Down wind faster than the wind

[Michael C]

The only “endurance test” I’ve seen is the video where the cart hovers towards the back end of the treadmill and ends up falling off that end. Spork said something to the effect that a constant hover isn’t possible to hold indefinitely and that eventually the cart will go off either end. He said it was a “coin toss” as to which end it went off. I have never seen a video of a cart going off the top end from a hover however. The advantage of a turntable is there is no ends to fall off and any variance in “expected” speed can be easily observed and monitored.

You're certainly talking about this video:



To make the cart stay so long in more or less the same position on the treadmill, the makers had to balance the force to the left (mechanical friction) with that to the right (thrust of the prop through the air). The state of equilibrium is clearly hard to achieve, and not very stable: the tiniest increase in the frictional force (some dust got in the bearings, there was a slight deformation due to the bearings heating up...) will cause the cart to start moving to the left, whereas the tiniest decrease in this force will cause the cart to move the other way. Therefore it's a "coin toss" as to which end of the treadmill the cart will eventually fall off.

Here's a question: can you design a non-motorised machine, as small and as light as Spork's cart, that does not use the thrust of a propeller powered by the wheels, that touches only the belt of the treadmill, but which will stay in the same place on the running treadmill for more than a minute?

 

[Michael C]

Nobody has yet chimed in to help me with these three technical terms in Marks Drela's pdf:

- air prop induced efficiency
- air prop profile efficiency
- air prop swirl efficiency

Can somebody point me to an explanation of these terms?

 

[ThinAirDesigns]

The only “endurance test” I’ve seen is the video where the cart hovers towards the back end of the treadmill and ends up falling off that end. Spork said something to the effect that a constant hover isn’t possible to hold indefinitely and that eventually the cart will go off either end. He said it was a “coin toss” as to which end it went off. I have never seen a video of a cart going off the top end from a hover however. The advantage of a turntable is there is no ends to fall off and any variance in “expected” speed can be easily observed and monitored.

If one understands physics, and particularly the physics of this device -- it's easy to see that with it repeatedly climbing a ~4.5 degree treadmill, it has steady state, indefinite thrust to hand out like cheap candy.

After the above, running unattended and untethered on the treadmill is nothing more than a stunt -- a stunt we managed to pull of for near two minutes.

Knock yourself out with your own stunts.

JB

 

[H'ethetheth]

Nobody has yet chimed in to help me with these three technical terms in Marks Drela's pdf:

- air prop induced efficiency
- air prop profile efficiency
- air prop swirl efficiency

Can somebody point me to an explanation of these terms?

I can't point anywhere, but I can perhaps help you out a bit. I haven't taken out pen and paper to do the calculations yet, so I can't say I fully understand it. I do understand what he's driving at in general though, but maybe you do too.
Anway, here goes:
Usually, drag on a wing is thought of as consisting of two components. One component is the result of viscous friction as the fluid flows over the surface. It's called profile drag.
The other component, induced drag, depends (more or less quadratically) on the lift force produced.
The swirl efficiency tells you how much of the thrust is in the right direction. Due to the rotation of the propeller, the air will be accelerated radially and tangentially relative to the propeller disk. These radial and tangential velocity components do not produce thrust.
So he breaks up the total propulsive efficiency η_p into
a factor η_v that tells you how much viscous drag affects the power generated.
a factor η_i that tells you both how much drag the blades generate per unit lift and how much of the lift is turned into effective thrust η_swirl.

Did that help?

 

[jjcote]

Nobody has yet chimed in to help me with these three technical terms in Marks Drela's pdf:

- air prop induced efficiency
- air prop profile efficiency
- air prop swirl efficiency

I'm not a prop expert, but I can provide at least a partial explanation of the first one. "Induced drag" is a feature of any wing, and it has to do with the fact that there's higher pressure under the wing than there is above it. As a result, the air tries to sneak around somehow, and the one place where it can do that it at the wingtip. As a result, you have air slipping around the tip, which on e.g. a glider wing turns into a corkscrew of moving air known as a tip vortex, and this vortex carries away energy. Depending on the design, this can actually be the dominant form of drag on an aircraft. Making the wing long and skinny reduces this effect because (loosely speaking) the air has a long way to go to get to the tip. In the case of a prop, the same thing applies, with air slipping around the prop tip and leaving a corkscrew embedded in a bigger corkscrew as its wake.

I'll take a shot on the other two: profile efficiency might be just the drag associated with moving the blade through the air, irrespective of the fact that it's generating lift. Swirl efficiency I'm going to guess has something to do with the fact that the rotating prop has a tendency to make the air in its vicinity rotate, which also robs the prop of energy.

In any case, I'm pretty sure that these are the major components of drag that a prop designer considers when evaluating the overall efficiency of a prop. For those not concerned with the details, you can just assume that for a given operating point, there's an overall efficiency, just as for an airplane, you can think about the overall drag without breaking it down into induced drag, skin friction drag, parasitic drag, etc.

 

[Archangel]

Archangel, we no longer discuss "The validity of classical physics" in this thread. That discussion has been split to here so that this thread can be used to discuss the real wind powered devices that propel themselves downwind faster than the wind.

Sorry about that, I was ~20 pages behind when I made that post last night before heading home from work and only just got to the split notification.

 

[John Freestone]

I was thinking about the RA rules again, as I pondered some alternative designs for windcarts. It occured to me that defining the windspeed as measured at the height of the CoG of the cart might encourage some unhelpful 'evolutionary pressure' on design: using ultralight frames to raise a lightweight propeller as high as possible, and even slinging some heavy ballast under the chassis, could give thrust at a much greater height and greater windspeed than that which the rest of the cart is moving through. Of course, some of this may not be possible to get out of: it's likely that a low, sleek body close to the boundary layer would be preferable, with the impulse provided as high as possible, but I wonder if the windspeed height should be defined more in relation to where that impulse is located. Even making it half the height of the cart might be way higher than the CoG. Of course, in terms of just having fun racing things, you could allow those light high props and heavy ballast, but it takes it significantly away from the principle of DDWFFTW in my opinion. I could stick a sail into the wind at 100 ft and beat the wind at 3 ft with no mechanical advantage of props and stuff at all.

I was also wondering why there was a rule specifically saying that storage of energy was allowed after starting off. What kind of energy storage, and why would it be allowed? I can see that it will happen inherently from parts gaining momentum, but is the rule saying you can build in a heavy flywheel if you like? Also, is this not actually rather a problem, as energy could be stored over a period of less-than-windspeed travel and then released somehow to boost speed unfairly beyond windspeed steady state? I can't see how this is possible mechanically without a gear change, but 'Energy can be stored..." even suggests you could have a battery and motor!

There are probably quite a few general rules missing, defining what the essential challenge of a windcart is, which would cover such loopholes. Some might be difficult to get round except by the usual things that happen - a committee has to sit down, discuss a result or a design, and make a judgement. For instance, gears might be described in a rather too specific way (imagining gear wheels, etc.), but someone could then come along and, applying the storage of energy rule, use some of the lower-than-windspeed time of their run to wind a lump of lead up a pole, which, when it reaches the top, trips a lever allowing it to fall, driving the wheels or prop instead of being winched up by them. There might be general ways of describing 'gear changes' of course, that get rid of these cheats.

I've just had a crazy vision of my cart. We calculate its CoG at about 2' at rest where there is a light breeze, although bystanders are curious as to why it has a mast sticking up 50 foot into the air, apparently not doing anything. I set off, and the wheels drive a winch, pulling a sail up the mast, with a lead weight tied to it...

 

[mender]

It occured to me that defining the windspeed as measured at the height of the CoG of the cart might encourage some unhelpful 'evolutionary pressure' on design: using ultralight frames to raise a lightweight propeller as high as possible, and even slinging some heavy ballast under the chassis, could give thrust at a much greater height and greater windspeed than that which the rest of the cart is moving through.

4. Total C of G of vehicle to be used in calculating DDW velocity and finishing position.

Height of C of G for the vehicle is open. Rule #4 relates to finishing position for the course. A factor will be applied to the finishing time to equalize the transit time of the C of G of the system. Example: a wind turbine used in a stationary position must have its weight and any transmission cable connected to the cart considered when calculating finishing time and position. (Note: wind turbine is used as an example only. Rule #6 prohibits stationary components.)

Rule # 6. Ground interface must travel with vehicle (no stakes, etc.).

I could stick a sail into the wind at 100 ft and beat the wind at 3 ft with no mechanical advantage of props and stuff at all.

3. Average height of "sail" from ground TBD.

Rule #3 restricts use of the wind speed gradient above the ground.

I was also wondering why there was a rule specifically saying that storage of energy was allowed after starting off. What kind of energy storage, and why would it be allowed? I can see that it will happen inherently from parts gaining momentum, but is the rule saying you can build in a heavy flywheel if you like? Also, is this not actually rather a problem, as energy could be stored over a period of less-than-windspeed travel and then released somehow to boost speed unfairly beyond windspeed steady state? I can't see how this is possible mechanically without a gear change, but 'Energy can be stored..." even suggests you could have a battery and motor!

5. Variable speed or pitch mechanisms allowed.

9. No energy storage before start allowed.
10. Energy storage after start is allowed.
11. Contest will begin with an unassisted start in a stationary position.
12. In case of #10, start is considered to have occurred when the start signal is given.

Use of a system to store energy after the start to be used to increase speed at any point along the course is allowed. Battery and motors are allowed. Weight and pulley systems are allowed. Flywheels are allowed. Use of the propeller as a flywheel is allowed. In all cases, carts may use only the energy of the wind to power the cart, that energy being used or captured from the time of the official start. No prior energy storage allowed.


In the spirit of allowing open competition and encouraging innovation, these rules have been made as general but as fair as possible. In the event of unforeseen advantage, rules may be altered or added to equalize competition. Please submit all entry designs two months prior to the event for approval. When the vehicles are submitted for approval at the event, the tech officials have discretionary powers and their decision is final.

 

[Christian Klippel]

Hello,

i'm not sure that the combination of battery, generator and motor is feasible. Same goes for a flywheel, but to a lesser extent. The problem i suspect comes with the added weight, additionally with the charge required in the case of a battery. If the course is not that long, it might take too long to just accelerate the whole thing, especially when a rather big flywheel is used. In case of a battery/generator/motor combination, one would need a rather long time to charge the battery to any useful level.

One might use a bank of super-cap's instead of a battery, which would reduce charge time significantly, but then rather high currents are required which will impose a rather big load on the generator (and thus the whole cart). Also, in such an electrical system you have losses in the generator and motor, so that you get significantly less out than you put in, making it a rather short-timed help to maintain the speed or increase it.

But then, these are just my opinions. Personally i would go with a cart as lightweight as possible to get a fastest possible acceleration.

Greetings,

Chris

Edit: Imagine everyone would use a generator/battery/motor combination. The start signal would be given, and people start to charge the batteries for the first x hours before really moving. The whole thing would become a competition about who has the most efficient generator and best charging circuit. For me, that would pretty much defy the original intent of such a thing.

 

[John Freestone]

Hi mender, whoops, it looks like I've been a bit dim. I checked back and remembered rule #17 concerning the throwing of fruit. D'oh!

 

[mender]

No problem, John, all in good fun!

Chris, I expect that a "pure" cart will end up being the winner on any length course, but the expected course length will be around 3000 feet, typically a runway at an old abandoned airport. The efficiency losses and weight penalty of the more complex system (not to mention the possibility of outright failure) would in my mind cause more issues than advantages.

The fact that the cart extracts power from the wind while moving faster than the wind is what usually gives people fits. I see the allowing of energy storage as a method of demonstrating the concept of DDWFTTW in a way that might help people understand how it works by separating the extraction of the wind power from the actual propulsion of the cart.

The idea of energy storage and variable speed/pitch mechanisms is also to allow the prop cart a method of getting through its lowest acceleration period, the self-start from a stationary position. To me, that is its only weakness.

I feel that the prop cart or some minor variation of it will perform the best and win the race. But that's why we have competition, for people to prove otherwise!

 

[jadebox]

Excuse me for jumping in here when I've been reading the posts occassionally (and not undertanding) 90% of them.

I was skeptical of the claim that the cart could go faster than the wind, and all the talk of treadmills and all just confused me, so I tried to visualize what's happening in as simple a way as I can.

My first thought is that there is really two problems - accelerating the cart up to the wind's speed then going faster than the wind's speed. The first problem is trivial if you ignore little things like friction. The cart is just like a sail boat.

But, the sail boat analogy breaks down when the cart reaches the wind's speed because there would no longer be a relative wind on the "sail."

So, to begin an examination of the second "problem" I'm picturing the cart suspended in the wind as if under a parachute. It's drifting with the wind so, relative the cart, there is no wind.

Now, I set the cart down on the ground (which, to the cart, is moving). The moving ground causes the wheels to turn and the propeller to spin, pulling the cart forward. Assuming that the propeller provides enough thrust to overcome drag and friction, the cart moves faster than the wind.

Does this make sense? Did I over-simplify it?

-- Roger

 

[John Freestone]

Nup, you've confused me again, mender. It's all good fun, or there's a 3000 foot abandoned runway and rules of the RA? It's gonna be damn near impossible to have a race meet unless you can turn the runway round to match the wind, assuming there is some wind.

If I can start off, then go quite slow for an extended length of time, with my variable pitch prop set to generate electricity to charge my battery, then, at some time later, whiz off downwind faster than the wind powered only by the... ??? ...with my prop at any height I can get it to while the windspeed is calculated at the c of g... ??? I mean, there's having design freedom, and a bit of fun on a runway, and then there's DDWFTTW. There needs at least to be a rule about how long I could go slowly charging a battery. Allowing batteries seems utterly stupid anyway, and only gives one more area where cheating could happen. Whose to say you can't have a little hidden switch - trip it one way and the examiner says there's no charge in the battery - trip it the other and you've got some juice you put in earlier? What does a windpowered cart need a battery for?

I'm sorry if I'm sounding a bit off with you, but I get different messages - one minute you sound like you're actually taking charge of the situation, inviting people to follow your lead, setting up an association for racing this thing, and the next it's just a bit of fun and we can collect energy to store in a battery. Are you serious?

Chris, I don't think added weight of a battery and motor is going to affect the performance that much, compared with its ability to power a cart. It takes accelerating, of course, but if the course is long enough to give the carts a fair trial (and 3000' might not be), then they will gain a little of advantage in momentum if there's a momentary drop in wind. [ETA: whoops, missed a 0 there - 300 certainly won't be!]

 

[mender]

...with my prop at any height I can get it to while the windspeed is calculated at the c of g...

Rule #3 requires that your prop be at a specific height from the ground - to be determined what that height is.

John, don't forget that while you're sitting there charging a battery (at ? 60% efficiency by the time that energy is at the wheels and doing something), the other carts are on their way to the finish line. You can't store energy before the start (rule #9). With only 3000 feet to work with, there isn't much time to charge a battery before you have to use it or you'll lose the race. Also, if you don't use up all the charge on that run, the officials will "help" you discharge it to a preset voltage before the next run to make sure you don't have an advantage. I'm sure we have enough EE people here to be able to police that!

If you go too slow for too long, you'll lose! Do some figuring based on a wind speed of 15 ft/sec over 3000 feet. If you go half the windspeed for 1/4 of the course, how fast will you have to go to catch the other carts by the finish line? Assume an acceleration rate of 10 ft/sec/sec to the top speed that you think you can maintain after your small amount of battery charge is gone. As a guess, a really good time for a prop cart would likely be a little over 2 1/2 minutes. How fast do you have to go to beat that?

Scheduling race meets will be a bit of a problem and will require some flexibility to get the right wind direction and speed. Not too much different than land speed record stuff.

It is all in fun, nobody's going to get rich doing this, but rules are needed to keep things fair. I proposed a set of rules to get things started. They are open to modification as needed, nothing set in stone yet. I've been racing cars for over thirty years, from local ice racing up to national class racing; serious racing but still fun! With my experience reading rules and figuring out how to best make use of them, I think I have a good chance of countering any attempts at cheating!

If you can get to the finish line in less time than a free-floating balloon can, you are traveling DDWFTTW regardless of how you do it, as long as you use only the energy in the wind.

 

[John Freestone]

Mender, I see. First up, I was actually thinking that the 'race' was more like a competition to see who could build a cart that would beat the windspeed by some particular percentage or whatever, rather than a simultaneous start or speed to a finish line. Sorry.

I guess it's just that your way of designing these rules are quite a way off how I would do it, and your suggested procedures are too. For instance, I'd hold tests at a wide-open space known for good even winds, but allowing for travel at a range of directions - a dry lake bed, salt flats or even suitable disused industrial area. I guess I don't know much about airfields, but the ones I've seen would only allow a very particular race direction (since this is DDW) and that's a lot less predictable from weather reports than just that there will be some suitable windspeed. So if you've got several carts all trying to go down a runway, but the wind makes the DD line off from the runway, and you've grass or something on either side, you either cancel or have carts going off the runway. I would then suggest that some kind of bubble-generator could be used at the start, so that everyone can see continually where the wind is going on that line, and carts could follow the line of bubbles within a certain degree of drift perhaps. But then my tests would be about making a cart that is most efficient at DDW, and the record would be a factor multiple of windspeed as best that can be calculated. I see this just comes from our different interests, and mine is pretty likely to remain academic anyway, at least for the foreseeable future.

I understand that you've decided that the cart's prop must be at a specific height, which begs the question that there's a single prop, and carts aren't going to use other means of harnessing the wind's energy. This is not spelled out in the rules, and also seems like an uninteresting limitation. It would, however, get round my objection of the boundary layer issue. If I had two props, I could put one [ETA: large one] very high, another [ETA: small one] low, and their centre would be unrepresentative of the wind at which most of their thrust is delivered. I don't know if a prop is the only viable contender - perhaps someone might come up with alternative propulsion types, like longitudinal impellers or whatever. Even if they're not as good, it might add to the interest and fun if you allow them.

I guess that kind of half answers my question about why you'd allow electrical power and other means of storing energy on a DDW cart - it could just be more interesting and fun to allow these additions of weights and batteries. Personally I feel it offends the principle, but whatever.

You say, "I proposed a set of rules to get things started. They are open to modification as needed, nothing set in stone yet.", and that's great. It's just that my questions and concerns about your rules, I felt, were met with a bit of a stone wall, rather than interest and a readiness to look at other points of view. I don't mind a bit if you said "Look, I'm setting up a particular race, and working out some rules. I'm the head of it, and things will be done my way." I haven't read that, but I don't quite know whether you mean the rules are open to modification and you are open to working with others to find good ones, or just that they're open to modification, but only by you, and if anyone has any alternative thoughts they won't be of any value whatsoever.

 

[jjcote]

I wonder whether it would be more useful to specify measuring the wind speed at the center of pressure (i.e. the center of the frontal silhouette) of the machine rather than its center of gravity. I recognize that it will take some work to come up with a clear definition of this when you're talking about machines with moving parts.

Alternatively, you might just want to specify that the wind speed will be measured at some particular height above the ground, and also put a height restriction on the machines.

 

[CyCrow]

Why not use a format with heats, 2 vehicles racing down the same line on parallel tracks? That would make excact windspeed measurements unneeded, and would probably be more exiting to watch.

As for height, limit total height to 4-5 meters. Also limit total width. That effectively limits usable sail/prop area.

No chemical batteries. They don't work well near total emptiness, and have somewhat undefined energy reserves. Very hard to prevent cheating, as even small amounts of electrical energy can be very useful. Capacitors that can be shorted down to 0 V before start would be OK, but maybe disallow all electromagnetic propulsion?

The surface should be specified. Some wheels work well on tarmac, but not on dry lakebeds.

// CyCrow

 

[mender]

John, the competition is open to any type of wind powered device. Rule #2 covers that:

2. Maximum "sail" area TBD.

The reason "sail" is in quotes is that it will apply to a sail, propeller, parachute, impeller, etc., equally so that many approaches can be used. Equivalency between the different wind power capturing methods will have to be decided on to make it a fair race. Several carts racing at a time would be the preferred format (alluded to in rule #16).

The height of the sail area is already covered and would be calculated using the average of the area (not allowing what you suggested, John):

3. Average height of "sail" from ground TBD.

There seems to be some confusion about the centre of gravity rule. This was suggested a while back by JB (Thin Air Design - which I thought was a beauty salon for men and was going to ask about pricing) as a method to compensate for any "unusual" wind capturing methods, specifically allowing the use of a parachute with a pulley attached placed partway down the course with one end attached to the cart and the other end being reeled in as the cart moves over the ground. The centre of gravity of that system would need a factor to equalize the performance. The size of the parachute and the height off the ground also needed to be specified to prevent the use of kite sailing techniques.

Cycrow, the venue will be whatever can be found that is suitable. Surface traction may be an issue as mentioned.

Capacitors could be allowed but unless someone can come up with a convincing argument supporting chemical batteries, it sounds like they should be scratched off the list.

Sorry if my way of responding was off-putting, it wasn't meant to be. This is what you encounter in every form of car racing though! The people running the show are usually very particular about rules and they are the only ones who have an opinion! I on the other hand am looking to get some rules hammered out and don't expect that the ones that I threw together will be the ones we end up with. I have put together sets of rules before to make the events more enjoyable. I have found that having a poor or vague set of rules can ruin what could otherwise be a lot of fun! It's all about giving everyone an equal opportunity to show their ingenuity.

All suggestions must be submitted in triplicate and notarized by at least two DDWFTTWRA officials in good standing ... or just put your comments on this thread!

 

[ynot]

Here's a question: can you design a non-motorised machine, as small and as light as Spork's cart, that does not use the thrust of a propeller powered by the wheels, that touches only the belt of the treadmill, but which will stay in the same place on the running treadmill for more than a minute?

Here’s an answer: Yes of course I can. A ball bearing will do this if I angle the treadmill downhill towards the front the correct amount so the ball will hover (not against the rules you set).

Holiday travelling is over and am starting to build my new turntable and cart for testing.

 

[ThinAirDesigns]

Here’s an answer: Yes of course I can. A ball bearing will do this if I angle the treadmill downhill towards the front the correct amount so the ball will hover (not against the rules you set).

This is a perfect example of your lack of depth when it comes to testing. It will be far more difficult than you imagine to keep a BB within the length of a treadmill belt without running off one end or the other.

Try it and see -- be honest and show all attempts until success. Oh, and no BS "spot the flaw" games this time around.

JB