speed slower than this magical transition point the cart will still need a force applied to keep it going at constant speed, yet the slightest bit more is enough for the propeller to suddenly generate enough new thrust for the cart to leave our hand behind? It won’t happen with the hand pushing it in still air, and essentially the same argument applies to the tailwind also.
Having the propeller connected to the wheel is not a lot different from the idea of putting a fan on the back of your yacht in the hope it can fill your sails and move you along at a decent clip, when actually you’ll go nowhere! The wheel will be doing “work” on the ground against the direction of travel, equal exactly to the work being done by the propeller (plus any friction in the mechanism, and from the air, etc.)

318. Don

I don’t think you are wrong, Mark.
First, the treadmill test is meaningless. The treadmill motor is driving the prop so that it is grabbing the still air and accelerating it to produce a force to the right that nearly exactly balances the forces pushing it to the left.
I predict that you could run the treadmill slightly slower and make it level and the cart would remain in place on the belt. By their reasoning, this is equivalent to the cart running with a wind exactly at wind speed. But it should also be able to do this if the treadmill was running slower, ie it should still remain on the belt, it is just equivalent to a slower wind speed. But I predict it would move off the belt to the left. And if you speedup the belt the cart will move off to the right. If they are correct about the treadmill representing wind speed then the cart should move to the right for *any* belt speed.
Sailboats and iceboats have no connection with the cart at all. They extract energy from the wind using aerodynamic effects available while running fully or partially cross wind. Running directly with the wind only allows energy extraction from drag. Sailcraft on reaches experience relative winds greater than the ambient wind, This is never true for the cart.
Now for the large cart on the road. The prop acting as a sail is a red herring. With a good, low rolling loss cart it will get up to wind speed just from the drag on its frame. The prop or sail will just get it there quicker.
With the prop there, let’s ignore the frame drag for now. As it starts from standing, there is drag on the prop so that the wind speed is lowered on the downstream side (in front of the cart). This starts the cart moving and the prop turning. As long as the screw velocity of the prop is less than the cart speed, this continues. At the point that the screw velocity matches the cart speed, the drag is zero but the drag on the rest of the cart continues to accelerate it so that the prop’s screw speed is greater than the cart speed. Now the prop is ‘grabbing’ air and accelerating it back so that the air directly behind the cart is moving slower than the wind. The energy to power this comes from the wheels so this slows the cart. But the energy taken from the motion of the cart to drive the prop is almost all recovered from the thrust from the prop so the cart travels very close to wind speed.
The air directly behind the prop is traveling slower than wind speed so the wind indicator behind the prop will indicate the cart moving faster than *that* air but it is still slightly slower than the wind.
Repeat the demonstration with the wind indicator outside the effect of the prop to prove me wrong.
Also repeat the treadmill experiment as I have indicated to prove me wrong there.

319. Anonymous

“Now if only I can convince my PhD physics colleagues that this works……”
And that’s exactly why this has to be on Mythbusters. It’s astonishing the controversy this 5.7 oz toy can cause. How is it possible that even experts in the field can argue over a simple prop geared to a wheel!?
As far as your PhD physics colleagues… do they believe ice-boats can maintain downwind velocity components of 3X to 4X the wind speed? That’s basically common knowledge in that community, and backed up with plenty of GPS data, analysis, etc.
Did they study physics or intuition?

320. Michael C

In #341, Clive says: “Imagine the cart sitting motionless on the ground and also that there is no wind at all. Assume also the cart has perfectly frictionless bearings and gears. Clearly it doesn’t move at all to start with. Now, we’ll simulate the force of the wind by pushing the cart gently by hand – enough to get it started and then moving at some slowish but constant velocity.”
Giving the cart a push does not simulate the force of the wind. It’s perfectly clear that if there is no wind, the cart won’t start. In this situation, if it is given an initial velocity and then left to move, it will slow down and eventually stop.
The cart can only work if there is a difference between the speed of the air and that of the ground.
My demonstration at

shows the mechanical analogy:
– The cotton reel exploits the difference in speed between the strip of paper and the surface of the table to move faster than the strip of paper.
– The cart exploits the difference in speed between the air and the ground to move faster than the air.

321. davem

Clive @341: Your example has you pushing the cart itself. But the wind pushes the propeller, which is moving slower than the cart as a whole.

322. AirShowFan

I have put the time into making one comprehensive page that gathers all the points made to show that DWFTTW is possible. This should be useful in anyone’s analysis of DWFTTW:
http://dwfttw.blogspot.com
Please please please check it out, and let me know if I missed anything so that I can keep it complete.
And feel free to link to it (or link straight to one of its sections) when debating DWFTTW.
I will add some diagrams and animated GIFs soon, but for now I need sleep… I created this page after writing hundreds and hundreds and hundreds of words about DWFTTW on the BoingBoing threads… I’m really just trying to be helpful, this is not spam. (Notice that I did not put the page on my own website, but created a Blogger blog just for this debate).

323. Greg London

So, after finally reverse engineering this thing down to the bones, I wrote up a spec that describes how this bloody thing actually works, and how to build a device that will go directly downwind faster than the wind without any moving parts.
Now that I’ve written it up, I don’t know why there is any controversy around this other than no one understood it well enough to explain it in plain english.
Spec is here:
http://www.greglondon.com/tumbleweed/

324. Clive

@335, @344, @345 and all the other comments I didn’t read the first time…
Having just read this entire thread from beginning to end (but still skimming a bit towards the end!) I can see that I really should have looked more carefully before leaping so enthusiastically in the fray a little earlier (but still also far too late as several others had already said similar things to me)!
I only glanced through the first 100 or so comments (at a guess) the first time through, before fast-forwarding to the end, glancing though a small selection from the tail of this very long tale (but enough to see that Mark sounded like he was changing his mind), and then letting my intuition run wild (as I often do) instead of sucking in a bit more Oxygen before typing!!
My intuition is still not totally happy, but it’s at least willing to negotiate. So I’m going to sleep on it, and let smarter people than me write it all up in terms I might be able to understand more easily later. Someone mentioned the cylinder of tacking ice boats – that seems pretty clear to me, although I also initially found the idea that the downwind component of a tacking yacht could exceed wind-speed hard to accept.
For anybody else joining this party late like I did, it probably pays to read everything, but perhaps the second half especially, not to mention the other discussions to be found elsewhere!

325. Anonymous

So… Greglondon has been kind enough to demonstrate how NOT to apologize. He’s now willing to explain to us exactly how this works, after offering the following quotes as recently as yesterday over here:
http://www.boingboing.net/2008/12/01/downwind-faster-than.html
The overall summary seems to be that:
A) this thing can’t possibly work – it’d have to be perpetual motion.
B) I clearly don’t understand this thing that I built.
The reality is that it does work, I understand it quite well, and I’ve explained it 1000 times in 100 ways over the last three years. I’ll be happy to dredge it all up.
So now Greg is kind enough to explain to us how it “really” works (which is at least a bit better than the insults and claims that it can’t work), but without having read his explanation, I’m perfectly comfortable offering a bet that he has it wrong.
Here’s as much of his recent spewage as I could stomach for the moment from Boing Boing as recently as yesterday:
It’s just that it’s either some hidden transfer of energy (suddenly spinning the wheels adds a lot of energy that takes a while to bleed off) or they’re cheating somehow (going downwind, getting a burst of energy from gusts and then bleeding it off over time)
Actually, if the device can move faster than the wind, then the device should be able to move in zero wind if you give it a push
so it seems like you’re moving faster than the air. But you’re only moving faster than instantaneous air. you’re still moving slower than the airspeed over time.
If the ground is stationary and you have to move the wind, I think you’ll see the wheels don’t move as fast as the wind.
I’d be willing to wager \$50 to the charity of your choice on that. Not the treadmill. stationary ground with a controlled wind blowing over the prop.
Dude. You just created energy out of nothing with that explanation. The only energy you’ve got is the wind
This is why the treadmill experiment is invalid.
But all you’re doing is argument from ignorance.
“Of course the Bible is true. Nobody can prove otherwise.”
You can’t explain it, so you don’t really know if you’re misinterpreting your data
Uh, the issue around you beign known only as “spork33” is that any anonymous knucklehead could claim to have a masters degree in aeronatical engineering, or doctorates in Theology and Canon Law *, for example.
I think I saw Wiley Coyote use one of these to chase the Road Runner one time.
This isn’t an aeronautical engineer talking. I don’t know what this guy does for a living, but if he designs aircraft I’d sure as hell not want to be flying in one at altitude.
If they’re aeronautical engineers, I want to know who they work for so I never fly or ride in any of their aircraft.
But you all have made abundantly clear that even if it does actually work, you have no idea how it works.
What you’re yapping about here though is pure pseudoscience.
You either don’t know what your talking about, or you do know but are completely unable to articulate it in any fashion
This news flash just in: several small children were swept away today while running around the schoolyard with pinwheels in their hands. They started running, the pinwheels started spinning, and before you know it, “the wind” picked them up and carried them off to places unknown.
There’s plenty more there – but that should be enough.

326. SC

Many developments since yesterday! I look forward to reading the continuation of this discussion.
Just delurking again briefly to thank SteveM @ #275 for the MIT info – I’ll look into it!

327. Anonymous

Yeah, pretty funny, GregLondon goes on forever telling us it’s not possible and then rather than say he was wrong, suddently he’s designed one of his own. LOL
Think about putting it on the treadmill and showing us what it does Greg. If it works on the treadmill, it works on the street and visa versa.
We have never made any claim that the prop-cart design is the only way to do it — in fact we have described and presented drawings of numerous other designs.
I’ll congratulate you on your design as soon as it climbs the treadmmill. Until then, it’s just a pretty picture.
JB

328. TFox

What I love about this problem is that everyone who studies it eventually comes to the same conclusion. Airplane on a treadmill caused lots of argument, but it was based on a cleverly hidden ambiguity. There’s no ambiguity here, everyone has always had the same understanding of exactly what was claimed, and yet many many technical people insist that it can’t work. After further study… we get a consensus.
“Science. It works, bitches.”
(@348: Let’s try to be kind to Greg: people say a lot of things in the heat of an argument. He’s not only finally understood, he’s also built and tested a new design, with what I see as potential added pedagogic value. Besides, he uses OpenOffice.)

329. SteveM

It’s astonishing the controversy this 5.7 oz toy can cause. How is it possible that even experts in the field can argue over a simple prop geared to a wheel!?
Perhaps you don’t remember the great “Monty Hall problem” controversy. A very simple probability problem had PhD’s in math and statistics arguing for and against Marilyn Vos Savant’s solution.
Secondly, I think that like Mark says “the worst math is no math”, meaning the verbal description of how this works sounds very much like how most perpetual motion machines are described, and that just gets people in the mindset of proving it wrong rather than analyzing what is really happening. When this thing is summed up as “wind powered moving faster than the wind” it sounds a lot like “battery powered generating more power than the battery” (and that is a real claim by one device).

330. spork

“Let’s try to be kind to Greg: people say a lot of things in the heat of an argument. He’s not only finally understood, he’s also built and tested a new design, with what I see as potential added pedagogic value.”
Sorry, no can do. Greg has followed the script to a tee. He argued it was impossible and called us fools all along. When the tide turned he decided to reconsider. Once he decided it was in fact possible, he explained that he in fact understands it better than we do (only to demonstrate that he doesn’t). And then he builds a particularly poor version of a vehicle I’ve posted a thousand times to explain his brilliant new concept.
But somehow he was never wrong!? And still, his vehicle won’t work in real life – and our inferior model works quite nicely. As I said, conceptually a vehicle along those lines can work (I’ve used it as a way of describing DDWFTTW travel many times). It’s just that he’s designed it so incredibly poorly that I’m willing to trot out my same old tired \$100K bet that his won’t work – before I even see his attempt.

331. Anonymous

“Secondly, I think that like Mark says “the worst math is no math”, meaning the verbal description of how this works sounds very much like how most perpetual motion machines are described”
Perhaps, but I’ve been posting the math for 3 years on several forums, along with many perfectly valid and rigorous text descriptions that I’m quite certain Mark would have accepted as proof had he troubled himself to ask me for instead of ridiculing me first.
I don’t intend this as a dig against Mark. His apology is quite sufficient. I just don’t accept your premise that no valid proof has been available.

332. Mark C. Chu-Carroll

As I work on the new post and look over comments, something spork said in one recent comment twigged me to what I’ve realized is a very simple way to explain what I misunderstood about it, and looking over a lot of the comments, I think a lot of other people are misunderstanding the device in the same way.
The wind does not directly spin the propeller. What I thought was going on was that the wind was spinning the propellor; the spin of the propellor turned the wheels. If it worked that way, then you’d be looking at a perpetual motion scenario: wind turns the propellor; propellor turns wheels; turning wheels turn propellor; propellor pushes cart forward. That’s totally circular, with no energy source once you reach windspeed.
But that’s not how this works.
What’s going on is that the propellor is just a sail. The wind isn’t spinning the prop! The prop is a dead wind catcher.
So – the wind pushes the cart forward. The forward motion turns the wheels, which turns the prop. The prop is spinning from the frictional rotation of the wheels.
The spinning prop becomes a fascinating sort of sail. The easiest way that I’ve found to think about it is that the propellor plus the cylinder of moving air that it produces act as a sail. So you’ve effectively got a sail that’s moving slower than the vehicle.
The maximum speed of the vehicle is going to be complex to compute; probably beyond my fluid dynamics. You’ve got drag from spinning the prop, which turns into a braking force against the wheels. But you’ve got a very good sail mechanism. The equilibrium is going to depend on the ratio of the gearing, the size and pitch of the propellor, and the diameter of the wheels. But it’s pretty clear that you’ll be able to get it above the windspeed with the correct setup.

333. Scott

“Secondly, I think that like Mark says “the worst math is no math”, meaning the verbal description of how this works sounds very much like how most perpetual motion machines are described”
I disagree completely. Mark used math to buttress a false premise — one that he did not arrive at via math. His premise was that the only source of energy was the difference in velocity between the air and the vehicle. When that premise was refuted he did not attempt to engage in a reasoned discourse.
The last paragraph of the post was going to be rather harsh. But I refreshed the discussion is a second window before posting and saw #356. Mark seems to have tacitly acknowledged that this problem can be explained without math, and indeed the math needed to precisely describe what is happening may be so complicated so as to prohibitive in the context of a blog discussion. Well, better late than never.

334. SteveM

I just don’t accept your premise that no valid proof has been available.
I did not say that no valid proof was available. I was only talking about initial impressions.

335. spork

“What’s going on is that the propellor is just a sail. The wind isn’t spinning the prop! The prop is a dead wind catcher.”
Mark, I’m extremely pleased that you’re engaged in the workings of this thing now. You’re right that the prop is initially a dead wind catcher as the cart starts from a stand-still. However, as the cart picks up speed and the prop starts to spin, the prop starts to see a relative wind that comes more from the leading edge. At some point still below wind speed, we get laminar flow over the prop, and it begins to act as a lifting surface rather than a bluff body. It gets particularly interesing above wind speed – as now the air passes from the front side of the prop disk to the back – moving from low pressure to high (proving that the prop acts as a prop and not a turbine). But… the relative wind is still striking the back surface of the prop. This is possible because of the angle of the prop blade and the path it follows through the wind. It is directly analagous to the ice boat on a 45 degree downwind course (see my vector diagram)
“The maximum speed of the vehicle is going to be complex to compute; probably beyond my fluid dynamics.”
If you consider a lossless system, the theoretical max speed is very easy to compute as a multiple of the true wind speed. It’s a very simple function of what I call the “advance ratio” of the vehicle. That is the distance the prop woud theoretically advance in one rotation divided by the distance the wheels advance in the same single rotation of the prop.

336. spork

“Mark seems to have tacitly acknowledged that this problem can be explained without math, and indeed the math needed to precisely describe what is happening may be so complicated so as to prohibitive in the context of a blog discussion. Well, better late than never.”
In fact I believe a completely rigorous proof without math is both possible and more intuitive in this particular case. However, the necessary math can be given with a relatively straightforward vector analysis and some appropriate description.

337. The Science Pundit

SteveM@#353

Perhaps you don’t remember the great “Monty Hall problem” controversy. A very simple probability problem had PhD’s in math and statistics arguing for and against Marilyn Vos Savant’s solution.

I’m very familiar with that. In fact, I’ve been in arguments over it. Another one which gets similar reactions is when you try to explain that 0.999… = 1. Granted, nobody who’s received a passing grade in calculus–much less any PhD’s–would ever fall on the wrong side of that argument, but the attitudes of the denialists is identical.
I’m looking forward to Mark’s new post. For the record, I knew he would come around. It’s not that I’m smarter than others here (I’m definitely NOT), but I have been reading Mark’s blog since he was on blogger. But since we’re doing Mea Culpas, I admit that when I first read this post, I didn’t bother watching the video and bought into the “device is a perpetual motion machine” narrative. I declined from commenting at the time because I didn’t feel that I understood the problem well enough to add anything. But after reading the comment exchange as it was happening, then reading Dave Munger’s post, and finally watching the video, I began to grasp how the cart worked.
Anyway, this has been a really fun discussion and once again, I’m looking forward to Mark’s new post.
ps—I don’t know about that greglondon guy, but as a general rule for this problem–its staunchest critics tend to become its greatest defenders once they truly understand the problem, so they should be welcomed. Cheers!

338. Dave M

Thanks to everyone for this great thread. Just for the record, I found Michael C’s explanations the most helpful (to this non-physicist!). I look forward to Mark’s post as well.

339. spork

“I found Michael C’s explanations the most helpful (to this non-physicist!).”
I still watch his spool video twice a day!

340. intuit

I had been a doubter, for the same reasons as Mark. The tacking sailboats example convinced me that the concept was not impossible, and once I realized that, it wasn’t hard to look at the design of the cart with an open mind and realize that it did indeed work.
But the sailboat example doesn’t work with everyone. The fact that it zig-zags tricks people into thinking that their are somehow different forces at work. For people still having trouble with their intuition, another example illustrated by David Munger might help.
Imagine a normal sailboat. Making the sail larger will never make it go faster than the wind, directly downwind.
But it *will* make the boat harder to brake, to an extent greater than the mass of the sail.
Obviously, then, more energy goes into a wind-powered vehicle than is necessary to move it forward. The cart is only one means of harnessing this energy.
Once you know that DWFTTW is possible, and that more energy goes into a vehicle than necessary to move it forward, the only question is whether this exact design does it.
In this design, the energy required to rotate the prop does not come from the energy that pushes the cart forward at wind speed.
You can think it through, or you can think that the cart-builders are wrong or hoaxers, seeking to trick us or themselves into building a vehicle THAT DOESN’T EVEN DEMONSTRATE ANYTHING NEW.
Man, if I were going to perpetrate a hoax, I would pick something more spectacular. Kudos to the designers of this cart, though, for demonstrating another way that people’s intuitions about the physics of motion, like their intuitions about probabilities, can lead them awry.

341. Mark C. Chu-Carroll

Re #357:
Sorry, but I still maintain that math is the only way to really conclusively prove things like this. The math isn’t a smokescreen, and the fact that I tried an informal explanation for the comment doesn’t mean that I’ve admitted that you don’t need math.
Like I said before, and like I’ll say again: intuitive explanations can easily contain subtle errors or deliberate misdirections. The only way to be sure about something is to do a careful analysis.
Do a websearch for “Brown’s gas”.
Brown’s gas is a mixture of hydrogen and oxygen in exact two to one proportions. Proponents of Brown’s gas claim that you can get more energy from burning Brown’s gas (combining the hydrogen and oxygen to get water) than you’ll consume breaking water into hydrogen and oxygen.
They’ve got a very nice sounding explanation for why it works. If you listen to it, it sounds very convincing. (The explanation involves compressive effects, because supposedly at the right proportions, the mixture “implodes” while burning, which produces more energy than simply burning things.) If you listen to some of the guys who believe in this, it sounds really convincing. Only one problem: it doesn’t work. When you actually sit down and do the math, it just doesn’t work. You can create a system which appears to net some energy. But it turns out to be a result of compressing the gas. When you add the energy it took to compress the gas into compressed air cylinders, you end up with a losing system. But it’s not clear until you actually sit down and do the math, working out all of the energy inputs, and all of the energy outputs. And without the math, you can argue for months with these people, who’ll insist to the end of their days that the amount of energy coming out of the “implosive combustion” is really more than they put in.
Similarly, there’s a ton of “free energy” systems out there that purport to work with magnets. People using this scam have managed to scam investors out of millions of dollars. (For example, look up “Steorm” in the archives of this blog.) Again, the proponents have some really impressive sounding explanations for how their magic machines work, and those informal explanations can sound awfully convincing. And they can argue with the best of us, making their arguments sound really compelling, and making it sound like we skeptics are just being a pack of jerks who refuse to see the truth. Only if you try to do the math to see how it works, it doesn’t, and it can’t. (And amazingly, in every single case of these magnetic free energy devices, the
machines either break down or get stolen or sabotaged before they can be demoed. If you look up Steorm, they made a huge deal about how they were going to do a public demo with a live video feed to the network… Only it turned out that the lights for the webcams were too hot, and the heat damaged the bearings on their machine, which stopped it from working. )
In some cases, the people behind these devices don’t even know that they’re wrong. They really, truly believe in it. For example, there’s a gentleman named, I think, Ron Bearden(?) who’s been travelling around giving talks about his magnetic engine. He’s not making any money from it, and he seems to really sincerely believe in what he’s doing. He really believes his arguments. But that doesn’t change the fact that the damned thing doesn’t work.
The proof of whether physical devices are really doing what their designers claim that they’re doing is a careful quantitative analysis of energy inputs and energy outputs – and that can only be done with math.
In the case of this wind device, I can’t do all of the math in a purely analytical way; it’s frankly a bit beyond my abilities in fluid dynamics. But I can do enough to see how a net force can be generated that will allow it to go faster than the wind. That’s why I’ve changed my opinion, and I believe that it can work. Because the math works.
(An easier way to do it would be to use measuring devices on the wheels, the propellor, and the wind, and show how the forces all work out in a measured situation; that’s also difficult to do, because measurements can perturb the system, and you need to make sure that you’ve got measurements on all of the places where energy can enter or leave the system. That’s easier than the pure analytical solution from a math point of view: you just take the measurements, calculate energy in, and energy out. But it’s a whole lot harder from an engineering point of view.)
And that’s also how I got it wrong in the first place. I screwed up the math of it. If you assume that propulsion comes from the wind directly pushing the propellor to spin, it doesn’t work. The math is the key to figuring out whether or not things like this work. But you’ve got to do the math right.
The whole reason that we’ve invented mathematical abstractions and notations is because informal language is too imprecise. Math is an irreplaceable tool that makes it easy to precisely describe and analyze things.

342. TFox

@354
I don’t think I suggested that he wasn’t wrong; I suggested that we try to be kind to him. Given what he has written, though, I can certainly understand your reluctance to do that.
I also share your skepticism as to whether his build will actually go faster than wind. Nevertheless, this design might still conceivably have instructional value, for those who understand the spool but believe that a propeller cart is somehow different. It seems to have been the tipping point for Greg.

343. ThinAirDesigns

Mark C. Cu-Carrol:
>Like I said before, and like I’ll say again:
>intuitive explanations can easily contain
>subtle errors or deliberate misdirections.
Unlike math. Mmmm Hhhhh.
In the past few weeks, we have seen hundreds of people (including you) convince themselves through bad math that this is impossible. We have yet to see anyone build one that does it and say “it can’t be done”.
>Sorry, but I still maintain that math is the
>only way to really conclusively prove things
>like this.
When your toolkit is limited to a hammer, everything looks like a nail.
Math is wonderful — as is something that is repeatably demonstrable.
JB

344. spork

“Sorry, but I still maintain that math is the only way to really conclusively prove things like this”
I’m a pretty big fan of math. But if my math and physical reality disagree, I’m going with the physical reality every time. Moreover, if I have the physical evidence in hand, that feels like better proof than the math.
“intuitive explanations can easily contain subtle errors or deliberate misdirections.”
As can the math.
“The only way to be sure about something is to do a careful analysis.”
I’d say that’s true whether using appropriate analogies or math.
“In the case of this wind device, I can’t do all of the math in a purely analytical way; it’s frankly a bit beyond my abilities in fluid dynamics”
I actually believe you can. In aerodynamics there are lots of things we do by using simple models based on measured quantities (such as Lift/Drag). We don’t have to derive and solve the Navier-Stokes equations for every problem.
“And that’s also how I got it wrong in the first place. I screwed up the math of it.”
And this is exactly my point. You can do either one wrong. In fact you can use either one to deliberately mislead.
Please don’t get me wrong. I’m all about the math. In aero we take literally years of math just for our B.S. degree. I simply feel that for most people other approaches are more accessible and can be just as convincing.
For what it’s worth, the math for this problem can be done in many ways – vector analysis and energy balance are just two ways. Yesterday JB and I spent an hour at lunch simply reviewing the specifics of the flow over the prop as it relates to the flow over the sail of an ice-boat, and when either could be considered as a prop or a turbine (which gets into energy measurement and reference frames). In fact there are extremely analytical books covering the math and aerodynamics of propellers alone.

345. spork

“I also share your skepticism as to whether his build will actually go faster than wind.”
Skeptical may be an understatement. I’m willing to bet \$100K on it.
“Nevertheless, this design might still conceivably have instructional value, for those who understand the spool but believe that a propeller cart is somehow different.”
I have used the yo-yo analogy for years. It is both different from the prop-cart in critical ways, and similar in other critical ways. It absolutely has instructional value, and I’m serious when I say Michael C’s video might be the best I’ve seen on the topic.
Greg has ignored a couple of basic principles of aerodynamics in his design. The general concept has merit (which is why we’ve used a very similar concept for so long). He simply implemented it wrong. By not feathering the vanes as the move through most of their motion he gives it NO chance of beating the wind. Consider the fact that the vanes on top will be approaching 2X wind speed, and that drag is proportional to the square of the wind speed. Cupping the vanes is the right idea, but it won’t do the trick.

346. Anonymous

Tfox:
>I also share your skepticism as to whether
>his build will actually go faster than wind
Not to rain on Greg’s little “look at my cool little tumbleweed device” parade, but we’ve used a device almost *exactly* like this to demonstrate the principles of our device.
We use it to show to doubters that air blowing at vanes *below* the hub of a spool move the spool the same direction as vane on *top* of the spool. It’s a aero version of the yoyo and string or paper and spool (thank Michael C) demo.
Here’s the video of that device and note the the posted day preceeds Greg’s design. We were having an exhange with a “Rob” on our home hang gliding forum and I made this video over lunch as a demonstration.

I really like people who play with stuff, but to say we don’t know what the hell we’re talking about and the ‘invent’ a version of something we’ve already posted is just a hoot.
Incidently, neither our version of the spool nor Greg’s version of the spool will go DDWFTTW. That’s why he hasn’t shown it doing just that. And yes, I’m quite willing to eat crow the day he proves me wrong.
I’ll be posting a video in a couple hours showing what he won’t.
JB

347. ratwerks

Mark @ #364
I’m glad the math works for you, but I don’t think you get to say math is the only thing that works. I am quite certain that my mechanical sense far exceeds my mathematical sense. In the Brown’s Gas example you cite I would have spotted the compression issue (if you want to jam scuba jugs in a hurry you have to put them in a shower to keep them cool) way before I could sort it out with math (which, in my case, would be never.)
If nothing else I think this puzzler shows that there are many many ways to arrive at truth and that by insisting, not on truth, but rather on some ‘one true path’ to truth we can miss the boat (or cart.)
I congratulate you on your clearly evident mastery of your chosen path. My path involves manipulation of physical objects and their mental analogs and my experience has been that I get there more often and more quickly than almost everybody. Other paths seem to work well for other people. Perhaps the coolest thing about (or defining characteristic of) truth is that it remains invariant regardless of how we describe it.
I look forward to reading your math when you publish. I’ve got to say, though, that it will be my understanding of the physical that illuminates my understanding of the math rather than the other way ’round.

348. Don

Here is a simple way to convince me it works.
In that demo using the larger cart on a treadmill, run the treadmill at, say, five different speeds (1,2,4,8,16 mph). Hold the cart in place until everything is up to speed and let go. If the cart always moves ‘upwind’ then I will believe I’ve misunderstood and you are right.

349. TFox

JB, spork:
I hope it didn’t seem like I was claiming that Greg was the first to come up with the impeller spool, as intermediate between a thread spool and a prop cart. His was the first I saw, but folks have been working on this system for decades. To be honest, by claiming it had instructional value, I was mostly just trying to find something positive and encouraging to say to him.
It’s not immediately obvious to me that it *couldn’t* be made to work. I think the system needs to be reasonably efficient, but paddlewheels got discarded a long time ago. However, I wouldn’t have guessed an horizontal impeller could power an airplane, but apparently it can (fan wing).
More interesting to me is the question of the maximum speed in air, but my propeller theory isn’t up to it.

350. J-J

The “theoretical” top speed of an idealized prop cart is unlimited. In the real world, there are various things that will limit the speed. Asking for the maximum speed is kind of like asking how fast a bicycle can go with a cadence of 60 rpm. If you keep gearing the bike higher and higher, it would in theory go faster and faster. With the cart, if it had a variable-pitch propeller, you could keep cranking up the advance ratio as it speeds up. At some point, between transmission losses, aerodynamic drag on the chassis, etc., you reach a point where you can’t get ahead any more. Calculating the top speed for any real device would require you to know a lot of efficiency values for the device (rolling resistance, prop efficiency, Cd of the chassis, etc. etc.). In the case of the cart that spork and JB built, it’s a simpler matter, because they don’t have a variable pitch prop, and the limit can (I believe) be calculated geometrically from the transmission gear ratio (1:1 in this case), the wheel diameter, and the advance ratio of the prop. For any given windspeed, I believe there is a speed at which the prop will no longer be providing any thrust, and the device will never quite reach that speed.

351. spork

>>
I hope it didn’t seem like I was claiming that Greg was the first to come up with the impeller spool…
>It’s not immediately obvious to me that it *couldn’t* be made to work…
Fair enough. But look at the orientation of the vanes. The vanes on top will present the same coefficient of drag to the headwind as the bottom vanes present to the tailwind. At wind speed this will have zero net result (but all the vanes will produce drag). Above the speed of the wind the upper vanes will be fighting an increasing headwind, while the lower vanes will benefit from less and less tailwind. It’s a losing battle without feathering vanes (and just a darn tough battle with feathering vanes).
>>
I believe there is a speed at which the prop will no longer be providing any thrust, and the device will never quite reach that speed.

352. TFox

@376
“At wind speed this will have zero net result (but all the vanes will produce drag). ”
Hm, the total symmetry in still air (ignoring the ground) does prohibit net thrust, I hadn’t appreciated that. In all functional paddle/impeller designs I can think of (water wheels, paddle wheel boats, water turbines, the fan wing plane) we arrange to break the symmetry by having the fluid flow cross only one side of the paddles. So Greg would need to add some kind of fixed cowling (breaking the “no moving parts” esthetic) to have any hope. Or feathering vanes, as you suggested. Regardless, propellers seem like a much better way to convert torque into thrust. Thanks!
As for Greg, over on BoingBoing it looks like he’s denying Galilean invariance again, so I may have to retract my earlier statement that he’s “finally understood”. *sigh* I hate being wrong đŸ™‚

353. spork

>> *sigh* I hate being wrong đŸ™‚
Once you admit you made a mistake the pain is all over. Mark did this with unusual style and grace. Greg seems insistent on prolonging the pain.

354. AirShowFan

I think that it’s interesting how Mark CC said that one big obstacle that kept him from understanding how the cart works was the idea that the wind spins the prop like a turbine. From the first time I saw the video that this post is about, where they put the cart down on a treadmill and spin it up and let go and watch it move forwards, I knew that the spinny thing was clearly a propeller powered by the wheels and providing thrust. On BoingBoing I must have written “It’s a propeller, not a turbine” on like 10 different comments. I don’t understand why this is non-obvious.
(However, I do understand why it is non-obvious that this is not a perpetual motion machine. Even given that it’s a powered prop spun by the wheels, someone could say “So the wheels spin the prop, which pulls the cart faster, which spins the wheels faster, which spins the prop faster… that’s a perpetual motion machine!”. The only way to get past that obstacle (to see that this mechanism works in an environment with wind to be slowed down, but NOT in air that is at rest relative to the ground) is to do an energy balance, thrust times speed).
And yes, I’m also getting the feeling that Greg does not see the equivalency between the treadmill test and the outdoors wind test. I’m not sure how many more ways there are explaining this. The best way I can come up with (talking about a balloon, a wind turbine, a Prius, and a combination of them, showing how some of the Earth’s KE is used to power the vehicle, just like the treadmill’s KE) is here.
Oh, and…
Besides, he uses OpenOffice
…cracked me up

355. spork

>> However, I do understand why it is non-obvious that this is not a perpetual motion machine.
I agree completely. We never would have made the vehicles or the video if it weren’t something of a mind bender.
“The only way to get past that obstacle (to see that this mechanism works in an environment with wind to be slowed down, but NOT in air that is at rest relative to the ground) is to do an energy balance”
I don’t completely agree with that. There are several approaches that lead to a satifactory conclusion in my opinion. Energy balance is absolutely one of them. It still amazes me how such a simple toy can continue to help us hone our understanding and question definitions regarding work & energy, turbine vs. prop, inertial reference frames – and it goes on and on.

356. Mender

You’re right there, spork! I’ve learned more about aerodynamics because of this device than I thought I didn’t know – and quite a bit about humans as well. It has been a fascinating journey and I applaud you for selling tickets!

357. spork

“It has been a fascinating journey and I applaud you for selling tickets!”
Shoot – am I supposed to get a piece of the gate?

358. JM

Take up sailing. Any elementary instruction book on learning to sail will explain the physics behind this.
Sailing is also fun, challenging and if you take it up seriously you will eventually get yourself into some serious trouble which will make you look at life with a new sense of balance, priorities and sense of responsibility.
And that can happen on a mill pond like the time I took my 18 month old daughter out in 2 knots of wind on near mirror glass water (don’t ask, I don’t like to admit how much of an idiot I was, but I can be persuaded to brag about how I got of it – if you buy me a drink).

359. ratwerks

Credit where credit is due. Any original work of authorship is indeed protected by copyright as soon as it is fixed in a tangible medium. Kudos to GregLandon for sharing his rights with the rest of us and for sticking with the thinking to the end. Each of us have our own obstacles to understanding, the real test is whether we recognize them as such and continue working until we’ve cleared them.
JM @ 383, surely you just pointed a big fan into the sail, no? I’ve seen Tom & Jerry water ski with that very arrangement.

360. Anonymous

“Credit where credit is due. Any original work of authorship is indeed protected by copyright as soon as it is fixed in a tangible medium.”
Agreed. But most of us don’t go to the trouble to mark it as such on such a horribly concieved piece of work that goes out of it’s way to insult much better work.
“Kudos to GregLandon for sharing his rights with the rest of us and for sticking with the thinking to the end.”
We’ve come across plenty of people that don’t agree with us, and that have proposed other approaches that we may not think will work. We have NO problem with that. We enjoy debating approaches with them. But they don’t insult us continuously while telling us they understand it and we don’t. There are some that do – and we treat them as we treat Greg.
“JM @ 383, surely you just pointed a big fan into the sail, no? I’ve seen Tom & Jerry water ski with that very arrangement.”
You joke, but it actually IS possible to point a fan at a sail and make forward progress – not just in the cartoons.

361. HUDMAN

You joke, but it actually IS possible to point a fan at a sail and make forward progress – not just in the cartoons.

I think you mean while the fan is attached to the sailing craft. It’s also possible to attach a fan to a sailing craft, point it at the sail and make progress in reverse. It all depends on the sail.

362. Anonymous

“I think you mean while the fan is attached to the sailing craft. It’s also possible to attach a fan to a sailing craft, point it at the sail and make progress in reverse. It all depends on the sail.”
Nope – you can actually point the fan straight forward into the sail and cause the boat to go forward. It just depends on how the sail is shaped. This is exactly how thrust reversers work on jet engines.

363. john brand

Hope someone can answer this question. A little over a week ago I followed a link to this blog on wiki. I could swear it did not have Mark’s retraction note on top.
Is there any way a link to the original article on ddwfttw minus the update is out there in the fog and could possibly be the one I read? I really would appreciate an answer. I’m at the point of questioning my sanity as I checked twice to be sure I wasn’t seeing the one with the retraction pasted in, which I had seen way back elsewhere. Thanks.

364. Mark C. Chu-Carroll

@393:
No.
The “original article” is this one: I updated it in place. So there’s no way that you could have come to this blog a little over a week ago, and read the post without the retraction and apology.
It’s possible that someone, somewhere, archived the original post – but if so, they’re doing it without my permission, and I’d be seriously ticked off.

365. brahmabull77

I wonder when the note in the beginning was made. Was it just made now that it was on Veritasium?

Anyway, i will not be calling the auther stupid or a bozo but for once i will understand if someone does. I think it’s uncalled for to call others bozos or stupid or that their ideas show mind-numbing stupidity. Even if the author was 100% certain it could not be otherwise that he was right, namecalling should not take place where people try do explain their thoughts on a scientific subject. Ideas should be heard.

1. markcc Post author

The original post was written on december 2nd, 2008; the note in the beginning was written on december 3rd, 2008. So yeah, definitely not related to veritaserum.

My friend David Kroll pointed out that the mechanism was essentially the same as a tacking sailboat. I still thought he was wrong, but since I know David’s a smart guy, and it *is* true that a tacking sailboat is getting power from the wind to go faster in the direction of its motion that the wind in that direction, I sat down with some paper and started trying to work through how it could work. And doing that, I realized my mistake.

It’s a pretty subtle mechanism that really *looks* wrong. But when you do the math, and you actually work it out,it really does work.

For me, the key was understanding that the propellor doesn’t turn the wheels; the wheels turn the propellor. The propellor is a sail, catching the wind and pushing the car forward; and that forward motion driven by the wind spins the wheel, turning the propellor, and creating an additional forward force. There’s no energy appearing from nowhere: it’s still drawing energy from the wind – it’s just exploiting the difference between the velocity of the air and the ground. The wheels are an engine that does that.

366. Anton K

Long live the scientific method of settling debates and uncovering the truth. You are all beautiful people

367. Mickey Farley

Yeah, maybe you should look up the definition of the word “relativity” before calling anybody using that word an idiot. There is such a thing as Galilean relativity.