Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Downwind faster than the wind

  1. Nov 7, 2008 #1
    Not too long ago Topher started a thread asking: is it possible to construct a wind powered vehicle that goes directly downwind - faster than the wind - steady state?

    I claimed this could be done, and I posted vector diagrams showing exactly how. This earned me a number of responses similar to this one from Shroder:

    I see. So the treadmill is exerting a force on the cart which is trying to accelerate it backwards. The restraint prevents this, so the turning wheels transmit force to the prop which then tries to accelerate the cart forward. And, the force of the prop on the cart is greater than the force of the treadmill on the wheels, so it moves forward. So what you are describing is a force multiplier. A simple lever is a force multiplier but the output force cannot be used to drive the input! What you are really describing here is a perpetual motion machine and it is an elaborate hoax. I am surprised that this thread has been allowed to continue for as long as it has.

    And this one from Topher:

    Also, I have constructed a vector diagram and analyzed it as I have asked spork to do several times. Its pretty clear just by looking at the diagram itself that the thing will never have a downwind component faster than the wind.

    Note that I did post a vector analysis, and it appears earlier in that thread.

    Aparently the administrators agreed that such a vehicle could never be built - because that would in fact constitute perpetual motion. As such, they locked the thread.

    So.... a friend and I went out, spent hundreds of dollars, and quite a few hours, and built a vehicle that does exactly that. And here that vehicle is:


    I assume this thread will be closed shortly as well. But at least for now, here is the physical proof that the vehicle originally asked about can be built to operate as described.
    Last edited by a moderator: Sep 25, 2014
  2. jcsd
  3. Nov 7, 2008 #2
    I'm the guy you see in the video -- spork is holding the camera. (he's too ugly so we had to keep him out of the frame).

    The basic design comes from the Bauer cart and has been also executed and filmed by Jack Goodman: (his video was posted on the thread that was closed).

    We've grown so tired of people telling us that to achieve success with a device like this would be to achieve perpetual motion (see shroder's quote above -- and many others) that when we got the cart finished and filmed it advancing on the treadmill we decided that we'd just post it as "perpetual motion realized". That way when everyone came along and said "that's not perpetual motion" we could say (as we've been saying) EXACTLY -- IT'S NOT PERPETUAL MOTION! (If any of you are fans of Monty Python, think "only the true Messiah denies his identity" and you'll get the picture.)

    Unfortunately, that thread got closed also -- not sure why.

    Anyway, it hardly seemed fair to finish and film the cart and not share the results. We find the hubub rather humorous as it's a quite simple device that will repeat it's feat as many times as you wish until you tire of the game.


    PS: we've almost finished a newer lightweight version that rips up the treadmill rather than barely advances -- we'll post that in a day or so when it's done. That form of that design came from another member of this site and is quite clever.

    PSS: Spork and I work at the same company, and built the device together but are not the same person as some have suspected.
  4. Nov 7, 2008 #3
    You can just imagine how ugly I must be if you see what we DID allow on camera. But the truth of the matter is that JB was on camera because he's not qualified to operate complex lab equipment (i.e. the treadmill). : )

    I think your chick still suspects that. But that works out OK for me : )
  5. Nov 7, 2008 #4
    That other thread was a real train wreck, with an over emphasis on ice boats rather than the particular machine in question and I was glad when it was finally closed. Hopefully you will receive a fair hearing here if you properly present the mechanics of the machine you built. For what it is worth, looking at the video, I can’t say that I am convinced of your claim but neither am I convinced that this is a hoax. What immediately comes to mind is that the rear drive wheels, when you are holding the machine back, are making better friction contact with the tread than when you release the machine. This can allow a transient condition in which the momentum of the spinning blades will temporarily overcome the drive force of the tread. But it is only a transient state and very soon the machine and treadmill return to the steady state condition of the machine being pushed backwards. Transient conditions do not constitute an greater than unity effect, in my opinion.. But I am not the judge and jury, let others have their say.
  6. Nov 7, 2008 #5
    Yes it was. We were attacked as charlatans when we presented perfectly valid analyses and analogies. In the end it comes down to this - we were told no such vehicle could be built. We have now built and demonstrated it.

    Despite your theory - no transient conditions are involved - nor is any "greater than unity effect". We take the laws of physics pretty seriously where I come from. I wouldn't break even one.

    You may not be convinced by the performance in that video. But I assure you there will be no room for doubt with the video we will post within the next day or two.
  7. Nov 7, 2008 #6
    I don't see enough of the design to be able to judge the device in the video, nor have I seen the locked thread spoke of. However, such a thing should at least in principle be possible. It is certainly not perpetual motion. The counterintuitive aspect of it is the mistaken notion that air resistance can be thought of as directional momentum collisions with the air molecules, like being drove back with a rapid fire BB gun. In fact it is a pressure difference where the faster moving air is less dense than slower moving air, the venturi effect. The air molecules can still be thought of as BBs of sorts except it's a questions of which side is receiving the most BBs rather than the velocity (f=mv^2) difference in the individual BBs. It's the same kind of silly physical mis-perceptions that aether theorist base the aether inflow theory of gravity on.

    Simple demonstration. Take a helium balloon and tie it to the center console in a car so it floats just below the roof. Get the car up to speed and break fairly hard. While everything else is being accelerated to the front of the car the helium balloon will dart to the back of the car. The video is no more perpetual motion than the balloon is a negation of inertial forces.
  8. Nov 7, 2008 #7


    User Avatar
    Science Advisor

    Greetings Spork and ThinAirDesign. I don't have any role in moderation here but I'm about 99% certain that the other thread was locked not because it was considered a "banned topic" (as in perpetual motion) but rather I think it was just an issue with the way the thread was heading with the goading of $100,000 wagers and a little bit of sarcasm and insults creeping in. I think if we keep this topic civil there's no reason for it to be locked.

    For the record I was convinced in the first thread that this is possible and doesn't violate any physical laws.
    Last edited: Nov 7, 2008
  9. Nov 7, 2008 #8
    The cart is no more perpetual motion than a sailboat is. The principle is straightforward even if not intuitive. The trick is to extract energy from the ground/air interface. When trying to sail a traditional boat directly downwind, we can't take advantage of that interface - thus when we approach the speed of the wind, we feel no relative wind, and can't accelerate any further.

    In the case of this cart, the prop blades are no different than the sail of an ice-boat on a 45 degree downwind tack. The only difference being that the prop tips maintain a continuous spiraling downwind tack while the cart itself goes directly downwind.

    Perhaps even less intuitive is that a sailboat proves this every time it tacks *upwind*. We just have to look at it in the frame of the wind, rather than the frame of the water. there's a relative velocity of the air and the water. In the case of an upwind tack we can just as easily look at this as the current propelling the boat upwind with a VMG greater than the speed of that current.
  10. Nov 7, 2008 #9


    User Avatar
    Homework Helper

    PhysicAddict, when are you going to post the video of your mini-cart?
  11. Nov 7, 2008 #10
    I will limit my comments to this particular type of machine, meaning I will steer clear of the diversion into ice boats and sailboats. As designers and builders of this machine, you should be able to inform interested readers of the principle involved which would theoretically allow it to advance in a steady state against the treadmill. You have not done that. You should be able to demonstrate that it can, in fact, advance in a steady state against the treadmill. You have not done that either. What you have shown is a machine that can make a transient lurch forward from a constraint, only to quickly come to a halt and fall back against the treadmill. According to the law of the conservation of momentum and the conservation of energy, if the machine stays at a constant mass, it cannot generate more energy or momentum than it receives from an external source. Based upon those inviolable principles, the machine cannot advance in a steady state against the treadmill. Now, if you contribute to the mass of the machine while holding it, and then reduce the mass when you let it go, it can have a momentary transient energy and momentum greater than what the mass of the machine can have on its own. That is apparently what is happening to account for the transient condition. I have presented my argument based upon sound established laws of physics. It is now up to you to show that this subject belongs in a serious physics forum.
  12. Nov 7, 2008 #11


    User Avatar
    Homework Helper

    I was going to wait for PhysicsAddict to post his video, but since it can be found via search on YouTube, here's the link. Note that his cart manages to accelerate forwards as opposed to just lurching forwards. In the last segment, his mini-cart runs into the far end of the treadmill.


    Note that the power source is the difference between the wind speed and the ground speed, which is independent of the vehicle speed. I'm not sure about the math for the prop blade's spiraling path, but for a landsail or icesail, the component of apparent wind perpendicular to the vehicle's movement is equal to the wind speed times sin(angle between wind and vehicle velocity direction), and is independent of the vehicles speed. For example, if the wind speed is 10 mph, and the vehicle is moving downwind with an offset of 30 degrees, then the vehicle experiences a relative crosswind of 10 mph x sin(30) = 5 mph, regardless of it's forward speed. If the vehicle can go faster than 11.55 mph with a 5 mph crosswind, then it's downwind speed will be faster than the wind. Downwind speed = vehicle speed times cos(angle between wind and vehicle velocity direction).

    Link to vector picture showing that the crosswind component pependicular to the vehicles direction is independent of the vechicles speed:

    Last edited: Nov 7, 2008
  13. Nov 7, 2008 #12
    :surprised The last video is very impressive, but I still have serious reservations concerning the added mass of the hand holding it down and then releasing it. But the advance does come close to a steady state condition. I remain skeptical while I applaud the effort!
  14. Nov 7, 2008 #13


    User Avatar
    Homework Helper

    Other than the initial run, the final touches with the hand before each release are to prevent forward motion of the cart, which clearly accelerates forwards for a bit after release. Preventing the cart from moving backwards could add momentum, but it appears that he's preventing it from moving forwards, especially on the last 3 runs, which shouldn't be adding any momentum to the mini-cart's prop.
  15. Nov 7, 2008 #14
    In fact I have done that. You can characterize the ice boats and sailboats as a "diversion" if you like. In fact what they are is a simple way to explain exactly what is happening here. And I did go into detail explaining that with vector analysis and carefully chosen analogies in the thread that was locked - the thread in which you claimed such a device would require perpetual motion.

    I have done that as well. While our machine is currently marginal as compared to PhysicsAddict's machine, it does in fact take itself from a steady state condition of moving downwind at exactly the wind speed (the state of non-motion relative to the treadmill frame) to a state of steady forward motion on the treadmill - until it goes off the side of the belt.

    You're wrong. You should watch the video. And read the analysis in the thread that you helped to close by insisting such a machine could never be built - and would constitute perpetual motion.

    Agreed. And it doesn't.

    It seems you're failing to apply those principles correctly to this experiment. Reality seems to fly in the face of your theory. And science is more about finding theories that explain reality than the other way around. My theory (which is perfectly simple and straightforward) does that.

    This is clearly not what's happening, and it makes no sense. The cart is released from a steady state - as such your theory does not apply.

    I have done exactly that - and was laughed off this forum by people that claim such a machine would imply perpetual motion - in fact by people that don't understand the principle of equivalence of inertial reference frames. My analysis exists in the first closed thread, and I stand by it 100%. I will reproduce it in this thread if need be, but I give you my assurance that it is accurate and correct.

    I should point out that PhysicsAddict argued with me, exactly as you are, that this would never be possible. He believed and argued this point for over two years. He made and tested models that failed to advance on the treadmill. It was only after he saw our marginal success a few days ago, that he constructed the cart in the video linked here. He managed to use the principles we described to build a cart that is much slicker than our own. And he has become an immediate convert. Reality has changed his way of understanding the scientific principles. That's how it's supposed to work.
    Last edited: Nov 7, 2008
  16. Nov 7, 2008 #15


    User Avatar
    Science Advisor

    Hi schroder, here’s a little thought experiment that might help convince you. In the spirit of de-bunking perpetual motion devices you can usually assume frictionless ideal operation of most components and of course they still fail to achieve "over unity" operation. So in this spirit lets assume that we have an ideal lossless drive train (wheels, belts and gearing) and further that we can adjust the gearing ratio from the wheels to propeller to any desired ratio. Lets just concentrate on non-ideal lift/drag of the propeller.

    First we note that the turning of the wheels is driving the prop, so the inevitable blade drag will mean we require constant torque to keep the prop turning at a constant rate and this torque must be provided by the wheels, giving a retarding force on the vehicle.

    Second we note that the lift generated by the prop is providing a forward directed force. So we now have two forces in opposition, the lift on the prop giving a forward force and the rotational drag on the prop which, through the drive train, ultimately results in a retarding force at the wheels.

    Since at first sight this thing looks like an “over unity” device our first instinct is to think that perhaps the lift force must be less than the retarding force. However the retarding force at the wheels is dependant on the gear ratio, that is, if we gear it so that the prop turns fewer times for each rev of the wheels then the ratio of retarding force at the wheels to blade drag to is also reduced.

    So lets play devils advocate and assume that we set this thing up on a treadmill and hold it until it’s at steady state (wheels and prop up to speed) and we find that the retarding force is indeed larger than the propeller lift and our vehicle goes backwards.

    No problems, lets just reduce the gear ratio so that the prop turns less times per wheel rev, and this will reduce the retarding force at the wheels. Arh but you say, this will also reduce the prop speed and so reduce it’s lift. Again no problems, just increase the treadmill speed until the prop turns at the same speed as it did before! Now you cant argue with this, the prop is at the same speed so the lift is identical to before, and the drag at the blades (torque required to spin the prop) is also the same as before, but due to the modified gearing the retarding force at the wheels is now lower than before. Can you see that in principle there is no limit to how much we repeat this procedure so eventually it has to work!
    Last edited: Nov 7, 2008
  17. Nov 7, 2008 #16
    Uart, you're describing it perfectly: it's relatively easy to make this work if you have a treadmill (or wind) that is running/blowing like stink. The challenge comes making one efficient enough to work on an ordinary 10mph limit home treadmill.

    The harder the wind blows, the more energy there is to be extracted and everything gets easier.

  18. Nov 7, 2008 #17
    Well, in that case I will drop out of this discussion. I do not want to be accused of getting this thread closed also! I’m interested in seeing it stay open in the hope that some of the heavyweight mathematical physicists here can weigh in and offer their analysis and opinions. I am not a heavyweight mathematical physicist, so even if you should manage to convince me, it will not settle the argument. For the record, I now hold a neutral position on the matter but am very interested in seeing some more opinions and analysis.
  19. Nov 7, 2008 #18


    User Avatar
    Science Advisor

    Thanks. :smile:

    Actually I wanted to post that explaination in the last thread but just before I got the chance they locked it.:uhh:
  20. Nov 7, 2008 #19
    The carts don't care about overweight mathematical physicists -- they just settle the argument by climbing up the treadmill.

  21. Nov 7, 2008 #20
    Thanks Jeff! and sorry for being late to the post here. And thanks for the "fmt=18" tidbit on the youtube to get hi-res video.


    Please watch that video again and pay close attention to how I VERY CAREFULLY do nothing more than VERY LIGHTLY hold the cart back while I adjust it's position for the 7 second 100 ft. run down its narrow path. I EVER SO LIGHTLY release my 2 finger hold and pull my hand away. Notice how the cart never lurches forward. At the end of the video I managed three straight shots in a row. Each time the cart slowly accelerates up and hits the end of the treadmill.

    Also schroder take note that treadmill in the video actually has a slight positive incline. As incredible as it seems that cart is doing it job heading ever so slightly uphill.

    Jeff Reid has suggested that the ring vortex that the cart sits in at the "break-even" point is a difficult barrier to overcome. After playing with this thing on the treadmill alot I can see where as the cart moves forward it seems that it does break out of it's own disturbed air and as it starts to see it's own headwind it looks like it really wants to stretch on out. Too bad my short little treadmill won't allow it.

    Man the term "argued" is spork's quote is a understatement:eek: I'm lucky I don't have to fly out to his house to mow his lawn and cook him dinner for the next 2 years.:smile:

    Folks, this is the most counter intuitive brain teaser I have ever twisted my hair over. It is especially difficult for someone such as myself who knows NOTHING about sailing. I had to approach this from an entirely different angle other than sailing to get my head around it. On the immediate surface it appears as an over-unity paradox but uart does a pretty good job of explaining how it's not really "over-unity".

    As it stands, I still consider my cart "marginal". It is really difficult to minimize the friction of all the transmission and rolling components on the smaller cart in order to bring it's performance envelope under 10mph which is all my treadmill will do. My cart is break-even at about 8.5mph and takes off in the video you see at 10mph.

    I will be spending time optimizing my cart some more when I get a chance. Right now its a pretty scrappy mish-mash of slow-flyer propeller, RC helicopter parts, scooter wheels, carbon fiber rods and voodoo spells.

    I do plan on documenting more on how its built in the weeks to come in case someone else wants to give it a go.

    Mark C.

    Attached Files:

    Last edited: Nov 7, 2008
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?