This is a really good explanation I saw a while back of the phenomenon. This debunker (a PhD @ Google) originally said it was impossible - and then realized he was wrong.
Thanks! That was far more readable than the original article. I had dismissed the idea for exactly the same reasons as the author, and now understand how I was wrong about it. Glad to have learned something today.
Thank you! That link is much clearer than the article.
One thing wasn't in it that I found cleared this up a lot, though: with a normal sail, wind that gets caught still moves forward relative to the ground, just slower. With this thing, it winds up going backwards. That supplies the extra needed momentum.
What makes me really happy is that when me and my roommate discussed how this worked months ago, we eventually explained it exactly as he did in less-equation-y form.
Moral of the story: I don't fail completely at fluid dynamics.
Several comments in this thread say something to the effect of "sailboats do this all the time" or "sailors have known this for years."
Sailboats do not sail faster than the wind when sailing dead downwind. High-performance sailboats absolutely do sail faster than the wind when sailing upwind or on a reach. This is a result of the apparent wind velocity (the velocity measured on the boat) increasing as the boat accelerates into the wind. This apparent wind is what powers the sails -- more apparent wind, higher boat speed. Terry Tao's explanation is excellent: http://terrytao.wordpress.com/2009/03/23/sailing-into-the-wi...
A recent real-world example is the boat that won the America's Cup this time around, Oracle's 90-foot trimaran BOR 90. BOR 90 handily sails several times faster than the wind upwind. Watch chase boats try and keep in a mere 10 or 12 knots of wind as the boat is going 32 knots: http://en.wikipedia.org/wiki/USA_17_(yacht)
Downwind is a totally different story. As you approach dead-downwind you accelerate with the wind until your boat speed reaches the speed of the wind and your apparent wind approaches zero.
As for sailors knowing about the effect, I can assure you it is just as counter-intuitive to us as anyone else. Maybe more so since we spend a lot of time observing how boats behave downwind. I've raced in San Francisco for more than a decade and had this discussion with maybe a dozen racing sailors in the last few months. Things got heated every time. No intuition to be found.
[Edit: corrected sloppy about dead-downwind vs just downwind. As commenters have noted, sailboats will happily go faster than the wind at angles below 90 degrees, but not at 180 degrees.]
Nice links, thanks. I respectfully disagree with the upwind/downwind speeds though. In my experience the most force on the sailboat/windsurfer (and resulting speed of the boat) comes from sailing at a broad reach, roughly 45% downwind. Linked from your USA_17 article:
One's intuition works if you think about it like a heavy brick pressing on a slippery wedge shaped object between the brick and the floor. The wedge will "squirt" out the side at several times the velocity the brick is moving towards the floor.
Sailboats can't sail directly downwind (say, within 5 deg. of directly downwind) faster than the wind. However, they can and do travel downwind with net speeds faster than the wind. Indeed, "speed made good" directly downwind for sailboats and iceboats has been measured at several times that of wind speed. This has been recorded in the America's cup race, for example.
So this result shouldn't really be that unexpected. Sailboats can't sail directly downwind faster than the wind, but they can travel downwind faster than the wind.
To expand on this slightly - sailboats can "make good" downwind at higher than windspeed by reaching at an angle to the wind, then gybing back the other way. If the boat can reach at a high enough speed then they can "make good" the extra distance.
[EDIT: I started typing this up before smanek's post. It's essentially the same argument, but I intentionally left out the formulae.]
Really cool. Here's my best explanation of how they achieve a stable speed faster than the wind; this isn't explained that well in the article. (Disclaimer: I'm not exactly a trained physicist/aerodynamics expert.)
They've built a lightweight vehicle with a huge propeller directly linked to the drive shaft ("directly linked" is important here!) With the vehicle at rest, the propeller catches wind and begins driving the shaft; the car itself also catches some wind. Obviously, the car speeds up.
Once the whole contraption is at or near wind speed, the wind obviously stops driving the propeller. It's clear that the vehicle can maintain this speed just by using itself as a sail.
Now, note that the wheels, directly linked to the propeller, are now driving the propeller. Running through stationary air (since the vehicle runs at wind speed, the air is stationary relative to the propeller), the propeller begins accelerating the vehicle!
Of course, as the whole contraption accelerates beyond wind speed, air and ground friction increases, etc. But all of these effects are gradual, and since we have already seen that it can, in fact, accelerate while traveling at wind speed, it will continue accelerating for a while beyond that, eventually reaching some maximum speed above the actual wind speed. Apparently, this is ~2.8x the wind speed for this particular vehicle, which is an impressive engineering feat.
The article says that this offers opportunities for power generation, but I'm not seeing those. Still, very cool.
What I think you're missing is that the wind never drives the propeller - the wheels drive the propeller.
When it first starts moving, the propeller moves the opposite direction from what you would expect - so that the propeller is actually pushing back against the wind from t=0.
What I don't get is why people thought this was some heretical nonsense? It makes total sense, and is kind of obvious when you actually consider the design. A car with a big sail on it would only be powered by the wind and probably not exceed the wind speed, because nothing else is driving it. But it was clear that this thing had two power sources: the force of the wind moving it forward, and the force of the prop pushing the air backward (that may sound totally retarded, I know nothing of physics) and propelling itself forward. It's as if people just stopped thinking once they hit the conundrum part of it and purely started trying to form reasons why it could not work, and didn't really question or analyze it with the possibility of them being wrong.
Are you sure that is correct, that the wheels are powering the propeller at some point? The article explained it completely differently. The article said that the effect is similar to when a sailboat tacks downwind sharply, they can briefly accelerate beyond wind speed. The propeller is essentially shaped as if each side is a sail continuously tacking, so the force from each of the two "sails" drives the propeller in a circular motion.
I don't believe the wheels are providing any energy into the process, but I could be wrong. I think it is the shape of the propeller that causes the effect. The article also said it could most likely work on water as well.
The wind energy applications of this propeller design are intriguing.
The interesting part is not that they can briefly go faster than the wind, but rather that they can keep this speed up. It's easy to briefly go faster than the wind: use some gears to connect a windmill to a heavy piece of metal rotating on the end of a long pole. Assuming a strong wind, low-friction gears, and that you've chosen the gears such that the metal rotates faster than the windmill, the machinery will soon be spinning quite rapidly. Now, just put a hapless pig in the way of the rotating piece of metal, and you suddenly have a slightly-dented pig flying at greater-than-wind speed. ("With sufficient thrust, pigs fly just fine", although this will definitely annoy the pig.)
I'm sure you can think of other ways to store the wind's energy and use it for a speed boost; the interesting thing is that this vehicle can consistently outrun the wind.
Also note that my explanation does not require the vehicle to be on the ground instead of in the water - think "water wheel".
With respect to "tacking": you'll find that Chu-Caroll's article, which smanek linked to, provides the same explanation as I did (skip the initial part where he explains his error and start reading after "let's look at a new diagram"), and mentions the similarity to "tacking" (start at "Once it's moving, what's going on can be described in a bunch of different ways.")
However, note that not even the "perfect" sail that Terence Tao assumes in the article linked above achieves more than 2x wind speed; clearly, there is something about this real-world vehicle that makes it move faster than even theoretical (fixed-)sail-powered cars. I think my comment above is a plausible explanation for how this can happen.
Yes, the wheels drive the propeller directly. Read the articles posted in the first comment. They will explain it. Very short summary: The prop creates an air cushion that acts as a virtual sail.
It's the difference between the wind speed and the ground speed that's driving the vehicle. The wheels are always driving the propeller against the wind; and when the vehicle is travelling at wind speed (so that relatively speaking, the wind should be dead air), the propeller is pushing against the wind, making it go faster.
This seems really interesting but I found the article almost unbearable to read - it seems like the author's train of thought directly transcribed. The project's website is http://www.fasterthanthewind.org/ though I'm still looking for a good explanation on how it actually works.
EDIT: do follow the link in smanek's comment. It's good. I think I've understood it from that explanation.
It seems less paradoxical when you think in terms of energy instead of speed. In other words, thinking about it as (1) drawing energy from the wind and transforming this energy into forward physical motion instead of (2) the wind pushing something faster than itself.
Yes, this is why the concept of energy is given such a prominent position in physics. It is a great tool for predicting what might happen even when the detailed mechanism is unknown or seemingly impossible.
I haven't yet taken the time to understand the details of these explanations, but when I heard about this cart I immediately thought "hmm, the wind is blowing, there's energy all around, so some tricky person can probably figure out how to make use of that energy; hence, it's plausible".
Of course, the flip side of this principle is that physicists can be driven to spend years and years and years trying to extract energy that is obviously there, but that has no known mechanism for getting it out. The classic example is nuclear fusion. Everybody knows it's energetically favorable; nobody knows how to get over or under the activation barrier in any useful way, other than "build something the size of a star".
Do you have a filing system, or just a ridiculously good memory? This is hardly the first time you've provided a list of similar or identical submissions...
Doing this has often led me to articles of interest I would otherwise not have found, and I believe that cross-referencing like this is potentially of use, a little now, potentially more later.
However, with a karma of nearly 14k but an average of 2.29 it's clear that such cross-referencing rarely gets upvotes. A quick check of my comments shows that:
Net result is that I'm probably mostly wasting my time. More than once I've considered stopping, more than once my mild obsessive/compulsive tendencies have caused me to carry on. I am again considering stopping and using HN differently. My "contributions" are rarely upvoted much, and that suggests my interests aren't that good a fit for the current population.
It's not that I care so much - I don't really care about the karma - it's just that it's a source of information about the community and my actions within it. Karma is intended to reinforce "good behavior" and discourage "bad behavior" and so I pay attention to it.
Not least, as I'm running a closed community I'm paying very close attention to the effect, and to the discussions that occasionally pop up.
> I have the hacker friends extension ... set
> to highlight all of your comments and submissions.
For better or worse, karma has little to do with good. http://news.ycombinator.com/item?id=1872693 was near the top, mixed snark and a (decent) reference, cost maybe 30 seconds to type, and earned 36 upvotes as of this post.
I'm not ashamed of the first comment, but it saddens me to see that the second got so far fewer votes... and this seems to be the case for many other comment(s/ors) as well.
For what it's worth, I know your username and read your comments with interest.
Let C be a cylinder positioned at the center of the propeller with the same radius as the propeller and whose length = the velocity of the wind. Based on the density of the air and the volume of air contained in the cylinder and velocity of the wind we can calculate the total momentum available. There are theorems on how efficiently you can extract this energy - based on propeller theory. let this number be defined as "eta". if M * V * eta is the momentum available and the mass of the vehicle is less that M * eta where M * V * eta is the momentum extracted i.e.
m < M * eta - then the velocity of the vehicle "v" must be greater than V the velocity of the wind. If you notice the size of the propeller in the picture - it is huge - so that the mass of the wind it intercepts > than the mass of the vehicle
The sailing against the wind is not an intuitive, mostly because there is no simple mathematical formula for the underlying physics. However one can do some convincing models.
I have some matlab code which can, probably in the simplest way, demonstrate theoretically the principle. In short, if it is possible do sail down the wind faster the the wind than by having rotating sail (blades) DDWFTTW is possible. See: http://goo.gl/TtEfM
I don't see why anyone would think this to be impossible. Neglecting friction, a vehicle can travel at an arbitrarily high speed without requiring energy input. Proof by gedanken construction: construct a vehicle with an arbitrarily large collapsible wind turbine. Generate and store electricity while at rest, for say 1 minute. Collapse & store wind turbine, and use stored electricity to drive at ludicrous speed for 9 minutes. QED.
Like most people my gut reaction was "that's not possible" and in fact I believe that depending on how you define "sailing dead downwind" it may or may not be possible.
It is clear that the angle of attack of the sails is not at 90 degrees to the wind therefore strictly speaking perhaps this is not sailing dead downwind.
It is well known that sail driven vehicles can greatly exceed actual wind speed due to "apparent wind" effects among others.
It is not sailing in any conventional since. The purpose of the vehicle is to demonstrate whether any wind powered vehicle is capable of traveling directly downwind faster than the wind. Many people believed, or still believe, that it's impossible.
It's been long since the sailors out there know that fact. Ships create their own wind when close hauled. Check out http://en.wikipedia.org/wiki/Hydropt%C3%A8re for a particularly striking evidence!
The point that made it counter-intuitive, here, is the "under the wind" part.
The (false) idea is, if a 20km/h wind pushes your back, once you're close to 20km/h, you've got almost zero relative wind, so you shouldn't be able to accelerate further.
Boats, including the hydroptere (which is special because of vertical lifting, but behaves as a normal sailboat in the horizontal plane), can go faster than the wind when then wind comes from the front, not when it comes from the back.
The essential difference is, boats use water resistance sideways, i.e. they tend to behave as if they were on railtracks. The demonstrator in the article exploits rolling resistance, which boats try to minimize.
> [Boats] can go faster than the wind when then wind comes from the front, not when it comes from the back.
True, but it may be worth of mentioning that no regular sailboat (i.e. without underwater propulsion) can go directly to upwind (0 degrees). For modern sailboats, the maximum upwind direction is roughly 45 degrees, the sidewind (90 degrees) being the most optimal.
Yes, they create their own wind downwind with a propeller (+ wheels and ground resistance) where a sailboat creates her own wind with her speed and water resistance / aerodynamics (instead of mechanics) but it's the same principle underneath: use your own speed to create your own wind (that will push you).
Apparently sailboats do this all the time. Who knew? I can only assume they somehow use tacking techniques to travel against the (relative) wind. Though it's not clear to me how they accelerate past the wind speed in the first place.
I've seen an explanation [1] that in water it depends on having two aerofoils — one in the air and one in the water (they should be called fluidfoils. I guess the one in the water, the keel, is acting as a hydrofoil.) At any rate, in theory, with two fluidfoils and fluids with any different velocities you can go at any velocity (i.e. in any direction, at any speed.)
Yeah, they really do. I was looking into this a couple months ago when a friend at work, who has a sailboat, told me that they could do that, and that they could sail with a headwind!
I've noticed a nasty trend where people use the word "Impossible" as another word for difficult, time consuming or hard.
In order to prove something impossible you have to do two things, a) have a complete mastery over every property and tool that can exist and b) attempt to apply or implement every tool or property to the problem in every sequence.
In this case though, nobody was using it to mean "hard". There really were PhD Physicists arguing at length that it was physically impossible to do this.
No one ever argued that it was impossible to use wind energy to move faster than the wind. The real problem here was describing the effect as sailing directly downwind. The "sails" (on the prop) do not go directly downwind, and there's stuff going on that isn't best described as "sailing". But by describing it that way, people managed to extract blustery pronouncements and explanations of why it was impossible (which sailing downwind faster than the wind is).
You can prove impossibility. It's easy to see that you cannot sort any arbitrary sequence of N arbitrary items in less than N elementary operations, for instance.
That said, yes, it's use as a shorthand is misleading and annoying.
Colloquially, "impossible" = "I don't know how to do it", "trivial" = "I know how to do it". These two terms have very little relation to whether something is actually impossible or trivial, but are a pretty useful shorthand for "Get someone else to do this" vs. "I can do this". Most good managers understand this, and will push you to do impossible things anyway while padding your estimates for trivial things by a factor of 10 or so.
A corollary is that if you want to expand your skills, work on impossible problems. A second corollary is that more entrepreneurs should be working on impossible problems, because the small number that are actually impossible is outweighed by the heavy returns for the ones that aren't impossible but scare off competitors because they look that way.
It isn't about transferring "speed" from the wind, it's about transferring energy. Traditional methods, such as sails, didn't have a high enough wind capture/energy output ratio.
By using a propellor they are capturing and using the energy from the wind much more efficient than they would with a sail.
Obviously, that doesn't (and I can't) explain the finer detail, but I don't think it's hard to understand conceptually.
It's not about the efficiency, it's about the relative speeds: the blades of the propellor are moving relative to the vehicle whereas a sail is stationary.
So in your model, this car would run go equally fast without the gears between the propeller and the wheels? Why then do sailboats not have propellers instead of sails?
Initial claim of impossibility: http://scienceblogs.com/goodmath/2008/12/windpowered_perpetu...
Recognition of truth ;-): http://scienceblogs.com/goodmath/2008/12/the_real_bozo_attem...