Oh God, not again.

December 3, 2008 4:56 PM

There used to be this problem you see, until one of our own kindly settled it. His services are desperately needed once again.
posted by tkolar (191 comments total)

Oh, shit. Here it comes.
posted by hifiparasol at 5:04 PM on December 3, 2008

Sadly, I found that ep of MB to be one of the most poorly explained they've ever done. (Normally that's definitely NOT something I'd say about MB.) They had a great deal of confusion of terms over things like "taxing". Motion is relative, but they kept mixing reference frames without even telling which one was which.

Perhaps they'll do better on this one, which I'm sure they'll tackle.
posted by DU at 5:05 PM on December 3, 2008

Also, this myth is either ridiculous or suffers from the same definitional problem. What does it even mean to be going "faster than the wind but in the same direction"? If you are going faster than "the" wind, then the air should actually be blowing against you.
posted by DU at 5:12 PM on December 3, 2008 [4 favorites]

Gratuitous spork shot at 5:15 of last video. NSFW.
posted by Severian at 5:19 PM on December 3, 2008

Not Safe for Woks?
posted by BrotherCaine at 5:31 PM on December 3, 2008

It may theoretically be possible to accelerate wind in this way, but in the real world you can't help but brake wind.
posted by googly at 5:32 PM on December 3, 2008 [2 favorites]

If you were in a river, and you swam downstream with the current, you would in effect be going faster than the water was in reference to the shoreline.
posted by The Power Nap at 5:35 PM on December 3, 2008

Over a year ago on Boing Boing, I linked to this video from a guy who made a propeller-powered vehicle that he claimed could travel downwind faster than the wind.
This should be no problem at all - in fact, since a propeller pushes air backward, air moving in the same direction as your direction of travel would increase your speed relative to the ground.
Charles Platt made a miniature model of the vehicle and came to the conclusion that there is no such thing as a wind-powered vehicle that can travel faster than the speed of the wind.
My intuition says he may be right, but I'm no qualified scientician or anything. However, this is a wholly different premise to the above, since something that is propeller-powered is, at least in my humble understanding of turboprops, helicopters, hovercraft, and those funky Floridian swamp-going craft things, powered by an external power source (an engine or motor) and not the wind...
Problem of a misunderstanding as regards the problem to be solved?

(Also, first MeFi post, be gentle...)
posted by Brother Dysk at 5:36 PM on December 3, 2008

If we are allowed to decouple the reference frames, then here's my solution:

The wind is blowing at 50 mph. It blows on a windmill for an hour, generating electricity that's stored in a battery. At the end of the hour, the batter powers a go-kart at 100 mph for 10 minutes.
posted by DU at 5:39 PM on December 3, 2008 [4 favorites]

Isn't a wind tunnel and a little vehicle model of the proposed design all you need to test this?
posted by rivenwanderer at 5:44 PM on December 3, 2008

That test is in the video. They claim it moves forward. They make this claim right after the pushes it with his finger. It's all very convincing.
posted by DU at 5:48 PM on December 3, 2008

The treadmill test is seemingly invalid, because there is no way to verify that the tread is moving in such a way as to counteract the wind. This is different from the airplane question, because the wheels are actually propelling the vehicle, so the speed of the treadmill is a valid concern. Also, the surfaces of the vehicle will catch some of the wind, adding additional propulsion.

Finally, the largest problem with the treadmill experiment- on a treadmill, where "at speed", means still with respect the treadmill, the fan is blowing at a constant velocity. This isn't the case with a wind-powered vehicle on solid ground, where wind speed is measured as relative to the ground. For example, if you are driving down the highway at 30mph with a uniform 10mph wind at your back, you have a 20mph wind in your face. On a treadmill, if you are travelling 30mph on a treadmill with a fan blowing a uniform 10mph, you still have a 10mph wind at your back.

In short, invalid.
posted by potch at 5:50 PM on December 3, 2008 [1 favorite]

You'll notice that those treadmills were all in excellent condition, as they were probably rarely used...
posted by Chuffy at 5:50 PM on December 3, 2008

potch: what fan?
posted by Osmanthus at 5:51 PM on December 3, 2008

potch: This is different from the airplane question, because the wheels are actually propelling the vehicle

Surely not, airplanes are propelled by jets or propellers, not the wheels...?
posted by Brother Dysk at 5:56 PM on December 3, 2008

Osmanthus: "potch: what fan?"

Doh. Yeah, somehow I watched the whole obnoxious video assuming they had a fan behind the treadmill. I'll have to think about this some more.
posted by potch at 5:56 PM on December 3, 2008

The only way to truly settle this is mandatory high school Physics instruction. I'll never understand why people think they can reason their way to an answer. This isn't Philosophy, folks, it's Physics. There is a single right answer, and you find it with science.
posted by sdodd at 6:06 PM on December 3, 2008 [5 favorites]

Literal viking.
posted by cortex at 6:09 PM on December 3, 2008 [8 favorites]

Metafilter: There is a single right answer
posted by Lemurrhea at 6:10 PM on December 3, 2008

Obviously there was a fan behind the treadmill. If that were not the case the device would be going up a hill with no source of energy. A perpetumobile so to speak. On the anecdotal front, in sailing it is possible to sail faster than the speed of the wind if one is sailing a an angle approching 90 degrees to the wind. however as one "falls off" and starts running with the wind ones boat speed decreases. Thus if you watch the americas cup you will notice that the boats "tack" down wind rather than run in a direct line with the wind in order to increase boat speed and thus distance toward the downwind mark.
posted by phoffmann at 6:11 PM on December 3, 2008

Oh FFS, a seven minute long video, for what can be shown in about 20 seconds? And you know it's gonna be long-winded when they start with a sepia toned picture of some 19th century contraption.
posted by Xoebe at 6:15 PM on December 3, 2008

Two key questions:

Q1: What determines the total energy available to any wind powered vehicle?
A1: The speed of air relative to the vehicle, cubed

Q2: What effect does air friction have on a vehicle traveling along with the wind?
A2: Friction applies a force that tends to reduce the speed of air relative to the vehicle.

I'm pretty sure these two factors close the door on any faster-than-wind devices. As the vehicle approaches the same speed as the wind, the energy available to it approaches zero.

Some could say "Well, that doesn't stop you from overshooting the windspeed, then generating power from the headwind!". That's basically the same as trying to improve your mileage by sticking a wind turbine out your car window on the freeway. Friction is going to try and slow you back down to the same speed as the surrounding air, and the second law of thermodynamics is going to make sure you can't cheat your way out of it.

In this house, we obey the laws of thermodynamics!
posted by anthill at 6:17 PM on December 3, 2008 [3 favorites]

Hrm, I now realise how badly I missed the boat on the whole propeller thing. Whoops.

Also, I love how, in the video, they do that close-up of the spirit level to demonstrate that the treadmill is level, and it blatantly isn't - the bubble is all the way off to one side...
posted by Brother Dysk at 6:21 PM on December 3, 2008

wow guys better hush. This device is totally legit. Energy comes from wind in one case, treadmill in the other.
Think about it this way: when stopped the prop captures energy from the wind, when moving it captues energy from the kinetit energy of the wheels being turned by the ground/treadmill . Always there is an external energy being tapped so it's not perpetual motion.

Still unsure? You understand that a crosswind can lead to speeds faster than the wind, think why is this different?
posted by Osmanthus at 6:30 PM on December 3, 2008

Yeah no, this isn't a stumper at all. Here's Mark C. Chu-Carroll to explain it all: Wind-powered perpetual motion.
posted by Skorgu at 6:33 PM on December 3, 2008

Another definitional problem here is the phrase "powered by the wind". Does my windmill count? What about a pressure differential? That's how actual sailboat sail. The wind doesn't push a sailboat, the sail acts as a wing and "sucked along" by the pressure differential on the two sides.

So imagine a container at pressure P (which is also the pressure outside the container). You put your vehicle on the inside near the bunghole. Pop it open and the vehicle doesn't move.

Now blow some wind across the bunghole at speed W. This creates a pressure differential. When you pop open the hole, this differential causes the vehicle to come out at speed V.

The only motive power here is the wind, so I think I've met the letter of the challenge. Whether I've succeeded is another question. I've spent a few minutes trying to refamiliarize myself with Bernoulli but not to much avail. I see a velocity squared in there, so it is not clear to me that V is always less than W for all Ws and Ps.
posted by DU at 6:33 PM on December 3, 2008

Oh, crap, here too? Let's see if I can head this off before paulsc shows up: the propeller serves as a simple sail while the wind accelerates the cart from a motionless start, and is then spun ever faster by its linkage to the wheels as the cart approaches wind speed, until it starts generating power and can assist to push the cart past wind speed and into an apparent headwind. At that point, the forces acting on the cart are:

• gravity (irrelevant except to keep the wheels on the ground)
• drag from the air (minimal since we're barely moving relative to the tailwind)
• friction from the wheels and drivetrain (minimal since the cart is light)
• air resistance to the spin of the propeller blade, and
• thrust from the propeller.

As long as the thrust generated by the propellor is greater than the resistance to its rotation, so it's helping more than it's hurting, the cart can travel faster than the wind. The confusion arises when people lose sight of the fact that the prop is always driven by the wheels, rather than the other way around.
posted by nicwolff at 6:42 PM on December 3, 2008

until it starts generating power

When does that start happening, in your scenario?
posted by DU at 6:46 PM on December 3, 2008

As the vehicle approaches the same speed as the wind, the energy available to it approaches zero.

That is a pretty convincing argument.

the propeller serves as a simple sail while the wind accelerates the cart from a motionless start, and is then spun ever faster by its linkage to the wheels as the cart approaches wind speed, until it starts generating power and can assist to push the cart past wind speed and into an apparent headwind.

Ya, but where is the energy for this motion coming from? As soon as power starts flowing from wheel to propeller, instead of propeller to wheel, the wheels will slow down.
posted by Chuckles at 6:55 PM on December 3, 2008

Just curious, if you throw one of those bubble levelers in the freezer, will it freeze? Seems this would be an easy way to make any surface look like it's on an angle (or opposite angle). I'm not saying this is what's occurring in the video (but I agree that the device can't move directly downwind faster than the wind without another source of power). I'm just saying if this works, I'm going to have to make some perpetual motion videos and sell my device online.

p.s. - the plane/treadmill question depends on the plane.
posted by ShadowCrash at 6:58 PM on December 3, 2008

I raced sailboats for about 5 years in SF in the 90's so I know just enough to be an idiot.

I'm going to say that the answer to this is "yes" but;

It would have to be incredibly light.

You would need either a rigid wing sail (as in airplane wing) or a soft wing sail (as in parasail).

As to the hull it would have to be either a planing boat with perfect conditions or a hydrofoil with perfect conditions. It would cost millions and dissolve in a sneeze.
This is just a hunch but I think it falls on the near side of a Perpetual Motion Machine.

Also;
potch, you broke my brain.
DU, it's been quite well documented and after an hour is the key to your conundrum.
anthill, come out of the formicary and read Bernoulli

posted by vapidave at 7:12 PM on December 3, 2008

Yeah, what the plane/treadmill thing that mythbusters did basically proved that the people arguing the question were arguing different things. The plane took off because it was moving forward. No shit it will take off then. All they've proven is that a treadmill can't keep a plane from moving forward. Or maybe that was the original question. I dunno.
posted by Eekacat at 7:16 PM on December 3, 2008

I think the cart on the treadmill is at least plausible. Rather than thinking of a cart travelling on the treadmill surface, consider the system as a treadmill turning a screw (the propeller, "threaded" through the air) via a set of gears (wheels, shaft, etc.). If the air is sufficiently viscous, then there is no real reason why the treadmill can't exert enough force on the wheels to cause the cart to move forward. It's not a perpetual motion machine; the cart is capturing energy from the treadmill.

I have a little more trouble believing that a wind-powered cart of this design could start at rest, accelerate downwind, and then exceed wind speed. I'm not going to claim it's impossible, but it seems mechanically unlikely.
posted by Galvatron at 7:21 PM on December 3, 2008

When does that start happening, in your scenario?

Depends on the angle of attack and stall speed. It'd be tricky to calculate since the apparent tailwind airspeed is falling while the prop's rotation speed is increasing, but as long as it happens before the cart stops accelerating before the wind due to drivetrain friction, and generates enough thrust to get us past wind speed, we're in business.
posted by nicwolff at 7:26 PM on December 3, 2008

I think I can do a thought experiment to show that a glider of neutral buoyancy (to take the energetics of gravity out of the equation) can in fact, theoretically move faster downwind than the average speed of the wind as long as there are occasional gusts of wind (which will be computed into the average speed of the wind, of course). You don't need any of that obfuscating garbage about propellers and treadmills either.

First, consider a game of T-ball. With a perfectly elastic collision between ball and bat does the ball leave the tee at the speed of the bat, assuming much greater mass on the part of the bat?

No. It leaves at twice the speed of the bat. To see this, consider a frame of reference at rest with respect to the bat (when it's up to speed and no longer accelerating). An observer on the bat sees the ball on the tee approaching the bat at a uniform speed, colliding with it and then, because it is an elastic collision, leaving at the same speed in the opposite direction (this is true ultimately as a consequence of conservation of momentum). Therefore, an observer on the ground will also see the ball speeding away from the bat, but since the bat is moving in that frame of reference, the ground observer will see the ball moving at the speed of the bat plus the speed of the ball the bat observer sees-- in other words, at twice the speed of the bat.

Now translate that to the glider situation. A gust of wind smacks into the back of the glider (already moving at the speed of the wind), and even though it is not a perfectly elastic collision, sends it speeding away at least somewhat faster than the speed of the gust itself, and it will maintain a speed greater than the surrounding air for a little while before friction slows it down to the speed of the air again. Therefore, the average speed of the glider will be greater than the average speed of the wind.
posted by jamjam at 7:31 PM on December 3, 2008

Ya, but where is the energy for this motion coming from? As soon as power starts flowing from wheel to propeller, instead of propeller to wheel, the wheels will slow down.

As I said: The confusion arises when people lose sight of the fact that the prop is always driven by the wheels, rather than the other way around. The cart may accelerate before the wind slightly more slowly because of the added work of spinning the prop against the resistance of the air, but there's no reason this would necessarily limit the cart's speed to the wind's speed.
posted by nicwolff at 7:37 PM on December 3, 2008

It's completely possible to build a wind power vehicle that can travel faster than the wind.

Imagine a parachute that is pushed by the wind, now also imagine that inside this parachute is a device that acts something like a crossbow with a propeller on it, so that the wind can cock the crossbow-like device. Once the crossbow device is cocked, you can fire it, the energy of which pulls the edges of the parachute together, forcing the wind out behind propelling everything forward.

You will achieve a momentary spurt of energy that would propel you faster than the current wind speed.

However, an equivalent parachute contraption without the crossbow would win any race, because its average speed would always be greater than the crossbow powered contraption, and the spurt the crossbow contraption gets would never be enough to close the gap between the two.

The fallacy is to equate instantaneous speed with average speed. On average you can never get better than the average windspeed, but you can have instantaneous moments where you are faster than the average windspeed, but you have to pay for it by being slower those other moments, and due to the laws of thermodynamics, you'll end up worse off than if you maintained an average speed of your own.
posted by forforf at 7:40 PM on December 3, 2008

Here's my analysis: put the cart on a moving treadmill and prevent it from moving backwards, so we can assume the wheels are up to speed. Since the wheels are turning, the prop is turning, which generates a thrust T pushing the cart forward, and an induced-drag torque D on the prop axle. This drag torque is transmitted to the wheel axle, giving a slowing torque of AxD, where A is some constant depending on the gearing of the transmission (and including some friction). The thrust T puts torque on the wheel axle TxR, where R is the radius of the wheels. If AxD < TxR, then the cart should accelerate. I see no a priori reason why this can't happen, if you pick the right prop radius and pitch, transmission between the prop and wheel, and wheel radius (and make everything as light as possible to limit friction). I don't know enough about propellers to know if this is actually possible.

So where is the energy to accelerate the cart coming from? This is easiest to see in the reference frame where the road is still, and the wind and cart are moving the same speed S. Think about the air passing over the propeller. It used to be moving at speed S. After the propeller sweeps by it, it is moving at a speed less than S. So the air has less kinetic energy. Where'd it go? To the cart (and the prop). The cart is extracting energy from the difference between the wind speed and the ground speed by reducing that difference.

But here's something I'm confused about: in the treadmill frame, it seems like the energy is coming from nowhere, because both the cart and the air end up moving, where neither were before. This bothers me.

Mark C. Chu-Carroll and a lot of people in this thread ignore the fact that we have a propeller that always sees wind speed in the right direction because it is turning.

All that said, the boing boing video sure looks real, but the other video it refers to sure looks fake.
posted by samw at 7:42 PM on December 3, 2008

Follow the energy, people.

Assertion A: The wheels are driving the propellor!
Assertion B: The propellor is pushing the cart forward faster than the wind (due to being driven by the wheels!)

Therefore: The propellor pushing the cart faster makes the wheels roll faster, increasing the speed of the propellor, increasing the speed of the wheels, increasing the speed of the propellor...

Perpetual motion.

As many others have said, there is no energy available to this machine from air that is at rest relative to it. If it worked, it wouldn't need any wind to begin with -- you could just give it a little shove and it would accelerate until it reached light speed.
posted by rusty at 7:43 PM on December 3, 2008 [7 favorites]

Galvatron, I've been trying to bend my head around the idea you suggest for the past few minutes, and while I think there might be something to it, I haven't gotten there yet. There is another factor, however. The treadmill belt has friction with the air, so in the neighbourhood of the belt there will be a small headwind. Can a windmill powered car move? Why not?

Hmm...

Okay, I think it can work! See, if a windmill powered car is possible, at all, then you have your answer. And, it is not a perpetual motion machine. Air molecules are being slowed by the propeller, and that is the source of energy.
posted by Chuckles at 7:45 PM on December 3, 2008

As I said: The confusion arises when people lose sight of the fact that the prop is always driven by the wheels

So, why doesn't your magical cart accelerate forever, eventually reaching relativistic speeds?
posted by 5MeoCMP at 7:47 PM on December 3, 2008

On preview, what rusty just said.
posted by 5MeoCMP at 7:47 PM on December 3, 2008

No! Not preview. If it were preview, it would have been in the original comment!
posted by 5MeoCMP at 7:48 PM on December 3, 2008 [1 favorite]

(shut up, brain)
posted by 5MeoCMP at 7:48 PM on December 3, 2008 [1 favorite]

I always make sure I travel downwind faster than my wind.


*ahem* I can't quite get my head around the specific examples at this hour. But surely if you had any machine that could go faster than the wind powering it, you'd have an infinite energy generator?

You could set a whole bunch of them trundling around the equator of a small planet and harvest the hurricane speed winds they eventually build up.

Maybe I don't understand what "DDFTTW" is supposed to mean.
posted by lucidium at 7:49 PM on December 3, 2008

The only way to truly settle this is mandatory high school Physics instruction. I'll never understand why people think they can reason their way to an answer. This isn't Philosophy, folks, it's Physics. There is a single right answer, and you find it with science.

If I could favourite a 100x, I would.
posted by five fresh fish at 7:51 PM on December 3, 2008

I'm not being snarky here jamjam but, to quote;

A gust of wind smacks (bold mine) into the back of the glider (already moving at the speed of the wind), and even though it is not a perfectly elastic collision, sends it speeding away at least somewhat faster than the speed of the gust itself, and it will maintain a speed greater than the surrounding air for a little while before friction slows it down to the speed of the air again. Therefore, the average speed of the glider will be greater than the average speed of the wind.

I think that after the yielded energy of the gust passed, the glider would indeed travel faster than the now surrounding wind but I think that as the wind "smacked" into the back of the glider only part of the energy would be imparted. Mostly the wind would pass by the glider and as such still be ahead of the glider. Hence not faster than downwind.

Also and unrelated to mine above I heartily agree with the first sentence of what Eekacat said.
posted by vapidave at 7:53 PM on December 3, 2008

ShadowCrash writes "Just curious, if you throw one of those bubble levelers in the freezer, will it freeze?"

You'd need a pretty good freezer because levels are filled with alcohol, hence the term spirit level, which has a freezing point around -115. Some liquid nitrogen would do the trick.
posted by Mitheral at 7:55 PM on December 3, 2008

Holy fuc,k where's the flag for "terrible physics" when you need it?
posted by Shutter at 7:57 PM on December 3, 2008 [2 favorites]

samw: in the treadmill frame, the energy would come from the treadmill--because it is encountering resistance from the wheels of the cart, the motor has to do more work than it would otherwise.

rusty: the speed differential between air and ground is the fundamental source of energy. Regardless of the speed it is travelling, an ideal cart could potentially harvest energy by reducing the speed of the air relative to the ground. It's not perpetual motion--there is a large but limited pool of energy available.

This may all be a hoax, and the mechanical design may not work, but I don't think there's a fundamental physical principle that prevents this sort of design from working.
posted by Galvatron at 7:58 PM on December 3, 2008 [1 favorite]

Hurts so bad I'm losing commas in the middle of expletives.
posted by Shutter at 7:58 PM on December 3, 2008

lucidium,
the power of the wind comes from the sun.
posted by vapidave at 7:58 PM on December 3, 2008

The only way to truly settle this is mandatory high school Physics instruction. I'll never understand why people think they can reason their way to an answer. This isn't Philosophy, folks, it's Physics. There is a single right answer, and you find it with science.

So, what is the answer? Explain it to me like I'm a high schooler.
posted by R_Nebblesworth at 8:01 PM on December 3, 2008

The sum of energy input to the wind-powered device can not be more than the energy of the wind.

Otherwise it's not a wind-powered device, but a dual-powered device, i.e. only partly wind-powered.

I expect the wind is nearly 100% efficient: almost all its energy is in motion. It has little to no internal resistance, and the only external resistance is the wind-powered device. Except for the influence of the device, it makes little to no heat, little to no noise, etcetera.

It is highly unlikely the device can extract energy from the wind and convert that to forward motion in a manner that exceeds wind's own ability to move that energy forward. Indeed, I can't imagine how it could. Magic, maybe.

Or DU's suggestion that it uses wind to store energy, and then moves downwind in a different wind. Decouple the energy extraction/storage from the energy expenditure. Which doesn't make sense for the scenario.
posted by five fresh fish at 8:02 PM on December 3, 2008

That looks like reasoning to me. How is your explanation different (as in "reasoning vs. science") from what others have offered? I am really trying to understand what's going on here, sorry if this comes off as snarky/dickish.
posted by R_Nebblesworth at 8:06 PM on December 3, 2008

rusty: the speed differential between air and ground is the fundamental source of energy.

?! Only if your wind->(power source) convertor is stuck in the ground. Which, unless you're using an electric rail system, isn't going to work so well at outpacing the wind. (Transmission losses prevent the electric rail idea from going very far.)
posted by five fresh fish at 8:07 PM on December 3, 2008

Again it's possible to go faster than the wind, but still be completely within the laws of thermodynamics. All it requires is a mechanism to change the kinetic energy of the wind into stored potential energy. That's it ... no magic, no perpetual motion. It's just releasing stored energy faster than it was accumulated. Like dumping all the coins out of your piggy bank at once. Hey look! I made $100 in 6 seconds, or $60,000 an hour and I only had to save at $0.10 per hour!
See the flaw? I had to save $0.10 per hour for 1000 hours just to get 6 seconds of fun. It's the same idea with energy (except with energy, its like some of your money spontaneously turned into heat).
posted by forforf at 8:08 PM on December 3, 2008

Galvatron: Ah, of course! Thanks!

FFF and others: the air does not have to be moving relative to the cart to have energy. Even if the air and the cart are moving the same speed, all the cart needs to do is extract some of the kinetic energy of the air. You're assuming that the cart cannot do this if it is moving the same speed as the air. Why do you think this is the case?
posted by samw at 8:12 PM on December 3, 2008

DU, the device needs to move at a faster than wind speed indefinitely, whereas your solution would only provide for bursts of faster than windspeed travel.
posted by spacediver at 8:13 PM on December 3, 2008

forfor, see above.
posted by spacediver at 8:14 PM on December 3, 2008

No. It leaves at a speed dependent on the amount of energy transmitted from the bat into the ball, and the mass of the ball; and the loss of energy to inefficient transfer of energy. "Twice" doesn't enter into it.
posted by five fresh fish at 8:16 PM on December 3, 2008

Galvatron: the speed differential between air and ground is the fundamental source of energy

Sorry, no. The source of energy (to power the cart) comes from the speed differential between the wind and the cart.

Consider this in a slightly thicker fluid: water.

Remove the consideration of wheels, they're irrelevant for this thought experiment.

Let's postulate a submarine with a "magic" propulsion source that doesn't affect the water surrounding it. Could that submarine, powered only by a mounted propeller (driven by the flow of water) move faster than a water current in which it's floating? Once the submarine itself and its attached propeller are moving at the same speed as the water, there's no "flow" past the propeller to cause it to turn.

In fact, once it starts moving faster than the water, the propeller turns *backwards*. Perhaps if you have a gearbox (again, 100% efficient), you could turn this into energy generation -- but at that point, you start reaching relativistic speeds again because it's a perpetual motion machine.
posted by 5MeoCMP at 8:16 PM on December 3, 2008

samw: consider frames of reference. A cart travelling at exactly the same speed as the surrounding wind is equivalent to a cart sitting still with no wind; the surrounding air is moving at exactly the same speed relative to the cart.

How can a stationary cart extract energy from stationary air?

Now, if you were proposing some exotic heat transfer/vacuum point energy device that would extract *heat* from the surrounding air, then that'd be one thing -- but you're not, you're suggesting that just because the surrounding air is in motion *relative to a stationary observer*, that the cart should be somehow able to take the observer's frame of reference into account.

I wish there were a tax on pseudoscience.
posted by 5MeoCMP at 8:21 PM on December 3, 2008

The fallacy is to equate instantaneous speed with average speed. On average you can never get better than the average windspeed, but you can have instantaneous moments where you are faster than the average windspeed, but you have to pay for it by being slower those other moments, and due to the laws of thermodynamics, you'll end up worse off than if you maintained an average speed of your own.

Nailed it. Nicely worded.

So this is dependent on whether one is measuring instantaneous speed or average speed. I assumed average speed, as that is the obviously more interesting challenge. A vehicle that can move faster for an instant, but ulitmately delivers you to your destination slower, is a rather useless vehicle indeed.

They should have used the word "device."
posted by five fresh fish at 8:22 PM on December 3, 2008

The sum of energy input to the wind-powered device can not be more than the energy of the wind.

Otherwise it's not a wind-powered device, but a dual-powered device, i.e. only partly wind-powered.

Lots of Hmmmm in this one.

Don't windmills actually generate energy? Or produce more than they consume?
Otherwise why have them?

Hmmmm.

Sails harvest energy. Boats harvest wind energy. It's free and has nothing to do with the mass of the boat. I think what this is going to boil down to is can the boat harvest enough energy to generate a velocity in excess of it's surrounding windspeed. (I just restated the question) I'm still stuck on "yes".
posted by vapidave at 8:24 PM on December 3, 2008

5MeoCMP, your submarine example is not applicable. The cart has wheels which are coupled to the ground, and a propeller which is (loosely) coupled to the air, and is therefore able to change the speed of the air relative to the ground. Your submarine is not coupled to the earth/seabed.
posted by Galvatron at 8:27 PM on December 3, 2008

vapidave: windmills don't really consume energy to operate; it's more like they lose some of the generated energy in conversion (and a teeny amount perhaps in running their own diagnostic/monitoring systems)

Windmills take energy out of the wind and turn it into electricity (minus losses to heat/noise). The wind that's gone past a windmill is a little slower, though since there's so much of it (and it's powered by the sun), we really wouldn't notice.
posted by 5MeoCMP at 8:27 PM on December 3, 2008

vapidave, oh, I know. I was just trying to point out that if this hypothetical machine worked without slowing down the wind it's getting its energy from, it would have to give some energy back to the air it's in, speeding up the wind. So even if you closed the system after the initial kick start to give them some wind, they'd keep going. It wasn't a very good example.

If they're slowing down the wind to go faster than it then you could do it by putting a big sheet of cardboard in the way of a wind tunnel. It would slow down all the air and, for a little while, be going faster than the "wind".

But as I understand it the suggestion behind this thing seems to be a machine that, on an infinite plain with a permanent wind blowing across it, would continually be able to travel faster than the wind. Which would be impossible.
posted by lucidium at 8:29 PM on December 3, 2008

(shut up, brain) posted by 5MeoCMP

Eponomystericalist. Or however it's spelled. I hate it when people point them out anyway.

the air does not have to be moving relative to the cart to have energy. Even if the air and the cart are moving the same speed...

...then it's dead air: it's moving the same speed as you, just like it would be if you were both enclosed in a train car. The best you can do is extract heat energy from it. Which isn't really extracting energy from the wind, per se.
posted by five fresh fish at 8:30 PM on December 3, 2008

Galvatron: I was trying to simplify the experiment, but that's not really relevant.

Exactly what does happen to the propeller when the cart starts travelling faster than the wind, and now experiences a headwind instead of a tailwind?
posted by 5MeoCMP at 8:31 PM on December 3, 2008

5MeoCMP: I admit that I'm a little confused about where the energy comes from in the treadmill frame. Galvatron's explanation seemed right at first, but I'm not so sure now. But conservation of energy is a consequence of other laws. Is there a mistake in my analysis of the forces involved? If not, the energy must be coming form somewhere, even if we haven't figured out where.

Also, in the ground's frame, it's clear where the energy is coming from: the air is moving slower than it was, so some energy was transferred from the air to the cart.

So if you reply to one thing, reply to this: where is the mistake in my accounting of the forces involved? I'm willing to be wrong. It wouldn't be the first time.
posted by samw at 8:32 PM on December 3, 2008

Don't windmills actually generate energy? Or produce more than they consume? Otherwise why have them?

They convert energy. They convert wind energy (force of air movement against blades) into electrical energy (or, in the old days, rotational energy, which was then used to do drive machinery.)
posted by five fresh fish at 8:33 PM on December 3, 2008

If the wheels are driving the prop, then the prop can not be be moving the cart forward, because that would require it to be driving the wheels. The wheels can not contribute energy to the prop at the same time the prop is contributing energy to the wheels.
posted by five fresh fish at 8:36 PM on December 3, 2008 [1 favorite]

Mkay. Brain hurts.

Night night.
posted by Samizdata at 8:39 PM on December 3, 2008

No, this can't possibly work, and here's why: view the whole system from the perspective of the moving air. (The principle of Newtonian relativity means all out physical laws are equally valid in this frame.) There are exactly 4 external forces acting on the cart: gravity, the normal and tangent contact forces between the wheels and the ground, and the drag force between the cart and the air. Since the gravitational and normal forces act perpendicular to the level ground, we can ignore them.

By moving at a positive velocity relative to the air, the cart is continually doing work against the drag force to push the air forward. The energy can't come from the ground; at best the contact is frictionless, and at worst it will cause the cart to slow down through rolling friction. And it can't come from the air, because in this reference frame the air has zero kinetic energy. So there's nowhere it can come from (besides an onboard power source) and the system as a whole violates conservation of energy.
posted by teraflop at 8:48 PM on December 3, 2008 [1 favorite]

fff: No one is claiming the prop is driving the wheels through the transmission. It is driving the wheels by pushing the cart forward. Since the wheels are on the ground, this motion makes the wheels turn (faster than they were already). Just like in an airplane: the prop doesn't make the wheels turn directly, the prop pulls the plane forward. The wheels turn because the plane is moving and the wheels are on the ground.

Again: did I make a mistake in my first post in my accounting of the forces involved?
posted by samw at 8:50 PM on December 3, 2008

I don't know, couldn't you just have the device remain stationary gathering energy from wind for, say, 1 hour, charging a battery, (or just a blamed rubber band), and then get perhaps 5 minutes of faster-than-the-wind travel out of it? I know I'm cheating the rules a little, but that seems possible to me.

I always liked those roadrunner cartoons where the coyote stood on a boat, blew into the sail of the boat, and quickly achieved speeds of what appeared to be 40+ mph. On land!
posted by SomeOneElse at 8:55 PM on December 3, 2008

teraflop: you've ignored two forces: the thrust generated by the prop and induced drag on the prop. These are important!
posted by samw at 8:56 PM on December 3, 2008

Brother Dysk: "potch: This is different from the airplane question, because the wheels are actually propelling the vehicle

Surely not, airplanes are propelled by jets or propellers, not the wheels...?"

That's what I'm saying- unlike airplanes, wheels are what actually move this forward.
posted by potch at 9:05 PM on December 3, 2008

samw: Sorry, that was badly phrased. Where I said "drag", read "any interaction between the cart and the air". The point stands that those forces cannot transfer energy to the cart, since in this frame the air has zero velocity.
posted by teraflop at 9:05 PM on December 3, 2008

It took me a bit to realize why, but it works, and here's how:

Nobody disputes that by sailing crosswind, a sailboat can sail faster than the wind. This is because the wind is hitting the sail at an angle.

The propeller blades are angled. The wind isn't flowing over the blades straight on, it's flowing over them at an angle, just like the sailboat sailing crosswind. The extra power is transferred to the wheels, et voila.
posted by smoothvirus at 9:10 PM on December 3, 2008 [1 favorite]

I'm not being snarky here jamjam but, to quote;

A gust of wind smacks (bold mine) into the back of the glider (already moving at the speed of the wind), and even though it is not a perfectly elastic collision, sends it speeding away at least somewhat faster than the speed of the gust itself, and it will maintain a speed greater than the surrounding air for a little while before friction slows it down to the speed of the air again. Therefore, the average speed of the glider will be greater than the average speed of the wind.

I think that after the yielded energy of the gust passed, the glider would indeed travel faster than the now surrounding wind but I think that as the wind "smacked" into the back of the glider only part of the energy would be imparted. Mostly the wind would pass by the glider and as such still be ahead of the glider. Hence not faster than downwind.


No offense taken, vapidave, that is an excellent criticism!

However, I think it can be easily met as long as the glider reaches the speed of the gust rapidly compared to the duration of the gust, and the glider slows down more slowly than it sped up-- both conditions which ought to be readily achieved by a low-mass, properly aerodynamic glider.
posted by jamjam at 9:15 PM on December 3, 2008

For me it goes to the moment the cart and the wind are going at the same speed. This is the cheat buried in the video- as the cart is moving at "wind speed" on the belt the wind is still present. This would not happen in a real world setting. When the cart is going at wind speed there is no wind to power anything- the air will appear still on the cart. When we sail crosswind the apparent wind speed remains the same.
posted by pointilist at 9:23 PM on December 3, 2008

Where I said "drag", read "any interaction between the cart and the air". The point stands that those forces cannot transfer energy to the cart, since in this frame the air has zero velocity.

In this reference frame, the air does not have energy, but the ground does. If the air exerts a force on the cart in a direction opposite the motion of the ground, and the cart then passes a percentage of that force on to the ground, then the ground decelerates but the cart accelerates. Some of the kinetic energy of the ground is converted to kinetic energy of the cart (including rotational KE of the propeller/shaft).
posted by Galvatron at 9:23 PM on December 3, 2008

(I didn't mean to start a reason vs. science derail. Science is reason plus experimentation, with rules to help get rid of biases. The experimentation is vital because it grounds the reasoning in reality. Without that you just have philosophy, which quickly leads you astray.)

As penance for my bad behavior, here's a page that does a nice job describing the aerodynamic force that allows a sailboat to travel faster than the wind (but only at an angle to it):

http://www.physclips.unsw.edu.au/jw/sailing.html

As for the question at hand: I don't know what the right answer is, but I suspect smoothvirus has hit upon it. Here's a gedanken experiment that may help: In the outdoor cart experiment, what happens to the cart when the wind is from the side? From 45 degrees to the rear?
posted by sdodd at 9:28 PM on December 3, 2008

teraflop: see this comment I made above. In the treadmill frame, I'm not quite sure where the energy is coming from. But in the still-ground frame, I explained where the energy comes from. Also, here I analyze the forces involved. Did I miss something there? (Note: I did miss air resistance from the apparent headwind on the cart once it accelerates past wind speed, but this can presumably be made small (relative to the thrust provided by the prop) by making the frame of the cart very spindly!)
posted by samw at 9:28 PM on December 3, 2008

The make discussion has some interesting stuff..
posted by Chuckles at 9:31 PM on December 3, 2008

Some of the kinetic energy of the ground

The ground is moving?
posted by five fresh fish at 9:38 PM on December 3, 2008

Galvatron: Forces come in equal and opposite pairs. An interaction that pushes forward to accelerate the cart must also push backwards to accelerate the ground as well. Now you're just violating conservation of momentum instead.

samw: No, you showed that "[you] see no a priori reason why this can't happen". In fact, violation of conservation of energy happens in either reference frame. The key is that kinetic energy = 0.5*pv (p = momentum, v = velocity). If the momentum lost by the air as it passes over the propellor equals the momentum gained by the cart (as it must) then the cart, which is moving faster, has gained more kinetic energy than the wind has lost.

five fresh fish: In the reference frame of the air, the ground is moving backwards.
posted by teraflop at 9:39 PM on December 3, 2008

You guys keep thinking the "propeller" is providing some kind of thrust. This is not correct. It is not possible for the propeller to provide any kind of thrust in this situation. In this case the propeller is acting as a windmill, transferring the energy from the wind to the wheels.

As for what would happen to the cart if the wind was coming in at an angle - the propeller would spin slower and the cart would slow down. However, I expect there would be some weathervaning and the cart would tend to turn so it was moving downwind.
posted by smoothvirus at 9:42 PM on December 3, 2008

Forces come in equal and opposite pairs. An interaction that pushes forward to accelerate the cart must also push backwards to accelerate the ground as well.

The thrust of the propeller pushes "back" on the air, and the air exerts an equal and opposite force pushing the cart "forward". The wheels of the cart push "forward" on the ground (due to the rotational drag of the propeller), and the ground exerts an equal and opposite force pushing the cart "backward". The ground accelerates "forward", in the direction of the cart... but the velocity of the ground is in the opposite direction, so it it losing energy.
posted by Galvatron at 9:52 PM on December 3, 2008

... so it is losing energy. (The ground, that is.)
posted by Galvatron at 9:55 PM on December 3, 2008

A propeller/windmill driven boat.
posted by Chuckles at 9:59 PM on December 3, 2008

Warning: The Venturi effect has been deprecated as an acceptable theory for lift from airfoils in flight. (It's not Bernoulli's principle that is being used in airfoils, anyway - it's the Venturi effect, and the Venturi effect only works with pairs of "airfoils", IE, as in a Venturi nozzle.)

See these fine NASA links:

Wrong.
Right.

Also see this wikipedia entry on airfoils. The words Venturi and Bernoulli are not found.
posted by loquacious at 9:59 PM on December 3, 2008

teraflop: Once again, I admit that I'm not completely sure where the energy is coming from in the treadmill frame (it's coming from the treadmill, but I'm not sure I understand the mechanism of energy transfer). But in the still-ground frame, the cart can extract kinetic energy from the air by slowing the air via the propeller.
posted by samw at 9:59 PM on December 3, 2008

Yeah, I think I've talked myself out of this. When the cart is at wind speed, the air is motionless relative to the cart. But the cart's wheels are turning, so the prop is spinning, and a prop spinning in still air produces thrust. Could that thrust be greater than the resistance that driving the prop adds to turning the wheels? If so, then the cart will accelerate towards the speed at which that no longer holds true, even if that speed is greater than the wind speed. But I don't have any way to show that that's possible, so I'm gonna chalk this up to bad late-night physics and get back to work...
posted by nicwolff at 10:11 PM on December 3, 2008

Also, I do not know if the following exercise is accurate or not - I await more rigorous experimentation - but it's lifted from the BoingBoing thread:

OK, I think I know how it works. The "props" are actually sails. They don't blow wind backwards. The wheels rotate the sails. Now...

We know that a sailboat can travel faster than the wind, but only at an angle to the wind. If you wanted to travel faster than the wind in the direction of the wind, the best you can do is tack back and forth.

With this vehicle, the vehicle doesn't need to tack back and forth, because the rotating sails do the equivalent of tacking. This allows the vehicle itself to travel in a straight line.


This is how I'm visualizing the problem. Visualize the path of the "windmill" (not a "propeller") as helical in nature. The airfoil blades of the windmill are rigid sail foils "tacking" around the directly downwind vector of the cart. More, these orbiting rigid "sails" don't have to actually change direction when the vehicle changes direction to continue tacking the other way.

Unless I'm totally mistaken, it is indeed possible to have a total average forward speed than the true downwind speed through careful tacking - especially on fast, lightweight boats than plane and tack well - and it's apparently even easier to do on land or ice.


But by a strict definition of "downwind", perhaps a windmill isn't actually traveling directly downwind faster than the wind at all - and we're all merely quibbling over the ill-defined semantics of what "directly downwind" even means in the first place.

Yeah, that sounds about right. This isn't a physics hack - it's actually a language hack.

Beer me.
posted by loquacious at 10:11 PM on December 3, 2008

Galvatron: IMO, considering the cart to be accelerating is a red herring. Think about what happens when the cart reaches its equilibrium speed. The cart and ground would both have a constant velocity, while the air would be continually gaining energy from -- where?

samw: No it can't. As I said, transferring momentum from the air to the cart requires a positive energy input from somewhere. The air doesn't lose as much energy as the cart needs to gain to make the momentum balance.
posted by teraflop at 10:17 PM on December 3, 2008

teraflop: Why is the air gaining energy? At equilibrium, from the reference frame of the ground, I expect that the air is losing energy. Of course, assuming there's a whole lot of air, the global average velocity of the air doesn't decrease much.

Your conservation of momentum argument is the most compelling refutation I've seen so far, but I'm not totally convinced that your application of the principle takes into account all the elements of the system. I'll have to think about it some more.
posted by Galvatron at 10:27 PM on December 3, 2008

(It's not Bernoulli's principle that is being used in airfoils, anyway - it's the Venturi effect, and the Venturi effect only works with pairs of "airfoils", IE, as in a Venturi nozzle.)

Err, a refinement: Bernoulli's principle is used in designing the foils themselves - but this has nothing to do with generating lift through lower pressures on top of the wing and more to do with reducing turbulence in the laminar flow leading to a more effective Coandă effect which gives more efficient flow turning - which all together gives us the forces involved with wings and airfoils.

I am not an aeronautical engineer, but I used to design my own wing chords for free-flight gliders - and a really good free-flight glider is one that vanishes into the sky on a rising thermal.
posted by loquacious at 10:30 PM on December 3, 2008

Galvatron, your example was relative to the air, not the ground. (I'm not sure why the apparent energy flow changes with your reference frame, but I suspect it has something to do with the fact that the atmosphere appears to have an effectively infinite supply of kinetic energy, which makes reasoning about it hard.)
posted by teraflop at 10:55 PM on December 3, 2008

teraflop: Now I see what was bothering me. Momentum is not conserved in the collision between air and cart, because the system is not closed. The ground exerts an external force on the cart wheels, in a direction opposite the motion of the cart. So you have to look at the three-way collision between air, cart, and ground, and I don't think the outcome of the collision can be trivially determined.
posted by Galvatron at 10:58 PM on December 3, 2008

You lost me, teraflop. In the reference frame of the air, the air always has zero kinetic energy. How could it be otherwise? In that reference frame, the energy exchange is all between the cart and ground.
posted by Galvatron at 11:03 PM on December 3, 2008

The air starts with zero energy, but gains energy and momentum as it's pushed backwards by the propeller. What I don't understand is how there can be any energy exchange between the cart and the ground; at equilibrium, neither of them is changing velocity, no matter what frame you look at.
posted by teraflop at 11:23 PM on December 3, 2008

How about if we just let the wind push the vehicle up to the speed of the wind, and then we kick it really hard?
posted by davejay at 11:52 PM on December 3, 2008

Crap, I think I was reasoning about a non-inertial reference frame. I believe I follow you now.

However, I disagree regarding energy exchange at equilibrium. There is never really any equilibrium, because energy is always draining from the system; there is never a point of balance. But at the point of the cart's maximum velocity, the ground would have to be slowing relative to the air (due to the force applied to the ground by the wheels of the cart). The ground would continue to slow over time, and the cart would slow and ultimately come to rest as the energy is drained from the system.
posted by Galvatron at 12:01 AM on December 4, 2008

Imagine the cart moving forward at the speed of the wind. For this to happen the wheels need to be frictionless and have no resistance. If we then want to harvest energy from the motion of the wheels, we have to add resistance, which means the cart slows down. If we use the energy harvested to propel the cart, via a propeller or via the wheels or whatever, we can make up for some of the slow down, but unless the system is >100% efficient, it can never replace the kinetic energy lost from harvesting it in the first place. So for the best *net* speed, you'd be better off not harvesting.

(Of course there's nothing stopping you storing the energy up and replaying it in one burst and therefore momentarily exceeding wind speed)
posted by cillit bang at 12:35 AM on December 4, 2008 [2 favorites]

I think I made a big mistake in my first post. I should be comparing the thrust power produced by the prop and the drag power required to turn the prop. Doing this makes that problematic R term go away when airspeed is 0. Then the inequality
drag power < radius x thrust power
can't be satisfied, so the cart shouldn't accelerate past wind speed.

I am, as ever, willing to be proven wrong. I'd be interested to see the person in the boing boing video do the experiment with a level treadmill, and showing us the rest of the room. They seem to eliminate most of the obvious ways to fake it.
posted by samw at 1:28 AM on December 4, 2008

Why do I have this terrible feeling of deja-vu?
posted by pharm at 1:44 AM on December 4, 2008

These chaps spent the summer trying to beat the wind powered land speed record. Here is their simple expanation of why they think you can go faster than the wind you get your power from. They reckon they can do between 3 and 5 times the true wind speed.
posted by biffa at 4:23 AM on December 4, 2008

why they think you can go faster than the wind you get your power from

That's because you can make your craft have a lot of drag in one direction (to pick up energy) but very little in another (to make moving forward easy). If the wind is in the same direction as you're trying to move, it doesn't work.
posted by cillit bang at 5:36 AM on December 4, 2008

The video is intentionally misleading. It's telling you that moving uphill on a treadmill is the same as increasing your speed in real life. It's not. When you run on a treadmill, it feels as if you're exerting the same energy to run outside, so you believe this. However, if you were to hold stand on the treadmill with roller skates on, would you still think hanging on to stop yourself from falling off is the same as actually roller skating the same distance?

There is an outside power source, namely the force exerted downward by his hand and the forces imparted by the treadmill. Consider the following possibility:

I build a machine with two sets of wheels, connected by a gear and suspended at different heights with a some sort of spring mechanism. If I push down on the device, the smaller wheels make contact with the ground, if I let go the larger wheels make contact. I now place it on the treadmill, and push down on it. The small wheels spin and, due to my gearing, so does the larger wheels. Because of circumferences of the wheels, the outside edge of the larger wheel is now spinning much faster then then outside edge of the smaller wheel. I now let go of the device, and the springs push the larger wheel into contact with the treadmill. If we assume the larger wheel has significantly more mass than the rest of the device, the device will accelerate up the treadmill.

However, there's no gain in total energy, I added energy by creating a downward force to build up the kinetic energy in the large wheels.

Here's another similar example: Pick up a kids toy car, the kind that you can pull backwards to make it drive forward. You can pull the car back 6 inches, and the car will travel a few feet forward. The difference is that you used much more force for the 6 inches you pulled the car back then it used going forward for 6 inches, so it had enough energy to continue moving forward (the car also goes much faster forward then you pulled it back).

Now, let's add the prop and see how that would work. Let's say I build a similar device as the one in the video, but instead use a giant prop. If I hold the device down enough to cause the wheels to spin the prop, it's a simple gearing problem. If it's a one to one gear ratio, the prop is rotating once for every tire roation. If the prop is big enough, it's going to try and drive the car forward and eventually I'm actually hold the car back. If we release the car, the prop will try and push the car forward. The wheels would try and retard this, but that's easy enough to overcome. All we need is a way to disengage the wheel from the prop, so the wheel can spin freely.

So what I'm actually doing is using the treadmill to power up my prop, and then letting the prop drive the device forward. The only force the prop needs to overcome is the resistance required to push the device forward on the treadmill (think of the force it would require for you to push the the device and prevent it from rolling backwards on the treadmill). In the video, there's also a small gravitational force to overcome, since the device is pointed up hill. In my example, if the device rapidly drove up the hill, I don't think you'd be very surprised.

So now I create a video, mislead you into thinking this is the same as going faster than wind, and hide the fact that the prop and wheels can disengage. The tricky part is not having the device climb up the treadmill, but having it climb slowly enough to make you think it's creating a very small gain.

If we created a device that could go directly downwind faster than the wind itself, the device would in actually end up moving through the air faster than the air is flowing, which would be equivalent to driving directly into a headwind. If you could use a headwind to power yourself forward into it, you'd create an even bigger head wind, which would give you even more power, etc... etc... Eventually, you'd exceed the speed of light, the universe would explode, and we wouldn't be able to have this discussion.
posted by ShadowCrash at 6:06 AM on December 4, 2008 [2 favorites]

biffa, the link you pointed to isn't going directly downwind, they're going cross wind. It's well established in the sailing community that you can go faster than the speed of the wind in this case. However, you're actually going faster than the wind appears to a stationary point, not the speed of the wind passing over your boat.
posted by ShadowCrash at 6:09 AM on December 4, 2008

Btw, the airplane/conveyor belt question is also misleading. The actual question should be:
Is it possible to build a treadmill capable of keeping an airplane stationary, assuming the plane has free spinning wheels and can generate thrust throw a propeller/jet?
posted by ShadowCrash at 6:12 AM on December 4, 2008

through, not throw.
posted by ShadowCrash at 6:12 AM on December 4, 2008

Since I won't read BB anymore, I have no idea what this is truly about, but I'm going to assume steampunk dildonics with drm.
posted by waraw at 6:42 AM on December 4, 2008 [1 favorite]

A hamster is in his hamsterball traveling at 2 miles per hour on a treadmill with a 2% incline and a 1.5 MPH wind at his back. The hamsterball is frictionless, with normal gravity, but (obviously) there's a perfect vacuum outside. It's an opaque hamsterball that is completely mirrored on the inside with a single light bulb. If the hamster suddenly accelerates to the speed of light in synchronicity with the treadmill, how much faster is the light bouncing around the mirrored ball inside?
posted by yeti at 7:02 AM on December 4, 2008

Btw, the airplane/conveyor belt question is also misleading. The actual question should be:
Is it possible to build a treadmill capable of keeping an airplane stationary, assuming the plane has free spinning wheels and can generate thrust throw a propeller/jet?

As opposed to all the fixed-wheel, non propeller or jet planes?
posted by signal at 7:22 AM on December 4, 2008

As opposed to taking it as a given that a treadmill can exert enough force on a plane to keep it from moving forward. I think a lot of people overlook the fact that the plane has wheels that spin freely, so they have a hard time understanding the problem. If you directly reference the free spinning wheels and thrust, I think it's easier for people to understand keeping a plane still on a treadmill isn't happening.
posted by ShadowCrash at 7:30 AM on December 4, 2008

I call shenanigans. I say it can't be done and I paraphrase my illustrious ancestor in saying:
"The laws of motion aren't just good ideas. They're LAWS."
posted by DaddyNewt at 7:48 AM on December 4, 2008

I think it's easier for people to understand keeping a plane still on a treadmill isn't happening.

The kind of people who have a hard time understanding that a plane's wheels spin freely and that its thrust is generated by jets or propellers (not the wheels) are probably not going to understand the problem even if you hit them over the head with a mid-range Cesna.
posted by signal at 7:58 AM on December 4, 2008

Since most of the "yes" camp seem to be saying that the wheels are powering the propeller, not vice versa, may I just point out that that would fail the "being wind-powered" test, if you were being really nit-picky?
posted by Brother Dysk at 8:19 AM on December 4, 2008

Brother Dysk, the wheels have to be powering the propeller, since there's no fan (at least not one shown) and the propeller isn't moving until the car's on the treadmill. You're right, this isn't the same as a wind powered device moving faster than the wind powering it, no matter what the creators of the video claim.
posted by ShadowCrash at 9:07 AM on December 4, 2008

Read up on sailing and you will learn that energy comes from the difference between velocity of the water and velocity of the air. As long as there is a difference, it is possible to extract energy. All of this applies equally to the car -- if there is a speed difference between the ground and the air, there is energy to be extracted.

The question is then one of implementation, not fundamental physics.
posted by Chuckles at 9:26 AM on December 4, 2008

This is very simple.

No, an airplane cannot take off, because its not jet thrust that makes the airplane fly, its lift generated by air passing over the wings, which itself is generated by forward motion. No forward motion (like if you are on a treadmill) means no flight. A similar analogy would be sticking your hand out of a moving car vs. a car on a treadmill. In the former, you feel a force on your hand, caused by air hitting your hand as the car pulls you forwards through the air. In the latter, you feel nothing on your hand because the car is stationary, hence no forward motion, hence no force.

No, you cannot go faster than the wind and at the same time be powered by the wind. If someone is pushing you, can you go faster than that person? No, because if you did, you would separate from him, lose momentum transfer (because you are now two bodies with two separate masses, rather than one integral unit) and slow down.

In conservation of momentum terms, m1v1 + m2v2 = (m1+m2)v3, where m1 = mass of vehicle, v1 is velocity of same, and m2 is the mass of air pushing the vehicle, and v2 is velocity of the same. If v1<v2 (meaning the cart is initially travelling slower than the wind), then v3 (the final velocity of the unit together) can never be greater than v2.

Consider this myth busted.
p.s. can somebody explain how a crosswind and angled sail create a force-multiplier?
posted by wayofthedodo at 9:28 AM on December 4, 2008

Another way to look at it: can you move faster than your car, if you're sitting on the hood?
posted by wayofthedodo at 9:29 AM on December 4, 2008

wayofthedodo, you're wrong about the airplane. If you could invent a treadmill that could keep the plane stationary, yes, you'd be correct. However, you'd need to violate the laws of physics to create a treadmill that does that (you'd need a treadmill that can accelerate to infinite). So the plane will move forward on the treadmill, air passes over the wings, creates lift, and the plan takes off.

Your answer is exactly why I said if they rephrase the question, it would be easier for people to understand.
posted by ShadowCrash at 9:45 AM on December 4, 2008

Ah, I see where I made my error. Really the treadmill is irrelevant, since the wheels are freely moving and hence no energy is transferred from the treadmill to the plane itself. Well played, Shadow, well played.
posted by wayofthedodo at 9:55 AM on December 4, 2008

wayofthedodo - I'm afraid, you're wrong. The plane can move forward on the treadmill because its wheels are spinning freely and the propeller will provide forward thrust. Consider this analogy, you're on a treadmill wearing a pair of rollerskates and you're holding a rope attached to the wall in front of you. If you pull on the rope, you'll move forward. Similarly, the plane's wheels are like the skates, and the propeller "pulls" the plane forward just like if you pull on the rope.
posted by TungstenChef at 9:57 AM on December 4, 2008

Heh, should've previewed.
posted by TungstenChef at 9:57 AM on December 4, 2008

wayofthedodo: "No, you cannot go faster than the wind and at the same time be powered by the wind. If someone is pushing you, can you go faster than that person? No, because if you did, you would separate from him, lose momentum transfer (because you are now two bodies with two separate masses, rather than one integral unit) and slow down."

Except that collision is not the proposed mechanism for the air transferring energy to the cart. Lift over the prop blade is the mechanism. There's always plenty of airspeed over the prop blade (because the prop is spinning), even if the cart is moving faster than the wind.

Brother Dysk: "Since most of the "yes" camp seem to be saying that the wheels are powering the propeller, not vice versa, may I just point out that that would fail the "being wind-powered" test, if you were being really nit-picky?"

Sure, in the treadmill frame, the energy must (assuming the thing works at all) be coming from the treadmill. But the the ground frame, the energy is coming from the wind (because after the interaction of the prop and air, the air is moving slower).

Something tricky about the treadmill frame is this: suppose there is some mechanism to transmit energy from the treadmill to the cart. Would you see a difference in kinetic energy of the treadmill? No, because the motor of the treadmill will replace any kinetic energy the treadmill loses. So assuming the cart works, it looks like the cart and air end up getting kinetic energy from nowhere, because we don't see the kinetic energy being put into the system by the treadmill motor.

At this point, I'm not convinced the cart works, but I'm not convinced it doesn't, either. All the explanations why it can't work have holes (especially in the ground frame).
posted by samw at 10:20 AM on December 4, 2008

If someone is pushing you, can you go faster than that person? No, because if you did, you would separate from him, lose momentum transfer (because you are now two bodies with two separate masses, rather than one integral unit) and slow down.

So you're saying that if the cart were moving faster than the wind, it would "separate" from the wind? That doesn't make any sense; air is a continuous fluid.

If you went faster than the wind, you'd see a drag force. You would accelerate until that drag balanced the thrust pushing you forward, at which point you'd maintain a constant velocity.

As to whether coupling the wheels to a propeller can generate a thrust greater than the accompanying increase in drag from rolling friction, it's hard for me to say from first principles. The treadmill video certainly makes it seem so, however.
posted by mr_roboto at 10:22 AM on December 4, 2008

> p.s. can somebody explain how a crosswind and angled sail create a force-multiplier?

wayofthedodo, instead of just catching the wind, the sail acts like the wing of an airplane, creating an aerodynamic force. The shape of the sail changes the direction of flow of air, shoving the air aside; the air shoves back on the boat.

Failure to account for the aerodynamic force due to an airfoil angled with respect to the wind means you can't explain why a boat can go a little faster than the wind. Likewise, many people offering opinions here are failing to account for the aerodynamic force due to the angled airfoils of the cart's propellers and therefore can't explain the motion of the cart. In sailing, you get a modest speed boost (not continuous acceleration or perpetual motion!), and only if the boat is light and sleek. I suspect the same may be true of the outdoor cart experiment -- you see a modest boost that goes away when you try to scale it up and do real work with it.
posted by sdodd at 10:24 AM on December 4, 2008 [1 favorite]

What a stupid, obvious, attention-whore hoax. At least the cold fusion people were earnest. Don't people have better things to do? Like cheat the numerically challenged into taking unserviceable mortgages on such a vast scale that they destroy America? Why do they hate America?
posted by cytherea at 10:31 AM on December 4, 2008

Why haven't any of these guys fitted a small windsock to the cart for the outdoor experiment? I think that would solve the question to everyone's satisfaction. (If they *can* go faster than the wind, the windsock will change direction, right?)
posted by sdodd at 11:15 AM on December 4, 2008 [1 favorite]

I changed my mind. I don't think this is impossible, and I no longer think it would automatically accelerate forever. Something was niggling at me all along, and I think I figured out how this actually can work, and why it's not silly or a perpetual motion machine.

The first thing to do is simplify your mental model of what's going on. How to account for the propellor kind of had me stumped, but this high-school level page from NASA gave me a good way out. Imagine the propellor as a magic fluid-pressure-changing permeable disc. All the details of the airfoil and such are irrelevant here -- the prop is just a disc where air flows in at one pressure and out at another. The amount of force provided by the prop is proportional to the difference in pressure between the two sides of the prop. So it provides a force of F = delta p * A where A is the area of the prop's disc.

So you start your little cart off on a big flat plane with a steady wind. Put the bastard on rails so you can orient it directly downwind (why haven't any of the builders done that?). At first, the wind pushes it forward because it's light and has a relatively large sail area (the prop blades). It's starts rolling. As it rolls, the wheels engage the prop and start that turning.

Now your prop is a magic pressure-differential disc. So let the thing accelerate up to the speed everyone agrees it can go, the speed of the wind. It's tooling along, the air around it is at rest (w/r/t the cart) and the prop is spinning. I think everyone is still with me at this point, yes?

Now, if the prop is spinning, and the air pressure ahead of the prop is the free-air pressure, it is pretty much by definition producing thrust from the back. That spinning prop is going to compress the air passing through it, causing a pressure differential, causing a forward force. So, the prop will push the cart faster than it is already going, which is to say faster than the wind.

That's part one of the question. But the brain rebels, and says "WHAAA!? If the prop produces a force which accelerates the cart, that force will turn the prop faster and so on and it'll just keep speeding up!" That's what my brain said anyway. But not so! Here's why:

As the cart accelerates past wind speed, it begins to experience a headwind. Many people have pointed this out. And what is a headwind, from the point of view of our magic propellor disc? It is nothing less than an increase in air pressure on the front (low-pressure) side. The faster the cart goes, the greater this front-side pressure is. This is our missing limiting factor. There will come a point where the propellor is no longer able to increase this pressure from front to back. This is basically the prop's stall speed -- its airfoil can no longer provide any "lift" through the air. At this speed, the pressure in front of the disc (from apparent headwind) is the same as the pressure behind, and no more force is applied.

Figuring out what this maximum speed actually is for some particular cart design would require a lot of physical details I don't have. What I'm sure of is that every device will have such a speed. When it's reached, drag from the wheels will start to slow the cart down, until it gets below "prop-stall" speed again, and the prop starts applying force again, and so forth until, if you have enough track, some equlibrium speed is reached. The cart should travel along at this equilibrium speed as long as you have moving air around it.

So. I have demonstrated, to my own satisfaction anyway, two things:

* The propellor will produce a thrust force when it's moving forward at the speed of the surrounding air, and

* The increase in speed is self-limiting.

It's an open question, to me, how much faster than the wind it could go. My sense is not a whole lot, but some. Pretty vague, I guess. If I were to try to order-of-magnitude it, I would guess it would be pretty similar to how much faster than the wind a good iceboat could go on a reach. But that is just a guess.
posted by rusty at 11:20 AM on December 4, 2008 [2 favorites]

rusty, you're wrong. You're stall speed for the prop is the speed of the wind, and your mistake is thinking the propeller is spinning when you reach the speed of the wind. The prop spins because the air is passing over it and apply a force to the blades. If you hold the prop still, the prop spins as fast as possible. If you allow it to power a cart in the same direction as the wind, the air flowing over the propeller decreases, which in turn decreases the power it's capable of producing. Eventually, if you manage to reach the same speed as the wind, there's zero air passing over the propeller and it's not spinning.

Regarding sailing and moving faster than the wind, it's definitely possible. However, you are not moving faster than the wind in the direction the wind is going. If you're on a beam reach (sailing 90 degrees in relation to the wind), it doesn't matter what speed you're travelling at, the actual wind is still coming over the side of your boat at the same speed as it did originally (your relative wind though, will feel like it's coming from 45 degrees or so due to the wind created by moving forward).

On a boat, the wind can move the boat in two ways. The first is by using the sails to be pushed by the wind, and the second is by using the wind & sail to create lift. The fastest way I've ever sailed is close hauled, which means you're sailing upwind, at about a 45% angle to the wind (some racing boats can sail even closer to directly into the wind, but there's always a point where you enter the no-go zone, you can never sail directly into the wind for the same reasons you can never sail directly downwind faster than the wind).
posted by ShadowCrash at 11:37 AM on December 4, 2008

The prop spins because the air is passing over it and apply a force to the blades. If you hold the prop still, the prop spins as fast as possible.

You have the angle of the blades backwards. The prop is being spun by the wheels, which are geared such that the prop generates a pressure gradient as rusty described.
posted by mr_roboto at 11:42 AM on December 4, 2008

ShadowCrash: The propellor is spun by a gear from the wheels. If the wheels are turning, the prop is turning. When the cart is moving at wind speed, the wheels are most definitely turning, so therefore the prop is too. It's not being driven by airflow.

And it seems pretty clear to me that when the cart is moving at the speed of the wind, the air around the cart is still from the cart's perspective. No?

If so, you have a spinning prop in still air. It's hard to imagine how that doesn't produce some thrust. And the math backs me up on this -- the thrust depends only on the pressure difference from front to back of the propellor.
posted by rusty at 11:45 AM on December 4, 2008

And I request that everyone who favorited my other comment above please unfavorite it. It is wrong. :-)

Unless, you know, your favorites list is like a "most wrong" reference list for you. In that case, leave it alone.
posted by rusty at 11:54 AM on December 4, 2008

The propellor is spun by a gear from the wheels. If the wheels are turning, the prop is turning. When the cart is moving at wind speed, the wheels are most definitely turning, so therefore the prop is too. It's not being driven by airflow.

In the case of the treadmill video, maybe, provided that there's not a fan which we cannot see off-screen.

But that is not the case in the original video. In the original video, the prop is spinning because air is passing over it.

I found this video on youtube which helps to illustrate what I think is going on in the first video:

DDWFTTW

The propeller (in this case actually a windmill) is operating on the same principle, except that instead of a little sledge with a sail, you have two sails spinning around each other, powering the wheels.

There's no thrust being generated by the propeller. It's able to go downwind faster than the wind but probably not a lot faster. In fact if you reverse the camber of the propeller blades, by flipping the propeller around, you'd be able to move directly upwind as well.
posted by smoothvirus at 12:12 PM on December 4, 2008

But that is not the case in the original video. In the original video, the prop is spinning because air is passing over it.

No, it's not. What make you think that this is the mechanism?
posted by mr_roboto at 12:14 PM on December 4, 2008

What make you think that this is the mechanism?

Because a propeller blade is an airfoil. The wind moving over the blades makes it spin. Same as a pinwheel. Same as a windmill.

If it was the wheels providing energy to the propeller, the coefficient of performance > 1.0. This is an overunity, which cannot work.

However this does not appear to be the case with the treadmill video. In that case, the wheels are transferring energy to the propeller from the treadmill belt. Propeller provides thrust. I think you're right about what is at work there.
posted by smoothvirus at 12:24 PM on December 4, 2008

Ok, Rusty, I see what you were trying to say. The cart is pushed by the wind, and the wheels are driving a propeller, which adds additional thrust. In that case, the cart will never reach the speed of the wind.

The cart is being pushed by the wind, and as long as nothing acts as a break on the cart, it will attain the speed of the wind. As soon as you add a braking mechanism (even friction from the tires against the ground), it's no longer a perfect system, and you have a negative force that counteracts at least part of the force of the wind. If you increase the drag by forcing the wheels to power a prop, you've increased the negative force acting against the wind and slowed yourself even further. If it was perfect system, the drag created by spinning the tires to power the prop would be perfectly offset by the prop pushing the cart forward. If it's not a perfect system, you'll lose some energy to heat, some to noise, etc...
posted by ShadowCrash at 12:25 PM on December 4, 2008

The treadmill video is confusing. I see what the creator of it is getting at but it's a different principle at work. Originally I thought there was a fan off-screen that the treadmill was counteracting, but it appears that this is not the case.
posted by smoothvirus at 12:28 PM on December 4, 2008

This technical explanation explains how you can sail faster than the true wind.

Note that it requires some velocity component that is orthogonal to the direction of the true wind. Which is a fancy way of saying it doesn't work when you go directly into or away from the true wind.

I'm not convinced that you can perpetually exceed the wind speed. I am convinced that you can exceed the wind speed under various conditions, such as storing energy to be used later, or an initial push (until the energy of that push is lost), or while the apparent wind is greater than the true wind (which can't last forever, see technical explanation above). So I'm still in the boat where I agree you can go faster than the wind, but I don't see how one can do it perpetually.

Someone would have to explain to me how wind can have more energy to be extracted than that contained in just its movement before I can buy a perpetually faster than the wind machine.
posted by forforf at 12:30 PM on December 4, 2008

And not just perpetually, I don't see how one could do better than say a dandelion seed floating on the wind.
posted by forforf at 12:33 PM on December 4, 2008

The cart is being pushed by the wind, and as long as nothing acts as a break on the cart, it will attain the speed of the wind. As soon as you add a braking mechanism (even friction from the tires against the ground), it's no longer a perfect system, and you have a negative force that counteracts at least part of the force of the wind.

Correct if you're using a sail, and not a propeller/windmill.
posted by smoothvirus at 12:34 PM on December 4, 2008

smoothvirus, a computer generated video of fake balloons and fake carts doesn't count. The boat could tack back and forth downwind like the video and the boats speed relative to the ground may be greater than the actual wind speed (but not greater than the boats relative wind speed). However, the boats speed in relation to the wind (ignoring the speed side to side) is not greater than the wind.

Technically, you could count such a device as sailing directly into the wind, if it did exist. However, I'm not sure the cart itself would stay straight. When the boat is sailing into the wind and is tacking back and forth, the force acting on the boat also pushes it on it's side (the keel/water counter acts this and stops the boat from lying down). I doubt the boat would provide a perfectly straight forward push to the two cables, and I'd expect some torque to be applied. Maybe you could keep it going straight (say if it was on rails), but then you're introducing an outside force to add a counter-force.
posted by ShadowCrash at 12:37 PM on December 4, 2008

smoothvirus, what starts the cart moving? By sail I meant the wind blowing on the cart, it doesn't need to be an actual sail.
posted by ShadowCrash at 12:40 PM on December 4, 2008

If it was the wheels providing energy to the propeller, the coefficient of performance > 1.0. This is an overunity, which cannot work.

How so? The wheels are being driven by energy from the wind.
posted by mr_roboto at 12:42 PM on December 4, 2008

The treadmill video is confusing. I see what the creator of it is getting at but it's a different principle at work.

The carts operate the same way, and the reference frames are equivalent inertial frames. On the treadwill, we're simply moving at the same speed as the wind.
posted by mr_roboto at 12:45 PM on December 4, 2008

the boats speed relative to the ground may be greater than the actual wind speed Isn't that the question? A vehicle that is wind-powered, and can move directly downwind with a ground speed greater than the windspeed at a fixed point?

No, you're not going to go faster than the relative windspeed. But I'm convinced that a groundspeed greater than the surface wind speed is possible.
posted by smoothvirus at 12:47 PM on December 4, 2008

And after posting that I think I finally understand how it can be done. The secret is that the device has to turn the wind (which can be used to create a force differential, i.e., lift).
Now think of the device on rails (like rusty mentioned above), with the wind coming in at an degree angle to the device at X knots/hour. If you put a sail on the device, angled just so, you would be able to achieve a speed slightly greater than X knots/hour due to the added lift from the wind. This lift is caused by the wind being turned by the sails, and the difference in velocity between the inside the sail flow, and the outside of the sail flow (same as the wing of an aircraft). You get that energy because you are on rails, and the rails can push against that flow direction. In other words, this won't work unless you have a fixed direction (i.e, a sailboat keel, rails, or wheels that don't freely turn). I think that was the variable that I was missing.

Friction is being used (the rails in this case) to turn the wind, so you can capture more energy. The friction comes for "free", but rather than a bug, you make it a feature. The friction is resisting being pushed "off course" that resistance is transferred to maintaining a wing (sail, propeller, whatever) position relative to an angled wind, creating a differential in wind speed across its surfaces.

I'm not 100% convinced I'm right, but I'm getting there ....
posted by forforf at 12:47 PM on December 4, 2008

So the question is can that phenomenon be transfererred to a propeller based system, where the propeller is acting as a fixed sail/wing wrt to an angled wind, but is also rotating. That I'm not sure about.
posted by forforf at 12:49 PM on December 4, 2008

I'm not convinced that you can perpetually exceed the wind speed. I am convinced that you can exceed the wind speed under various conditions, such as storing energy to be used later, or an initial push (until the energy of that push is lost), or while the apparent wind is greater than the true wind (which can't last forever, see technical explanation above). So I'm still in the boat where I agree you can go faster than the wind, but I don't see how one can do it perpetually.

You seem to be confusing velocity and energy, and postulating a law of conservation of velocity.
posted by mr_roboto at 12:50 PM on December 4, 2008

How so? The wheels are being driven by energy from the wind.

Right. That's what I'm saying.

In the first video, not the treadmill video, the wheels are driven by energy from the wind.
posted by smoothvirus at 1:00 PM on December 4, 2008

In the first video, not the treadmill video, the wheels are driven by energy from the wind.

Right. The wind is pushing the body of the cart, like a sail. This turns the wheels, which turn the propeller, generating thrust. When the cart is moving at wind speed, there is no longer any net force from the wind on it, but the thrust from the propeller accelerates it. Once it is moving faster than the wind, there's a drag force on the cart, and it will reach a constant velocity when drag and thrust are balanced.
posted by mr_roboto at 1:02 PM on December 4, 2008

forforf

You're almost there. One half of a propeller is a wing.
posted by smoothvirus at 1:03 PM on December 4, 2008

ShadowCrash: In that case, the cart will never reach the speed of the wind.

I'm willing to accept that as a possibility, in the real world. It seems pretty reasonable that making the wheels push the prop around would act as a braking force, to some extent. I think it's an open question whether the prop turning could counteract or exceed that. It's not obvious to me how the real-world math would turn out on that.

As soon as you add a braking mechanism (even friction from the tires against the ground), it's no longer a perfect system, and you have a negative force that counteracts at least part of the force of the wind. If you increase the drag by forcing the wheels to power a prop, you've increased the negative force acting against the wind and slowed yourself even further.

The thing is, that it's not a flipping-switch kind of situation, where the cart is being pushed along like a sailboat running downwind until it reaches the speed of the wind, and then switches to being pushed by the prop. As soon as the propellor blades are turning, the prop is generating some thrust. And in fact, while the cart is still moving at less than windspeed, the prop ought to be generating relatively more thrust than it would be if the air was still (and it was turning at the same speed, you know?). The wind behind the prop just increases the pressure at the back of it, like any tailwind.

So that's why I say it's an open question. It seems like it depends a lot of the actual numbers. Judging from the videos I've seen of these carts, it looks to me like they're light enough and there's little enough added fristion and whatnot, compared to the size of the prop, that I can believe they work, to a small extent.

I don't think this design would scale up in any useful way though.
posted by rusty at 1:05 PM on December 4, 2008

The wind is pushing the body of the cart, like a sail.

The wind is pushing the body of the cart a little, but probably not enough to even make it roll. The wheels are not making the propeller spin, the wind is.
posted by smoothvirus at 1:05 PM on December 4, 2008

Think of it this way, what would happen if someone held the wheels off the ground? The cart wouldn't go anywhere while it was being held up, but the propeller would turn and the wheels would spin.
posted by smoothvirus at 1:10 PM on December 4, 2008

The wind is pushing the body of the cart a little, but probably not enough to even make it roll.

No one is claiming that it can get up to speed on it's own, just that when it's reached wind speed it can accelerate past.

We're arguing about different contraptions here. I'm right about mine, and you're right about yours.

Think of it this way, what would happen if someone held the wheels off the ground? The cart wouldn't go anywhere while it was being held up, but the propeller would turn and the wheels would spin.

Based on the gearing of the device I'm describing (and the one I believe is portrayed in all of these videos), they would spin backwards.
posted by mr_roboto at 1:13 PM on December 4, 2008

smothvirus: The propellor would turn backward and the wheels would spin backward.
posted by rusty at 1:15 PM on December 4, 2008

We're arguing about different contraptions here.

Yes and this, I think, is the difference in between the two videos. Because if the cart in the first video is geared that way then it would roll backwards. Or at least, into the wind rather than away from it.

However, in the second video, it's geared the other way, the way mr_roboto and rusty are describing. And in that case, the wheels are driving the propeller.
posted by smoothvirus at 1:52 PM on December 4, 2008

Because if the cart in the first video is geared that way then it would roll backwards.

It might, were it not for the push he gave it to start it off.
posted by mr_roboto at 1:55 PM on December 4, 2008

It might, were it not for the push he gave it to start it off.

Not for long, because the wind would be pushing the propeller in the other direction. The cart would slowly come to a stop, and then, start rolling backwards.
posted by smoothvirus at 1:58 PM on December 4, 2008

Well, that depends entirely on the gear ratio. It could be designed either way, really. This device is based entirely on the use of gearing to get a mechanical advantage.
posted by mr_roboto at 2:05 PM on December 4, 2008

How is it that my arguments have gone totally unheeded and unnoticed.

Did I write a FNORD! or something? WTF?
posted by loquacious at 2:35 PM on December 4, 2008

What Rusty said, way back there. Sticking to the treadmill example as it's simpler to describe and contains fewer red herrings:

The energy to rotate the propeller and create thrust is acquired at the cost of drag on the wheels. For this to work as described, you'd have to get more thrust out of spinning the propeller than it costs to spin it. I don't care what your gearing or propeller pitch is, this is getting something for nothing, period.
posted by George_Spiggott at 3:08 PM on December 4, 2008

The energy to rotate the propeller and create thrust is acquired at the cost of drag on the wheels. For this to work as described, you'd have to get more thrust out of spinning the propeller than it costs to spin it.

You need to generate more force than is exerted on the wheels. This requires only a simple machine. Remember mechanical advantage from high school?

If, at wind velocity, you generate a thrust force greater than the associated increased drag force on the wheels, the vehicle will accelerate.

You're treating force as if it is a conserved quantity. It is not. Energy is conserved; in the treadmill example, the energy to accelerate the cart is coming from the treadmill motor. In the outdoor example, the energy is coming from the wind.
posted by mr_roboto at 3:24 PM on December 4, 2008 [1 favorite]

Seriously, George. Your language up there is very confused, and you seem not to be making clear distinctions between force, work, and energy.
posted by mr_roboto at 3:25 PM on December 4, 2008

Your explanation doesn't help, but let me rework the question to put it in common terms -- let's stick to energy, and nothing else. To accelerate you need a net energy input. The continously supplied energy of the powered belt relative to the stationary air constitutes a net energy input which the device can exploit to move forward, overcoming the reverse drag.

Okay, I'll buy that as not describing a technical impossibility.
posted by George_Spiggott at 3:34 PM on December 4, 2008

The continously supplied energy of the powered belt relative to the stationary air constitutes a net energy input which the device can exploit to move forward, overcoming the reverse drag.

Yeah, that's it exactly. If we move back outside, the motion of the wind relative to the stationary ground is what supplies the energy. They're equivalent frames of reference.
posted by mr_roboto at 3:41 PM on December 4, 2008

rusty, you're right.

And, I hate that the only favorite I have is for something I no longer believe :)
posted by Chuckles at 4:40 PM on December 4, 2008

And I request that everyone who favorited my other comment above please unfavorite it. It is wrong. :-)

Must preview.. It is a horrible feeling, isn't it.
posted by Chuckles at 4:42 PM on December 4, 2008

placeholder
posted by five fresh fish at 5:11 PM on December 4, 2008

Oh I hate doing this. Let's just do a quick thought experiment. We have wind moving relative to some surface. Reference frames are not important, have the surface moving, have the wind moving, same physics. But let's choose the reference frame of the wind, so the road appears to be moving. And let's have the machine start off moving relative to the road at exactly the same speed as the wind, perhaps we held it motionless on the treadmill and then released it. Now, the wheels are spinning from road friction, and that energy is being transferred at a loss, due to the mechanical friction of the coupling (and the second law of thermodynamics) to the propeller, making it spin. Because the propellers are an airfoil, their rotation provides (rather lossy) forward thrust. But how could this thrust be more efficient than the wheels? Do you think that it would be better to have a car engine linked to a propeller, or an engine linked to a wheel in direct contact with the ground? And there you have it. It's a propeller versus wheels. Propeller loses. Why must people keep trying to square the circle?

In the first video they claimed that the first device was moving at 10 mph, while the wind was at 7 mph. But there's no evidence for that beyond their words. In the second video, I'm sorry, but there was the distinct sound of a fan blowing. And no wide angle shots. And, if you pay attention to the language of the voice over, those were not the words that someone would have used to describe a fantastic new invention, which should have been quickly patented. No, that was the language of a jerk trying to get some attention on that awful cable show.

I'm kind of disappointed that such a shoddy hoax would garner this much attention. If we can't discern the truth from a hoax in the most amendable and verifiable of subjects, newtonian physics, then how can we possibly hope for rational decisions in the social and political spheres?
posted by cytherea at 8:34 PM on December 4, 2008

What I said above about energy inputs, I still don't see an escape from the simple problem of where the thrust to overcome the drag comes from. In the treadmill model, the only energy input to the cart itself is the rotation of the wheels. The mechanical energy extracted from this is solely responsible for turning the prop. The energy thus extracted constitutes drag. For this to work, the thrust produced by the prop must be more than sufficient to overcome the drag that produced it.

A sailboat can in principle run faster than the wind because it extracts energy from more than just the force of the wind hitting the surface of the sail: the shape of the sail creates a flow which involves a larger volume of moving air than that which intersects its surface and thus presumably has a larger-than-intuitive amount of energy available to extract.

I don't see anything similar in the treadmill model. Mechanical drag on the wheels turns the prop, the turning prop creates thrust. If you're going to yell about "force multipliers" and claim that it's a confusion of terms, back it up with actual numbers. The reduction in velocity caused by the mechanical costs of turning the prop is, we are asked to believe, exceeded by the velocity added by the thrust of that turning prop. "Force multipliers" don't cut it. If you can use a block and tackle to lift a piano by yourself it's only because you're pulling your end of the rope two or three feet for every foot the piano goes up. I see no corresponding reduction available here no matter how you gear your transmission. The belt pushes the cart backwards against the air, the wheels turn the prop and the cart (we are told) goes forward more than it got pushed backwards. Even assuming zero mechanical loss I don't see how it does better than stay in the same place.

(On preview, looks like cytheria's on this.)
posted by George_Spiggott at 8:37 PM on December 4, 2008

that awful cable show

BLASPHEMER!
posted by smoothvirus at 8:42 PM on December 4, 2008

Should have previewed again. That first sentence should read "Forget what I said above about energy inputs." And sorry for misspelling your name, cytherea.
posted by George_Spiggott at 8:43 PM on December 4, 2008

Oh, everyone spells my real name incorrectly to. No harm. But yes, you're right about the block and tackle. You could have a propeller linked to a screw drive, and that could make headway against the wind, more slowly than the wind velocity. But not in a one to one situation. And I'm sorry, but that show is so loud! I did try my best to watch it.
posted by cytherea at 8:54 PM on December 4, 2008

Hm. Dammit, I think I can counter my own argument here. Sticking with the treadmill scenario for consistency, let's replace the propeller with something with real traction and see what we get. If the cart had traction against something that were physically stationary, it could convert backward thrust into forward motion in a reasonable way. If for example the wheels under the cart were connected via an axle to wheels that rode on the cart's outer frame rails (the circumference of the outer wheels would have to differ from the inner, and the heights of the rails relative to the belt adjusted in a pretty precise ratio for this to work smoothly without slippage) the backward thrust would be translated into forward motion. (The ratios matter a great deal: invert them and instead of going forward the cart goes backward twice as fast, more or less.)

So, if I've pictured this correctly, non-moving air can presumably be regarded as a lower-traction medium but comparable at least in principle. If we treat the surrounding air as a stationary object (albeit one with far less traction than wheels on the treadmill rails), the cart can utilize the force pushing it backward to drive itself forward. So it's technically possible for the cart to gain net velocity opposite to the thrust being applied to it, and the cart moves forward. Thrusting with air rather than wheels (or even toothed gears for that matter) is a lot lossier but this seems like it can work even for very lossy transmission, given that we have a lot of available input from the treadmill. This mental picture is a little clumsier to depict for the outdoor/wind scenario, but imagine the same cart and pair of moving outer belts in place of the wind and it sorta becomes clear.
posted by George_Spiggott at 12:59 AM on December 5, 2008

(Correction to the above, the outer wheels ride on the treadmill's outer frame rails, not the cart's. And I'm out of here before it gets any sillier, if that's possible...)
posted by George_Spiggott at 1:03 AM on December 5, 2008

Yeah no, this isn't a stumper at all. Here's Mark C. Chu-Carroll to explain it all: Wind-powered perpetual motion.

Chu-Carroll has now changed his position from (extreme) skepticism to support:

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.
posted by Mapes at 7:38 AM on December 6, 2008

Chu-Carroll has now changed his position from (extreme) skepticism to support

Man, that's satisfying. Also, it's good to know that someone who really knows enough physics to really do the math actually did the math.
posted by samw at 8:15 AM on December 6, 2008

I stand corrected. Yay science!
posted by Skorgu at 12:32 PM on December 6, 2008

Here's his explanation as to why it works. interestingly enough, he starts by illustrating something functionally equivalent to the word picture I tried to describe above, though his approach is easier to depict in two dimensions. He also explains why that approach, that involving the pure mechanical linkage. Will not work. (Oops again, it also applies to my model. Once the outer wheels are moving forward on their track they are ceasing to get thrust from that track. Damn.) But then he goes on to explain why the fluid dynamics of wind propulsion aren't limited in this way. Fascinating stuff.
posted by George_Spiggott at 12:36 PM on December 7, 2008

Most of the explanations, including this Chu-Carroll person's, are really terribly tortured. If you can't do the prop calculations, all the analogies are just so much hand waving.

And in addition, every time I read some of what he has to say, he seems to repeat this gross error:

So, initially, you get a forward accelerating force, which is proportional to the wind velocity.

No, it isn't proportional at all. Aerodynamic forces are proportional to velocity squared! And this is really quite important, because the non-linearity helps to throw our intuition off.

And this:

What stops it from being perpetual motion? As you accelerate above windspeed, you'll start to suffer a vaccum effect - the wind is moving slower than the vehicle, so it'll effectively be sucking away some of the force generated by the propellor. So you'll need to start subtracting a force proportional to the relative windspeed from the amount of force you can generate from the propellor. Eventually, that vaccum force will equal GAvcart/2πr. As you get closer and closer, the amount of force that you'll be generating by spinning the propellor will be completely consumed - and you'll be at the maximum attainable speed in that amount of wind.

Nonsense! That is an explanation of why the cart doesn't continue to accelerate forever -- why it reaches a 'critical velocity'. The reason it is not perpetual motion is that there is energy to be extracted from the wind (or the wind to surface velocity difference, if you prefer).

I dunno.. personally, I think analogies are fine in general. When they are simple and direct, it can really help understanding. You have to be very careful to fall back on the underlying physics at every opportunity though, or the analogy will quickly fall apart.
posted by Chuckles at 1:16 PM on December 7, 2008

Funny, in my own analysis I drew out a rigid gear diagram exactly like Chu-Carroll's. I think he concludes too quickly that a rigid gear system doesn't work, however; I'm pretty sure you can achieve v_cart > v_upper by adding a concentric gear into the mix. Specifically, mount a small concentric gear on the large lower gear, then link the middle gear to the concentric gear and drop the top gear completely.
posted by Galvatron at 10:20 AM on December 8, 2008

I think the key thing here is that even in that first video the craft doesn't maintain a speed greater than the wind - it is generally moving slower than the wind and 'gusts' of wind will push it abit faster.

which I beleive is due to a kind of momentary flywheel effect in the 'propeller' type device (which seems quite heavy). This stores the energy of the wind during a gust, and 'releases' it afterward giving momentary faster than wind movement.
posted by mary8nne at 4:16 AM on December 15, 2008