Sailing Faster Than The Wind Itself

If you search the outer reaches of the internet you will find all sorts of web sites and videos purporting to answer to free energy in the form of perpetual motion machines and other fantastical structures that bend the laws of physics to breaking point. We’d love them to be true but we have [Émilie du Châtelet] and her law of conservation of energy to thank for dashing those hopes. So when along comes a machine that appears to violate a fundamental Law of Physics, it’s reasonably met with skepticism. But the wind-powered vehicle built by [Rick Cavallaro] looks as though it might just achieve that which was previously thought impossible. It’s a machine that can move with the wind at a speed faster than the wind itself.

A fundamental law of sailing boats is that when they are sailing with the wind, i.e. in the same direction as the wind, they can’t sail faster than the wind itself. Sailing boats can go faster than the wind powering them by sailing across it at an angle to create lift from their sails, but this effect doesn’t work as the angle tends towards that of the wind.

The vehicle in the video below the break is a sleek and lightweight machine with a large propeller above it, which we are told is not the windmill power source we might imagine it to be. Instead it mimics the effect of a pair of sailing boats sailing across the wind in a spiral around a long cylinder, and thus becomes in effect a fan when turned by the motoin in the craft’s wheels. The drive comes from the wind working on the craft itself, and thus as can be seen from the motion of a streamer on its front, it can overtake the wind. It seems too good to be true at first sight but the explanation holds water. Now we want a ride too!

For fairly obvious reasons, the fantastical world of pseudo-physics isn’t our bag here at Hackaday. But if something might hold promise we’ll at least give it a look. Not all such things we cover turn out to change those Laws of Physics, though.

Thanks [Feinfinger] for the tip.

136 thoughts on “Sailing Faster Than The Wind Itself

  1. “But the wind-powered vehicle built by [Rick Cavallaro] looks as though it might just achieve that which was previously thought impossible. It’s a machine that can move with the wind at a speed faster than the wind itself.”

    Then the author explains that we haven’t previously thought this impossible since perhaps the 19th century.

    Oh me! Oh my! Oh HaD! Obligatory complaint about the title.

    1. A lot of comments to read about this and I haven’t read all of them. I did look at a fair bit and didn’t notice anyone mentioning history of this project. I read about it over 10 years ago when a challenge had been made to do this. There was a $1m prize put up to anyone that could achieve this goal. It was ‘done’ according to what I read some place like Popular Mechanics or Science, seems to me the top speed was around 2.5 x measured wind speed and the pilot abandoned any further attempts to increase speed due to the vehicles Prop developing enough torque to overcome gravity. The vehicle would not stay on all wheels but started to roll over, he could not control his direction of travel. And vibration became a major issue. Also one should consider the danger of having that big spinning blade up there!

      1. Why not 2 fans spinning in opposite directions (this gets rid of gravity influence of single fan)?? That has been done for years on aircraft. Also the airfoil on the blades if chosen correctly should provide lift in the direction of travel same as airplane wing. This means that the pitch on the blades will need to vary according to the wind speed to optimize force in the forward direction (and perhaps change airfoil profile).

        1. Well, fans are generally balanced, so there is no “gravity influence” involved. What he was talking about was torque, which can be balanced with counter-rotating props. Just the same, once you’ve met the specs with a margin of 150%, it’s a good place to stop anyway.

          1. Kinda. The comment section on this article is worse than most, with people saying things that are explained thoroughly in the article and/or video. But I was wrong anyway. His question was about the previous comment. I apologize.

        1. Technically it is at an angle to going directly down wind. The angle represents a sailboat sailing on a broad reach course to the true wind, perhaps 135 degrees away from the direction of the source of the true wind rather than 180 degrees. It is often faster to sail this course and periodically jibe (tack down wind) to the opposite 135-degree course such that the boat sails a zigzag course-made-good averaging 180 degrees from the true wind. The sail(s) provide considerable lift as opposed to mere drag and that is how a good sailboat exceeds the wind speed when sailing down wind.

          However, the relative comparison should be between the sails of a sailboat and the blades of this craft. If you look at the apparent wind relative to the blades then you will notice the apparent wind angle varies with the craft’s speed (and the rate of blade rotation). They start out providing 100% drag and zero lift while still stopped. But once moving the blades begin to turn and turn faster as the craft picks up speed. As the speed of the craft and the rotation of the blades increase, the apparent wind as experienced on the blades, narrows until it operates like a sail on a sailboat when sailing close hauled or at about 45 degrees away from directly toward the true wind. The blades now offer zero drag (in terms of pushing the craft) but considerable lift. Now, any drag will have negative impact on further increasing speed.

          At this point, the craft is moving in the same direction and at the same speed as the wind and the blades are performing purely as a propeller. There is no wind resistance acting on the craft at this speed, just rolling resistance. There is wind resistance to the rotating blades but there is also a net positive force applied to the craft in the down wind direction. If the propeller generates enough power to exceed only the rolling resistance, then it will accelerate the craft to a speed greater than the wind that is driving it.

    1. It’s different though, isn’t it? I get this, but I don’t get the sailboats.
      This works because it’s the wheels not the wind driving the fan. But in a sailboat, the wind is driving the sail.

      1. Perhaps I am benefiting from my experience being taught to sail Sunfish sail boats when I was a teenager. Having had the experience of piloting a boat tacking into the wind, picturing the motion of a pair of tacking sailboats doesn’t seem like a foreign concept to me.

        That being said, there is much that seemed easily misunderstood about this whole experiment, especially before I had the “ah-ha!” moment. For instance, I would have liked to have seen exactly how the wheels and prop/fan/sailboats-on-a-cylinder were geared together. Also, the knowledge that the initial push comes from the wind might have been far more useful to me had it come near the beginning of the video instead of being part of the wrap-up.

      2. A simpler device showing how this work would be a large Ferris wheel, with sails that are slack when at the top of the wheel and taut and catching the wind when at the bottom. The wind pushes on the sails under the wheel’s axle, and the axle travels over the ground faster than the sail, and eventually the wind. Essentially, the craft extracts energy from the difference in speed between the wind and the ground.

        1. but that’s not the principle by which this works. Its not a matter of wind-gradient, its a matter of slowing wind behind you (creating a high pressure zone) to extract work.

          1. I agree that this machine does not use the wind gradient as its main means of propulsion.

            However, I also question the concept of slowing the wind. I think the slowing of the wind is more an effect than a cause, much like the slowing of the wind behind a sail or a windmill is not the source of propulsion or of energy creation.

            The blades are like sails. At first, when the craft is starting out and moving slowly, the blades just produce drag for relatively slow downwind sailing. As the craft picks up speed, the blades turn such that they operate similar to the sails of a sailboat when sailing on a broad reach. This is more efficient than simple drag and increases additional propulsion. Finally, as the craft catches up to the wind speed and beyond, the blades effectively transition in behavior to sails when a sailboat is beating to windward. This is also similar to blades on a fan. The blades just need a means of energy to turn them.

            They get that energy from the movement of the craft. The craft gets its motion initially from the drag of the wind but increasingly it gets more propulsion from the blades as it picks up speed. Once the craft achieves the same speed as the wind or more, the blades provide all of the propulsion.

            To suggest that the wind pushing the craft could not turn the blades would mean that the wind could not move the craft at all. We know that not to be true. So the wind pushing the craft will turn the blades initially. The rotating blades then add to the propulsion. Even at the speed of the wind where there is zero apparent wind over the craft, the blades must still turn because the craft is moving and, like a fan, create wind in the opposite direction to the movement of the craft pushing it even faster. It’s quite clever. And I believe understandable without breaking the laws of physics.

            It might help to imagine the apparent wind relative to the rotating blades instead of relative to the moving craft.

      3. In a sailboat the hull is being squeezed along by being pushed by the wind on one side and the water on the other. I think that one source of confusion in this whole discussion is the use of the term tacking instead of jibing. Tacking is used to sail upwind.

    2. The two sailboats on a cylinder still wouldn’t outrun a balloon, though. They’re still each moving faster than the wind, but not in the same direction. A sailboat can’t overtake a balloon without repeatedly switching directions to leverage its momentum to repeatedly overtake the wind momentarily. Yes, the craft *does* work, but that explanation was bogus. See my other comment about gearboxes.

      1. The America’s cup AC75 boats have a VMG downwind faster than the wind they sail into. VMG is the component of speed towards the gate. Essentially, they can outrun a balloon by running a zig-zag path.

  2. I still don’t get what principle people think should be impossible about it. If you have access to two mediums with different velocities, you can extract energy from that. If you have access to energy and are sufficiently efficient, you can go very fast.

    Just recently on hackaday, we had a model glider that could accelerate close to the speed of sound just from a natural wind shear.

    Of course, actually building this is an impressive feat!

    1. It’s not as easy as “different speeds”. Take a sailboat. It has access to different media, no matter how efficient you make it (ice-sailboat?), it can’t go directly downwind faster than the wind.

      1. If you have two sailboats and define them as one “thing”, you can have them zig-zag away from the wind in such a way that their/its center of gravity moves directly downwind faster than the wind. If you like, connect them with a functionally useless rubber line.

        Maybe it’s a problem if you only have a rigid body to work with, i. e. just one big moving part. Like a sailboat, which may not be rigid, but stays in one fixed configuration for a given course and wind speed.

      2. Not true. America’s cup hydrofoiling catamarans routinely sail both upwind faster than the wind (!!) as well as downwind much faster than the wind. from wiki:

        “The catamarans used for the 2013 America’s Cup were expected to sail upwind at 1.2 times the speed of the true wind, and downwind at 1.6 times the speed of the true wind. They proved to be faster, averaging about 1.8 times the speed of the wind with peaks slightly over 2.0.”

        1. False, the wiki is referring to net speed, not the “down wind velocity component”. The term “down wind” in the wiki is not referring to a “down wind velocity component”.

          This vehicle is actually unique as far as I know.

          1. The wiki is likely referring to the “downwind velocity component” because most reasonable cruising sailboats today can exceed windspeed on some courses. That hasn’t been news for over a 100 years.

            I agree that this vehicle is unique.

      3. Also, as mentioned on this site before and on comments above, “dynamic soaring” RC sailplanes are nearing trans-sonic speeds using semi-normal terrestrial wind speeds.

      4. Of course it’s as easy as to mediums moving at different speeds. That’s what gearboxes do. They take two things moving at different speeds, (of rotation, in that case) one being the input shaft, the other being whatever the frame of the gearbox is anchored to, and harness the difference to drive the output shaft, potentially faster than the input shaft.

        This craft just “gears up” a difference in linear motion.

      5. The point is that the blades aren’t going directly downwind. They are crossing the wind and generating their own wind from the speed of rotation and therefore changing the apparent wind as seen at the blades. That apparent wind generates lift which is much more powerful than drag and extracts far more energy from a given area of wind. It isn’t the same as a sail pushing a boat directly downwind.

    2. Most people aren’t subjecting it to a first-principle analysis at all. They’ve heard about the fallacy of a perpetual-motion machine, and jump right to concluding that is what is being presented. It’s a confluence of several biases–including the conceit that most relatively smart people have that everybody else is most likely not as smart, that the internet is full of low-grade scams preying on the proles and nothing else, and that everything so simple would already be well known.

      The last one is a profound social fallacy–nobody put wheels on luggage until the 1970s. A lot of relatively simple and valuable solutions to common problems are still not widely practiced. We would be a lot better off if it weren’t such a tiny segment of the population looking for them.

      1. >including the conceit that most relatively smart people have that everybody else is most likely not as smart

        That’s not what Dunning and Kruger found. Quite the opposite – smart people think themselves as stupid, while stupid people think themselves as smart, because the relatively smart people are aware of people much smarter than they are and can understand their merits, while the stupid people are not and cannot.

        1. More advanced cultures, thousands of years earlier, developed the crab-claw sail along with the proa. European Explorers were astounded by their speed (and voyaging capability) when they explored the South Pacific. The aerodynamics of the crab-claw sail was applied to the Space shuttle and the delta wing, It draws more on vortices than laminar flow.
          As Harry Truman used to say, the only thing new is the history we haven’t read yet.

          The polynesians also discovered new islands hundreds of miles away by observing wave patterns on the ocean, and used that knowledge to navigate thousands of miles across open ocean (eg. Tahiti to Hawaii) without a compass. New Zealand is in the South Pacific. Its current name comes from the Dutch who explored those waters hundreds of years ago. The modern America’s cup technology is a resurrection based on more modern materials. The inspiration to go fast was always there.

          1. I have heard the claim of finding distant islands by wave patterns many times. I would like to see it demonstrated. Can you see these waves from space? How big are they that they don’t vanish into the noise? Why did they think this was true? As in how was the knowledge gained? Those who guessed wrong were not around to tell anyone they were wrong. I can see learning a little from drifting down wind beyond sight of your island, but that isn’t hundreds of miles.

      2. >nobody put wheels on luggage until the 1970s.

        Inventions are made to a need. I’m sure someone put wheels on luggage before; they were only PATENTED in the early 70’s.
        “If we look again at the picture of people carrying luggage above, we notice another addition: the porter. Prior to the introduction of cheap travel, most people would have someone else to carry their luggage when it really mattered. (…) The Social Trend in this scenario is the democratization of travel. Plane tickets started dropping in price and more people started visiting airports. Few people were accustomed to paying porters to carry their luggage.”

        Think about it. An upper middle class person of the “jet set” in the 60’s would expect to take a taxi from their home door – the driver picks up the luggage – to the airport where the porters immediately load the luggage onto trolleys and carry them to the plane, and the same thing in reverse at the other end. Why would you put wheels on them when you barely have to carry your suitcase 200 feet?

        1. Another thing was, people used to travel light. A suitcase was a suitcase – some shaving equipment, a book, and a change of clothes. At the destination, you’d use the hotel laundry services. Everything else you’d rent or buy at the destination. If you had something big to move, you’d mail it to yourself because travel took time – you didn’t go across the continent for just a couple days.

          Nowadays when people travel, they bring the whole wardrobe, a table full of electronics and gear, food, drink… everything they don’t want to buy on the way because it costs more, and stuff it in a backpack or a travel case which then weighs a hundred pounds or more and needs a pair of wheels to move.

      1. So they’re playing this like they are just now achieving their goal, but actually they’re just demonstrating something they accomplished years ago, using someone who has a YouTube audience to get some exposure. Nothing wrong with that, just ever so slightly misleading, I’m okay with that because I’d never seen this before, and now I have, and it is awesome.

  3. Next up, the faster you go the more wind speed you have coming towards you. Add a fan to the front and use a backwards differential to add the speed from the wind to the speed from the wheels and spin the fan faster… Scotty might have something to say about this one.

  4. Don’t forget, what works for sound works for light.
    Piezoelectric resonance to disrupt the air so you can break the sound barrier. So gamma rays to wobble the light that builds up ahead of your space craft. Hmm. :D Guessing a radiometer isn’t going to cut it in that situation but still. :D

    1. Piezo disruptors? Old school stuff. We use an annihilating Bergenholm to cancel inertia. The only thing that limits us is the mass of the interstellar medium! I would tell you more but we are in talks with Tesla.

    2. i’m not sure “what works for sound works for light”, but even if it’s true, there’s the little thing of “relativistic velocities.” if you accelerate up to near the speed of light then you get these huge lorentz factors that cancel out further acceleration. this is the same whether you are using wind or light to aid your acceleration. it’s just that you don’t expect wind to reach relativistic velocities…

    1. But they don’t go faster than the wind downwind, so they? If the wind is going N, they might go 2x that speed ENE, but less than the wind speed N. This is an entirely different achievement by my understanding.

      1. Actually, they do go faster than the wind downwind looking only at the speed in the direction of the wind; i.e., when the yacht travels NE with the wind going N, the yacht’s speed component in the N direction exceeds the wind’s speed, outrunning the wind. See the video (and high-performance sailing on Wikipedia) where they discuss how sailboats do this routinely and how it was adapted to enable this vehicle to do the same.

      2. The thing about the modern AC yachts is they don’t have sails. The big thing sticking up in the middle is a wing! It is mounted on a very strong hinge, and using hydraulic rams you can tilt sideways, and fore & aft, and change it’s angle to the wind. You can adjust the amount of “lift” you get and hence the speed. The hydrofoils reduce the drag of the water, and also act as keels to prevent the whole thing sliding sideways. The hydraulics are operated by a bunch of big guys winding handles! The only electrical power allowed is for the nav gear, and the cameras.

        Quote: “With the weather being once again a key player today and the SW breeze gusting up to 22 knots at times, a new speed record has been broken today as American Magic hit on 53.31 knots (almost 100 km/h) on race 1.”

        From here:

        Although I used to sail competitively, I never sailed with this type of rig. However a bit over 20 years ago I tried out a friends land yacht on an airfield and got up to quite frightening speeds!

        1. Land yachts and hydrofoils need technology (wheel bearings and carbon fiber) to reach high ratios.
          Iceboats have been doing it for more than 80 years.
          Skeeters reach 84 mph, DNs, a little less.
          An Iceboat established a record in 1938, on lake Winnebago — 143 mph.

      3. They most certainly do. In fact, they didn’t bother to spinnakers in the recent America’s Cup because the downwind performance was so good. They are hydrofoils, so the conventional limits of displacement hulks do not apply when they are foiling.

  5. For those who don’t understand how it works…. The big fan is not driven by the wind, but it is the propulsion of the vehicle! Energy is extracted from the wheels and is used to push the vehicle forward against the air that is already moving in that direction. Thus when going slightly faster than the wind, a lot of energy is extracted from the wheels (going fast) for a small amount of force. But when pushing against the wind a bigger force over a much smaller distance requires less energy, so there is some energy left for friction and whatnot.

  6. But they can go faster than the wind by tacking. Maybe, with similar arrangement as described in the article, but with wheels replaced with propellers or paddlewheels, going directly downwind faster than the wind on the water would be possible?

    1. Tacking is used to travel into the wind, not with the wind, and it’s much, much slower than sailing with the wind.

      When tacking, a boat sails at a sharp angle into the wind for a distance, then turns across the wind to take the opposite angle for a distance, them back again: it’s a zig-zagging course into the wind. It lets you said into the wind, but it takes a long time and it’s horrible inefficient compared to sailing with the wind directly behind you.

      The only way a boat will sail into the wind faster than the wind is if it’s not using the wind for power.

      1. America’s Cup would like a word with you. From wikipedia:

        The catamarans used for the 2013 America’s Cup were expected to sail upwind at 1.2 times the speed of the true wind, and downwind at 1.6 times the speed of the true wind. They proved to be faster, averaging about 1.8 times the speed of the wind with peaks slightly over 2.0.

        1. American Magic’s “Patriot” (AC75, America’s cup 2021) managed a 53.31 knots top speed in a wind gusting up to 22 knots. But New Zealand’s Te Rehutai also managed to reach 49.1 knots in as low as 15 knots, and both boats VMG downwind are faster than the actual wind. They can overrun the wind as long as they keep an angle to it and keep jibing.

      1. When a boat tacks, the wind blows on the opposite side of the sail from what it was. The apparent wind against these blades is always on the same side so I wouldn’t agree that they are tacking.

        However, I would agree that they effectively transition from dead down (stopped) to broad reach to beam reach to close reach to close hauled as the vehicle picks up speed.

  7. To really understand “faster than the wind downwind” you have to consider that the air surrounding the fan as part of the vehicle. This has been done before as was pointed out, so I don’t really understand why this is in the news again. It’s’ still cool though :)

    1. I suppose one way to look at it is that it’s harvesting some of the energy of the wind pushing it forward and using that to push back against the wind.

      It harvests that energy by means of the wheels making ground contact and utilizing the air speed over the ground even when the apparent wind speed is zero. That’s why this can’t be a flying vehicle.

      So now the obvious question beckons: could you make the same vehicle as a water vehicle?

      1. I won’t say “no”, but the problem is that it is the friction between the ground and the wheels that is required to spin the fan. I don’t know how you’re going to get this on the water. But… If you have a propeller on the boat, this can drive the fan. Okay, done. The wind on the fan forces the boat forward, the forward motion in the water drives the propeller, and that drives the fan, and you have the same interactions we see in the video.

    2. No, the *really* important thing to understand is *the ground*.

      The key is that the reference frame analogy is wrong: standing still is not equal to running at the speed of the wind, because in one case the *ground* is moving and the other, it’s not. From your point of view, you’re not pulling energy from the air, you’re pulling energy from the *ground*. You’re using the air as propellant.

      Get rid of the ground and the problem doesn’t work, there’s no energy or momentum to steal. I mean, imagine it on the Moon. Get rid of the air completely, and drop a completely mechanical object out the back of a truck Knight Rider style. You can *still* hook the wheels up to a generator and gain energy that way, right? Of course conservation of energy says “no, you can’t gain back more energy than you lost” – but you can always pull rocket-tricks and use that energy to shove parts of your vehicle backwards, even past the speed you were originally going (because you’re losing mass).

      Get rid of the ground in *that* case and you’re screwed – whatever you shove out the back has no ability to do anything.

  8. So, sailing boats moving faster than the wind is impossible? Tell that to the America’s Cup, which has been travelling faster than the wind for a decade or more. The modern ones travel faster than the wind INTO the wind!

    Absolute nonsense that it’s a ‘fundamental law thought unbreakable’.

  9. In a sense, this is a little like an autogiro, where wind through the rotor produces enough lift to keep it airborne. Now all we need is an autogiro that doesn’t need a thrust engine…

  10. To anyone still thinking this shouldn’t work consider this.

    Take a windmill, slap it on a stationary vehicle. Use that windmill to charge up a battery, then stow the windmill and use the batter to power the car upwind. However fast you wish.

    Conceptually, you are harnessing the energy differential between the stationary ground and the moving wind. Then using that energy. No one would say this is voodoo, nor claim that thermodynamic principles are being violated.

    this is… kinda the same idea abstracted.

    1. I’ve had to listen to people come up with the wonderful “idea” of putting a wind turbine on an electric car, to charge the battery as it’s going along…

  11. Are we sure this is not just an effect of wind gradient? That big windmill is up ~18 feet AGL, much higher than the ribbon telltale. One would expect higher wind velicity at that height. That’s one reason why we don’t takeoff downwind. Get up a little ways, lose airspeed and you come back down.

    I bet a square sail that high would do the same thing.

      1. Then the counter question is “but how can you *accelerate* using that energy?”

        After all, this won’t work if there’s no wind – if I shove the vehicle forward, the ground is still rushing past me, so why can’t I do the same thing there? And the answer is that in *that* case, if you shove the vehicle forward, there *is* wind – a headwind. And the drag on the fan from the headwind prevents that from working.

        The key is that when there *is* wind, and you’re moving with it, the air around you is still but the ground is moving. So the fan by itself produces no drag, and the wheels allow you to use the fan to scoop up the air and throw it backwards for propellant.

          1. It’s not oscillating. The math and physics here isn’t particularly hard.

            Consider steady state. The wheel’s generating power equal to the rolling resistance (Fr) times the speed of the vehicle (V), which is greater than windspeed (W, where W less than V). Consider the prop: the *airspeed* it’s seeing is V-W. So the power it’s producing is equal to the thrust it’s generating (call it Fp), times the velocity of the air through it, which is (V-W). The power it’s *consuming* is just that, divided by its efficiency.

            Ignoring the prop efficiency for the moment, that means the power *available* is Fr*V, and the power *consumed* is Fp*(V-W). That’s the key. It’s *easier to spin the fan in the wind*.

            In order to figure out the steady-state velocity, you need to add efficiencies here (believe it or not, with perfect efficiency the max velocity is unbounded – which isn’t that surprising since the available energy is, as well). So if we add in just a “pure” efficiency term folding in prop efficiency, gearing efficiency, etc., then we have:

            Input power to shaft = Ft*V
            Output power to prop = Fp*(V-W)/efficiency

            Setting them equal, and setting steady state (no net force, Ft=Fp), we get V = W/(1-efficiency). No oscillation. How could there be? This is steady state – no net force, with V > W.

            Note that this is just power transfer efficiency, so even if efficiency = 0, steady state is V=W. In this case the prop isn’t spinning (but the air is still, so no drag), but the reason there’s no loss is because there’s only an efficiency on the *output*. Putting an efficiency on the *input* allows steady state to be zero if the efficiency’s zero (since there would be no other way to balance the two).

            There are more recent papers on this that similarly cover all efficiencies in detail: see, for instance, “Analysis of down-wind propeller vehicle”, a paper from 2013 which has a good history of the discussions on this.

        1. They don’t really cover if this is sustainable, or temporary… Simplified(?), is this much different than regenerative braking, using the energy to power a fan? – That’s where I have a very hard time seeing energy pulled from braking (drag against ground speed) having any chance, even with 100% efficiencies, giving you more thrust from a propeller than you drew off the wheels. IE if you pull 100 watts off the wheels (100 watts of drag), you only have 100 watts to use as thrust (with a propeller no less). Maybe I just haven’t wrapped my head around it yet, but I can see this much more easily if you were in this example charging a battery with regen braking for a while (drag against wind, but still forward momentum), and then dump the energy to push faster than wind. Seems maybe same effect here due to variable pitch blades – you can coast/spin it up for a while, storing energy essentially in flywheel of wheels/prop/(maybe even a flywheel internally not shown?), and then change blade pitch to dump stored energy to push you forward faster tham the wind, but then this is not sustainable…

          1. It’s sustained.

            The reason i5s possible is because the generating medium (ground) is at a different speed than the thrust medium (air).

            Imagine your fan is generating 10 Pa of pressure over 1 square meter. How much power is it consuming? Depends on the speed of the air! If the air is moving at 1 m/s, it’s 10 W.

            Likewise, if a wheel is moving and you put a load of 10 N on it, how much power do you get? Depends on the speed of the wheel. If it’s at 10 m/s, it’s 100 W. Yes, it’s weird to think about the *linear* speed of the wheel rather than *rotational*, but they’re coupled in a rolling tire with no slip.

            So the gap between the two speeds (wheel, at V, and air, at V-W) gives you power available for use. The reason it’s even *hard* is that prop efficiencies are limited. Mechanically wheels can be *very* efficient.

            Think about it this way. Imagine a vehicle in a tube, where a track on the ceiling was moving constantly and the ground was fixed. Could you constantly accelerate? Hard to picture? Then decouple the track: now imagine a vehicle on flat ground with a treadmill on it that is constantly running. Now it’s easy to see you can accelerate forever, and it *doesn’t matter* what the treadmill speed or direction is. That’s just gearing.

      2. It’s a little like the old argument that a jet- or rocket-powered craft can never go faster than its exhaust velocity. This craft can still accelerate when its moving faster than the surrounding air, because it generates a high pressure area behind the propeller. Somehow. Still not quite sure how this isn’t like picking yourself up by your own bootstraps.

        1. Actually, when you consider a rocket’s acceleration, if you look at its exhaust plume after a certain speed it is moving along the same path of the rocket, only slower than it. Just because you reached a set speed you don’t stop having a reaction when pushing stuff backwards. Consider two ice skaters moving in unison. At a certain point one of the skaters pushes on the other to gain extra speed. The other skater loses speed, but keeps on travelling in the same direction. This is how a rocket accelerates even though it is going faster than its exhaust speed.

          1. I guess I was unclear. I meant that the notion that a sailing vessel can’t go faster downwind than the wind itself is LIKE those other notions – widely believed but false.

          2. Many *sailors* actually think it’s impossible for an unpowered boat to go directly downwind faster than the wind. Which is true for all the cases *they* care about, so I don’t blame them. You usually learn pretty early in lessons on sailing, for instance, that sailing aligned with the wind is bad.

            Fair number of websites and books, for instance, state it as absolute. Others have it as a footnote/addendum pointing out that it’s only true if you’re purely limited to a sail.

          3. “You usually learn pretty early in lessons on sailing, for instance, that sailing aligned with the wind is bad.”

            I guess that’s why you could never sell anybody a sail made for doing nothing but that.

  12. The propeller disk stores energy like a flywheel. It is possible to use this energy to go faster than the wind for brief periods. However, the propeller will begin to slow down once the relative wind across it goes to zero or less. Steady state velocity for the system is slightly less than wind speed. However, the test setup doesn’t allow the vehicle to reach steady state due to variable winds and limited travel.

    You have to do this in a wind tunnel, but no one seems interested in that.

    1. They effectively did that first, with the miniature on the treadmill. The important thing is that the wind (still air) and the ground (moving surface of the belt) were moving relative to each other.

    2. Agree, the experiment is far from providing definitive proof.
      But the idea is remarkable and should work provided friction is low enough.
      It’s a gearing problem that trades power for speed.
      This will spawn a new sport, I expect professional races of such contraptions on the salt lakes in the near future.

    3. In the video, they show tests with a scale vehicle on a treadmill. Their argument is that this is the same as a wind tunnel – the air and ground are moving at different speeds. From the vehicle’s point of view, there is no difference between the two.

      1. The test setup is needlessly complex. They need a freewheeling prop on a load cell in a wind tunnel.

        I feel the test setup is complex because they are trying to create a mystery where there is none.

    4. Nope, that’s not what they are doing. They are extracting energy from the speed difference between the wind and the ground. They are effectively gearing the chassis in such a way that it is pushing back the plane of the “virtual sail” that is the prop, fast enough that the chassis is moving faster than the wind relative to the ground. It is a steady state, not a momentary acceleration.

      1. If the relative wind speed is zero or less, the prop is expending energy. The prop will only expend energy until it’s kinetic energy is gone.

        The prop will slow as it expends energy. As it slows, the relative wind will become positive and accelerate the vehicle.

        The system will oscillate about a zero relative wind speed.

        1. The wind is pushing on the prop. the prop is pushing on the chassis, assume right now that the prop is not spinning. The chassis being pushed forward by the wind rolls on its wheels, causing them to turn. The wheels are geared in such a way that causes the prop to turn. If you were to slice the entire contraption vertically you would see that the prop section (that the wind is pushing onto) is moving backwards relative to the chassis. So the wind is pushing on a sail that is moving backwards relative to the chassis, because the sail is anchored to the ground. If you had a tall pole with a sail attached near the bottom and the wind pushed on the sail and caused it to topple it over, would you be surprised if the tip of the pole was moving faster than the wind?. This device essentially creates an ever toppling pole.

        2. “If the relative wind speed is zero or less, the prop is expending energy.”

          If the relative wind speed is zero, that prop sure as hell ain’t expending energy, because neither it nor the air are moving.

          That’s the key. It’s only consuming power on the *relative* wind speed, whereas the wheels are generating power from the *total* speed. Couple the two, and steady state is *above* wind speed. It’s not oscillating.

          Just imagine the ideal case: no rolling resistance until a load is on the shaft, no gearing losses, wind at 10 m/s, vehicle at 10 m/s. So no drag on the prop. Put a load on the shaft of 10 N, and you’re generating 100 W. Shove that power into the prop, and if you can generate more than 10 N of forward thrust in *still air* you *speed up*. And obviously that’s just a question of the prop geometry.

          Since the ideal case is obvious, the rest is just a question of efficiencies.

  13. Anyone watching carefully will notice that the fan effect produced by the propeller is actually reducing the air speed in front of the vehicle. Therefore, using a streamer in front of the vehicle as a proof of exceeding the wind speed is not really any all. To use the streamer as a proof, it would have to be installed such that it is out of the influence of the fan effect, for example well above the height of the propeller, thereby being within the actual wind speed.

    1. Maybe I’m wrong, but rather think that once going faster than the wind, the streamer is going through the air before the fan gets to it, so if anything, adds a wee bit of drag and slows it down. However, the fan is sucking air from the direction of the streamer and shoving it out the back, so might to some degree cause an exaggeration or slightly early appearance of going through still air. Quite far apart and fan is much higher up, so also maybe not.

      Is the propeller high enough to have slightly higher wind speed than the streamer? Is the world flat? We shall never know.

    2. I agree that the wind gradient and possibly the rotating blades could have had some slight affect on the streamer. But I also think they used the streamer because it is simple and easy to show on video.

      A mechanical speedometer would be better but then people might question its calibration. They could have used GPS but GPS isn’t perfect either, though on average it should be pretty good if the successful portion of the experiment lasts long enough to get reasonable averages. They could have measured the speed of the vehicle (where it is expected to attain the relevant speed) based on time recorded over a measured distance but that would leave the question of the average windspeed as experienced by the moving vehicle.

      I’m sure the wizards could figure this out for official runs especially if the results are dramatic enough such as nearly three times the true windspeed as I understand the official runs were.

      For this video, I think they just wanted to show something simple to get people thinking.

  14. How about this one: if it were impossible for a sail boat to sail faster than the wind, then powered, heavier than air aircraft would notbe able to fly…….. It is just as simple as ‘Aerodynamics 101’.

    1. The boat analogy is kinda misleading here. The car is flowing downstream (in the wind) at about up to the wind speed. It’s then using the wheels’ grip on the road to turn the turbine, which can push it along faster than the average wind.

  15. I remember reading about this about 10 years ago and the illustration of the two yachts on a cylindrical earth made perfect sense to me. I even bought a model off the net that I have somewhere.

    Another similarity I noticed was to a thread cutting die on a rod. The prop blades are like the thread teeth in the die. A small force needs to be given to the rear of the die for the teeth to bite into the material in addition to the rotational force. You could think of the rod being the air in front of the vehicle. Once the teeth are engaged on the rod, only rotational force is needed to advance the die forward along the length of the rod.

    But it’s almost 1am and I need sleep.

  16. Ice boats have been reaching speeds 2 to 3x the speed of the wind for decades so this is not some new concept. In laymens’ terms, the push effect of the wind would dictate a boat can never exceed the speed of the wind and then only in a downwind direction and could never sail upwind against the downwind push of the wind.
    The lift effect of the wind around curved sails and wing forms makes upwind sailing and higher than windspeed sailing possible so the fastest point of sail is at right angles to the wind. Fast sailboats don’t get pushed, they get lifted. Modern sailboats tack downwind because they can cover the greater distance faster than if they sailed the shorter and slower dead downwind course.

  17. This is my favorite physics question to date. The pure number of people in the “oh yeah, this is so obvious, why are we talking about it” camp and the “no way, they did thing [x] wrong, this is impossible ” camp is just so entertaining. Imma grab some more popcorn and keep reading the comments.

  18. Catamarans can accumulate net speed and then change direction going short period of time really faster than wind. Similar thing here with this trike. It accumulates kinetic energy to its big windmill then shift gear to break speed record… but it is not sustainable.

    1. I suspect minor course changes to right or left of dead down wind are creating enough wind lift to increase speed above maximum wind push speed when DDW. You’d need some decent instrumentation to measure for that (not just a freakin tell tale). Sailboat instrumentation measures actual speed over ground, direction over ground, actual wind speed and wind direction and then calculates apparent and true wind angle and speed. Reduced to graph format, accurate answers could follow. You have to begin with decent measurement.

    2. Nope. Steady state is moving faster than the wind. It’s just taking advantage of the fact that you can get power from one medium and impart it to another, and there’s a relative speed difference between the two. Takes less power to push the air, because it’s moving slower.

  19. So with this one should also be able to build the opposite as a corollary proof, and it may be easier to achieve a steady state to study it. Should be possible to make a ‘boat’ that sails upstream using a propeller that pushes it into the water stream, and also reverse that to make one that uses the water stream to into the wind.

    1. I doubt that would work for a vessel on the water because of considerable skin drag but I do agree the idea should be tested on a land vehicle with low rolling-resistance similar to this experiment. If the wind can be harnessed to generate power, then that power could be harnessed to drive a vehicle against the wind, although the gear ratio would need to be very low and move the vehicle very slowly. It might work. I proposed the same experiment below. The blades would need to generate more power than the forces acting on the vehicle to push it backward.

      The blades get their power from lift which is a more efficient use of the energy in a given area of wind as compared to the drag forces acting on the vehicle. In that sense, it would be the same principle as this experiment, though the application would be a bit different.

      I hope they attempt this experiment.

  20. This sounds like one of those ideas that wouldn’t work because it seems to generate more energy than it receives, breaking a fundamental law in physics. However, the law is not broken in this case because the amount of energy consumed is not more than what is available. Proving that a vehicle can move faster than its source of energy is not the same as proving that energy has been harvested more efficiently. This is true in this experiment and for any efficient sailing craft.

    The law is not about the rate of movement but about the energy available and the energy required for the movement. If one were to measure the energy available in the wind that was used for this experiment, there is no doubt that it would exceed the energy required to push the vehicle at the speed attained.

    In sailing, there are two basic ways to utilize the wind as a source of power. The first way is to use simple drag where a vehicle hitches a ride on the moving air. In this case, the vehicle cannot move faster than the wind, or even as fast as the wind given expected resistance to motion.

    The second way to get energy out of the wind is to redirect some of the wind creating lift, much like an airplane wing. By crossing the wind at an angle a sail can drive a vessel or vehicle faster than the wind itself. A sail operating this way does not create additional energy but it does make better use of the wind’s available energy.

    The blades on this propeller act like sails. Sometimes they use drag to capture the wind’s energy and sometimes they use a combination of drag and lift. Sometimes it is just lift that is used. Once this test vehicle reaches the speed of the true wind, lift is the only forward source of energy. The fact that the blades are turning because of the motion of the vehicle is not reason to disqualify the comparison to a sail.

    The comparison should be between the apparent wind as it passes over the sail or over the rotating blades. Imagine each blade as a sail. At rest, the sails operate like they are on a dead run. The power comes just from the wind pushing on them, drag. When the vehicle moves the blades turn. Each sail (blade) is now crossing the true wind at a slow speed but enough to alter the apparent wind direction over the blades. Power will still mostly come from drag but some lift will develop. As the vehicle speed and rotation rate continue to increase the apparent wind over the blades moves further forward creating more lift and less drag. Once vehicle speed reaches that of the true wind, the apparent wind over the blades will use only lift for forward motion. Lift is far more efficient than drag and utilizes far more of the wind’s energy. The rolling resistance of a vehicle will generally be linear so any increased use of available energy will be translated into higher speeds.

    Also, as the vehicle speed increases and surpasses the true wind speed, the wind’s drag on other parts of the vehicle will change from supporting movement to restricting movement so only so much additional speed can be attained this way. I suppose a similar concept machine could be perfected to exceed the speed of the wind by a significant multiple based on recorded downwind speeds of highly efficient sailboats.

    The idea that a sailing vehicle exceeds the true wind speed is not new. Sailboats do it all the time. This is just another example, though a little more thought provoking.

  21. I cannot resolve (in favor of or against) the issue of “how” the Blackbird goes faster than the wind propelling it. Does anyone know of any mathematical explanation? Has the small model treadmill experiment been reproduced by the kids at MIT or CalTech? That would go a long way toward convincing me.

    At about 12:20 in the video, there is an apparent “powered event” wherein the Blackbird’s propeller mysteriously accelerates from about 1 revolution per second to about 3 revolution per second. The craft is all but stopped when the acceleration begins. The “tell-tail” at the front of the craft shows the true wind direction. So, we may conclude that the prop acceleration is not due to that wind since it would cause the propeller to slow down … not speed up (in accordance with the experimenter’s design). It looks to me like a motor is driving the propeller and the experiment is about to be “undone” when the prop is abruptly halted (by some manual emergency stop mechanism not shown in the video). Notice also how the prop is going so fast as to cause the Blackbird to “lurch forward” a few feet as if being propelled by a motorized fan against the brakes which were applied prior to the propeller acceleration.

    The logic of the cylindrical world with wind driving two sailboats (at ~8:20) implies that if we run the Blackbird with free wheels and simply let the sail freely rotate … we’d also get faster than wind speeds. Why not attempt this as well? All that would be necessary would be to disconnect the drive mechanism between the drive wheels and the prop (that causes the reverse propeller, working as a “fan” rather than as a “windmill”) and let all … (wheels and propeller) … rotate freely.

    On the experimenter’s YouTube site, there are several videos of the treadmill-model experiments that appear to conclusively demonstrate the effect proposed. However, Penn-Teller could fake such things if motivated (by film-treadmill reversals, small magnets, magician’s thread, reverse speech mouthing with forward audio, collimated air streams, etc. … this is why it is important for the kids at MIT or CalTech to reproduce these results. They wouldn’t fake it because physics is to be their life’s work. So, reproducibility is essential.

    1. The wheels are geared to the propeller. What you’re seeing is the car being pushed by the wind, which turns the wheels against the ground, which turns the propeller faster, to push harder against the wind, which turns the wheels even faster…

      I’m still not convinced that the “cylindrical world” thing isn’t a red herring — it seems to explain how propellers work in terms of sailboats. None of this has to do with wheel-driven-propellers.

      1. What I meant about 12:20 the “powered event” is that the prop is increasing its rotation rate without any apparent cause. The wheels are hardly turning at all and the following wind would cause a decrease in rotation rate.
        It is unfortunate that the creators of the Blackbird did not see fit to include a schematic diagram of their build in the video. A three second screen shot would be enough to stop the video and examine their design. And … why not site other people’s models that verified their model experiments? These seem like obvious additions to the video that would not require more than an extra 10 seconds appended to the end of the video.

    2. The apparent powered event at 12:20 was a mechanical failure. I suppose there was a break in the chain or belt drive between the wheels and the propeller and the wind started driving the free-wheeling blades rather than the wheels driving the blades.

      The telltale at the front indicates the apparent wind, not the true wind. The apparent wind is the same as the true wind only when the vehicle is stopped.

      When the contraption is working properly, the wind doesn’t drive the propeller directly. The wind drives the car which drives the propeller in the opposite direction to what the wind would cause if the propeller were free-wheeling.

      If the propeller were free-wheeling it would not add forward drive to the vehicle because like any other part of the vehicle the propeller would “see” no wind on the blades when the vehicle and true wind speeds are the same. The propeller must be driven such that the rotation is against the flow of the true wind thereby creating its own wind at and beyond the true windspeed.

      The only way an unpowered free-wheeling sailing vehicle could surpass the true windspeed would be to sail downwind at an angle to the true wind such that it generates its own apparent wind like a good sailboat or an iceboat. Under this situation the sails are trimmed close as though they were sailing upwind, because they are sailing toward the apparent wind at an angle.

      Perhaps one thing that might not have been mentioned in these discussions is the fact that sailboats do not sail in true wind. They sail only in apparent wind. As soon as a sailboat or sailing vehicle moves, it is sailing in apparent wind. The only forces that matter are those that the sail “feels”. Sailing directly downwind at six knots in a 10 knot breeze will offer the sails just four knots of wind pushing on the sail. Sailing at ten knots at 90 degrees to a true wind of ten knots would result in an apparent wind of 14 knots at 45 degrees from the bow. Sailing at 20 knots at 135 degrees away from the true wind will also result in about 14 knots of apparent wind but with the apparent wind closer to the bow, perhaps 30 degrees or so (I didn’t do a calculation so this 30 degrees is just an estimate). It is this apparent wind that drives the sails on a sailing craft, not the true wind.

      In the case of this contraception, the blades can be thought of as sails. Think about the apparent wind that each blade “feels” as the propeller rotates against the will of the wind. Initially, at zero or very low vehicle speeds, the wind over the stopped or slow rotating blades causes only drag but that helps get the vehicle moving and the blades turning. With the blades moving, the apparent wind over the blades changes both in angle of attack and in speed. This change of the apparent wind to a narrower and narrower angle of attack and higher and higher speeds as the vehicle moves faster away from the true wind results in less and less drag but more and more lift. Lift is far more efficient than drag and this is how this vehicle can surpass the true wind speed directly downwind. The vehicle moves directly downwind in terms of the true wind but the blades operate in apparent winds at higher speeds and at narrower angles of attack.

      This will only work if the propeller is driven such that the blades rotate in the opposite direction to that which a free-wheeling propeller would rotate. Using the motion of the vehicle is a way to drive the propeller without resorting to a separate power source.

      It is also worth noting that the power taken from the wind is not more than what is available from the wind. The assumption might be that windspeed equals power. Power is of course proportional to windspeed but area is also proportional to power. There is 100 times more power in a 10 x 10 foot square area of a ten knot wind than there is in a 1 x 1 square foot area of the same ten knot wind. A larger sail obviously catches more wind and therefore more power. Using only as drag from the wind, it wouldn’t matter how large or small a sail is, it could not push a vehicle directly downwind faster than the true wind speed. So this experiment is not proving a failure of a law of physics. Rather, it is proving that there are more efficient ways of harnessing the wind as an energy source than to drift with it.

      Proving it from a trusted source would be good of course. But the physics is understandable. I just hope the above explanation is somehow helpful.

  22. I liked the treadmill concept. Here’s a mental image, “Place it on the treadmill (zero wind) and start the treads, soon it will overrun the far end, right??” Sorry, I’m just not buying it. Somebody added energy to the system. RR

  23. Sailboats sail faster directly downwind than the wind itself by jibing and a vehicle has been developed to sail directly downwind faster than the wind itself without jibing.

    Sailboats also sail directly to windward by tacking. The next challenge then should be to develop a vehicle to sail directly into the wind without tacking.

    The use of the wind for propulsion needs to be more efficient than the wind resisting the movement. A wind driven propeller would generate power from lift while resistance to movement of the vehicle would come from drag. We know lift is more efficient than drag.

    It might even be a similar vehicle with different blades and geared to move slowly and backward toward the wind.


  24. I don’t understand this website.

    At first I thought this was really interesting and new. But it turns out that this experiment or one exactly the same was conducted officially way back in 2010. The vehicle, in 2010, managed to travel directly downwind at almost three times the speed of the true wind.

    Further, either the same vehicle modified or a similar new vehicle two years later in 2012 proved that a vehicle could sail directly into the wind at more than twice the speed of the true wind while powered only by the wind itself.

    All of this was officially recorded.

    There is no debate here.

    So why are we having this discussion?

    And specifically, why now?

    1. Well, probably because most of the people here never heard about it. Why are we having this discussion now? Because someone brought it up, now. It’s really not that hard to understand.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.