Down wind faster than the wind

[John Freestone]

Yeah, I see. Cheers Greg. I don't even think it would get to windspeed. It's just rubbish! If my Edit link hadn't expired, I'd delete it. OTOH, as I was trawling through to find which b*d told me "Yeah, sure, it's fine to have a cowl over half yer vanes, man, chill...", the things I wrote about the cart and mechanics generally were a lot more embarrassing.

I think spork is right about the one-piece solution. I don't follow what you're describing, but anyway, I certainly count a few more parts than one there! A prop must be rotating round or relative to something, I presume, so there's two. I'm intruiged what it is you mean though. There are too many possibilities for what 'diameter' you mean, or what the geared ground is for in relation to the 'corkscrew', etc.

My only other design in progress (which no-one better blow out the, er, air, god help 'me) is not an attempt to reduce the number of relative parts, and it's nothing I thought up myself (more chance it will work! haha) - it's the tumbleweed principle again, but with the vanes geared from the wheels so that they turn at half the ang. velocity, vertical below, horizontal above, as they rotate round the gear at the axle. Someone on the thread put the idea out there and I'm just doing it for the exercise in mechanics, to see if I can figure out what it entails. It ends up with a lot of geared surfaces, which is inefficient, but at least there's no 90^o.

Now there's a thought - I've just realised that your gravity function thing is an option for this one as well. I wonder if there are any major disadvantages. I was gaining the torque off a chassis again with a front wheel. There needs to be something that can't rotate. A weight would get rid of chassis and one extra wheel, with its bearing and rolling resistance. I suppose one disadvantage of these single axle jobs is the steering isn't as simple.

Are you working on any designs now you've moved on from the simple anemometer-type tumbleweed? I like coming up with alternative methods. I played about with the idea of flapping vanes, as you describe, and hit many of the same problems. I like your saloon doors idea, and I didn't think of turning it that way. The pressure switch I don't like so much, because you need the doors open for a certain period, so your switch is going to have to be an arc, or it has to be well timed how it opens and closes the doors, like getting your valve timing right on a car engine. I imagine with all such flapping arrangements, efficiency is going to rapidly disappear, and gearing them seems a better option.

Anyway, if you feel like discussing different methods, I'm all ears. I'd love to see more on your balancing prop version sometime.

 

[spork]

...the things I wrote about the cart and mechanics generally were a lot more embarrassing.

You know John, you came on a little strong (and wrong) at first, but you fairly quickly did a bit of research, returned with an open mind, and have been contributing to the thread in meaningful and entertaining ways ever since. I don't think you have anything to be embarrassed about.

I don't follow what you're describing, but anyway, I certainly count a few more parts than one there!

I'm pretty sure I see exactly what he's getting at. Imagine the "ground" was actually a big giant threaded tube, and the wind is blowing straight down the axis of that tube. His cart would be the mating threaded tube that fit inside the "ground" threaded tube. Inside his tube would be a prop that's mounted to the tube so it's all one solid piece. I think it's a very clever approach to achieving DDWFTTW with one solid chunk of material, but I think it has two problems... it would be impossible to get it to move through the "ground" tube with low enough frictional losses, and as Greg notes, such a vehicle brushes up against the limits of the definition of what is the vehicle.

Admittedly, I'm turning his ground into a full tube. It could be half a threaded tube, or even less. But I think the concerns remain.

it's the tumbleweed principle again, but with the vanes geared from the wheels so that they turn at half the ang. velocity, vertical below, horizontal above, as they rotate round the gear at the axle. Someone on the thread put the idea out there and I'm just doing it for the exercise in mechanics, to see if I can figure out what it entails.

That idea can work theoretically, but will be quite a challenge for several reasons. If the vanes have any significant chord, there will be sufficient losses just from turning them. Also, the angle the vanes need to be at any point of the rotation will depend on the speed of the wheel relative to the wind. You'd almost have to make an active system to orient the vanes by measuring force-feedback. Finally, vanes are far less efficient than airfoils.

I really would like to see someone build a working non-prop design. I have no doubt it can be done; but it won't be easy (in my opinion).

I'd love to see more on your balancing prop version sometime.

What's this?

 

[mender]

I really would like to see someone build a working non-prop design. I have no doubt it can be done; but it won't be easy (in my opinion).

I'm still trying to get to it. Initial testing of my first pass at a non-prop mechanism was surprisingly successful but the test device was built with no concern for weight. Proper material usage would see a cart maybe 60% heavier than the prop cart. Another design with more efficiency is on the drawing board for the next stage of testing.

I'm worried that someone is going to beat me to an actual treadmill test though!

 

[spork]

I'm worried that someone is going to beat me to an actual treadmill test though!

I wouldn't worry about anyone beating you to a successful treadmill test of a non-prop DDWFTTW cart. It will be a challenge, but I have no doubt you can succeed if you put your mind to it (and are prepared to waste some time and money).

 

[mender]

I'll have to kick my humber habit first though! Man, I can't believe how much time I waste on him!

 

[John Freestone]

You know John, you came on a little strong (and wrong) at first, but you fairly quickly did a bit of research, returned with an open mind, and have been contributing to the thread in meaningful and entertaining ways ever since. I don't think you have anything to be embarrassed about.

Awww, thanks. But I'd forgotten just how arsey I was. Like I am now, but mostly just to humber.

I'm pretty sure I see exactly what he's getting at. Imagine the "ground" was actually a big giant threaded tube, and the wind is blowing straight down the axis of that tube. His cart would be the mating threaded tube that fit inside the "ground" threaded tube. Inside his tube would be a prop that's mounted to the tube so it's all one solid piece. I think it's a very clever approach to achieving DDWFTTW with one solid chunk of material, but I think it has two problems... it would be impossible to get it to move through the "ground" tube with low enough frictional losses, and as Greg notes, such a vehicle brushes up against the limits of the definition of what is the vehicle.

Admittedly, I'm turning his ground into a full tube. It could be half a threaded tube, or even less. But I think the concerns remain.

That idea can work theoretically, but will be quite a challenge for several reasons. If the vanes have any significant chord, there will be sufficient losses just from turning them. Also, the angle the vanes need to be at any point of the rotation will depend on the speed of the wheel relative to the wind. You'd almost have to make an active system to orient the vanes by measuring force-feedback. Finally, vanes are far less efficient than airfoils.

I really would like to see someone build a working non-prop design. I have no doubt it can be done; but it won't be easy (in my opinion).

Thanks. What's the vane's chord?

What's this?

#3930 [whoops 3130] above, a 2-part cart: twin wheels with some sort of gear; prop with gear - and somehow arranged at the gearing so that the weight keeps it from flipping, if I understand right. Crazeee duuuude!

I'm still trying to get to it. Initial testing of my first pass at a non-prop mechanism was surprisingly successful but the test device was built with no concern for weight. Proper material usage would see a cart maybe 60% heavier than the prop cart. Another design with more efficiency is on the drawing board for the next stage of testing.

I'm worried that someone is going to beat me to an actual treadmill test though!

I don't suppose it'll be me, but you never know. I guess there are two targets there if we did get competitive about it - first to test and get FTTW with a non-prop version, and the on-going record for efficiency I suppose, that breaks even at the lowest speed or whatever.

I wouldn't worry about anyone beating you to a successful treadmill test of a non-prop DDWFTTW cart. It will be a challenge, but I have no doubt you can succeed if you put your mind to it (and are prepared to waste some time and money).

Has it only been done with a prop?

 

[spork]

I'm not sure if that means you find the concept bizarre or that I just did a really bad job of explaining it.

Thanks. What's the vane's chord?

If the vane is attaching two wheels I consider the distance between those wheels to be the length or "span" of the vanes. The other dimension of the vane is what I'm calling the "chord".

Has it only been done with a prop?

There are plenty of vehicles that can tack their way downwind faster than the wind, but to my knowledge the only way it's been done directly downwind is with a prop - of course the prop blades are on one long continuous tack.

Or if you prefer the humberian explanation - it's never been done.

On a tangentially related note, would it be fair to say that humb's posts are humberific?

 

[Dan O.]

I posted this one a few hundred posts ago. This is just reworked to show the gearing.



I would call this a 3 piece design by not counting the replication of the vanes (a single vane version would work but not as smoothly and starting might be a problem). The gear in the center is weighted so the dark side is always down. It has 1/2 the diameter of the gears attached to the vanes so the vane makes a 1/2 rotation each trip around the wheel.

 

[mender]

On a tangentially related note, would it be fair to say that humb's posts are humberific?

Humb's parody of humber is definitiely one of the high points of this thread.

 

[John Freestone]

I posted this one a few hundred posts ago. This is just reworked to show the gearing.



I would call this a 3 piece design by not counting the replication of the vanes (a single vane version would work but not as smoothly and starting might be a problem). The gear in the center is weighted so the dark side is always down.

Use the force, Dan.

It has 1/2 the diameter of the gears attached to the vanes so the vane makes a 1/2 rotation each trip around the wheel.

Dan, I'm sorry that I laid into that idea all those hundred posts ago. I would have apologised earlier, but I thought it was someone else who posted that, and that they left the thread. I see you've made it even more ridiculously big and dangerous, so it might fall over and squash the model (a stupid job anyway, if you ask me, posing next to hypothetical machines). I flatter myself that's a friendly dig at me.

I'm torn between discussing my version and so collaborating to some extent on best designs, toward your successful testing and competing against you on it. I know my record probably won't have you desperate for my help or quaking in your shoes either way.

I'm not sure if that means you find the concept bizarre or that I just did a really bad job of explaining it.

I was just kind of shocked at the bizarreness of it in context - I suppose because trying to make a one-piece design based on the wind physically shoving a screw down its thread would just be in the bin before I mentioned it to anyone! (Oh, God, like the cowl idea should!) I didn't quite get my head round it, just from not reading carefully and thinking about it enough, and it was late. This morning it makes perfect, er, sense. I didn't quite get that the cart (screw) was also the prop (airscrew) as one piece (which obviously they would be, or both you and Greg can't count properly, and that's precisely what you said!). Actually, I've just realised, you've done a simliar thing to what I did with my pipe zizmo, turning the threaded 'ground' into a tube (or maybe that was Greg's intention). It's interesting, because with just a few bits in the way of bearings at the screw interface (breaking the pure intention) it would make a lovely example in practice. I think I do get it fully now. It's a prop that's pushed down the thread so that it has to turn, and it's screw pitch is greater than it's prop pitch. In another way, other than trying to make it one-piece, it's very similar to other variants with a geared 'ground'. I've posted the idea of hanging a prop cart from a bike chain, for instance, and driving the chain, and various people have suggested a land-type cart with geared wheels on a geared track, if only to confound humber's levitation theory!

I suppose there's that question hanging (as if it was important at all) of whether one-piece can be one very large piece compared to some very small bearings. And, as you say, whether DDWFFTW implies relative to a surface connected to it only by friction, or if a ruddy great screw thread is ok. Furthermore, whether the air flow has to be uncontained, or a pipe is ok. The latter could get into the realms of higher pressure air input.

Actually, I'm just thinking that it could even work in practice, that one. If the screw thread is nicely greased and of a very long pitch, and you blast a fair old gale through it, with a short pitch prop, especially one with a large-ish surface area, it can't do much but be pushed along, and at some point outpace the airflow. You know, this could make a really great sport - a sort of pumpkin-airgun games (you have them in the US, I think) for tech heads - a rifled barrel firing prop shot across a field would be a great way to spend a Saturday afternoon.

If the vane is attaching two wheels I consider the distance between those wheels to be the length or "span" of the vanes. The other dimension of the vane is what I'm calling the "chord".

I see. Thanks. It reminded me of the curve of Greg's vanes, and I just wondered if that was what you were referring to. I suppose there's little advantage to use curved ones. I think it was just because of his anemometer idea.

There are plenty of vehicles that can tack their way downwind faster than the wind, but to my knowledge the only way it's been done directly downwind is with a prop - of course the prop blades are on one long continuous tack.

Yes. What a wonderful invention the prop is. So simple, and yet it still seems quite mysterious. There's something about that in the DD cart discussion - the fact that there's a surface always moving slower than the vehicle almost seems like magic, over-unity itself. Something going backwards w.r.t. the cart must come the other way again at some point! But then a wheel is pretty mysterious.

Yes, the big problem with all these other prop replacements is that the bit going backwards has to return against the 'headwind'.

Or if you prefer the humberian explanation - it's never been done.

On a tangentially related note, would it be fair to say that humb's posts are humberific?

I think that would be fair, yes. The adjectives humberesque and humberian have their place too. Humb's posts are humberesque and terrific, so that makes them humberific. Humber's posts are more purely humberian, IMHumberO.

 

[GregLondon]

I don't follow what you're describing, but anyway, I certainly count a few more parts than one there! A prop must be rotating round or relative to something, I presume, so there's two. I'm intruiged what it is you mean though. There are too many possibilities for what 'diameter' you mean, or what the geared ground is for in relation to the 'corkscrew', etc.

Take a piece of 8 guage wire (really fricken stiff wire). Now go to the imaginary hardware store and get a big fricken bolt that's a foot in diameter and has big threads that are something like an inch apart. Take the wire and wrap it around the bolt so that it follows the threads of the bolt. Then unscrew the bolt and throw it away. (Don't worry. They're imaginary, so they're cheap if you need another one.)

That wire is the "corkscrew".

Now, get yourself a metal four bladed propeller that is 11.5 inches in diameter. Get your soldering torch out and braze the tips of the prop to the inside of the corkscrew.

you now have a single moving part.

Now, go to the imaginary hardware store and get yourself a big metal nut that would fit the 1 foot diameter bolt you just used. You want a nut that is a few feet long. Take the nut back to your imaginary shop and put it on the bandsaw and cut going along the shaft of the nut so that you end up with a slice of exposed threads on one side. You just want a little bit of curvature to the threads so that it will hold the corkscrew in place and keep it from rolling off to either side. The rest you want to cut away so that the wind can hit the prop, and so you don't have any more friction than you need.

This sliver of a bolt is the "track".

Put the track on the ground, with the threads facing up. Put the corkscrew in the track. Aim the track so its facing directly downwind. and if friction is low enough, it'll move.

Then its a matter of adjusting the pitch of the prop blades so that you get the gearbox ratio going on and you can go faster than the wind.

There are two moving parts, the corkscrew and the track affixed to the ground.

 

[GregLondon]

I would call this a 3 piece design by not counting the replication of the vanes

That's a pretty cool way to do it.

When I was dealing with vanes, I ended up making them shutter open and closed. They hung on the wheel like saloon doors, stuck otu into the wind when that part of the wheel was close to the ground, and laid flat against the wheel during the rest of the time to get out of the way. The idea was to use some sort of mechanical switch along the rim of the wheel that was underneath each door, so it would know when to open and close the doors.

Description on page 22 of my doc. Drawings on page 23.

http://www.greglondon.com/tumbleweed/tumbleweed.pdf

 

[GregLondon]

I think this is my favorite ever of all time non-propeller propulsion system I've ever seen ever at all:

http://en.wikipedia.org/wiki/Voith-Schneider

ever

 

[GregLondon]

for total steam-punkery type goodness, page 44 of my doc has a "sail-barge" that should go directly downwind faster than the wind without a propeller. It's just the yoyo with a parachute idea expanded upon until you get a vehicle that doesn't run out of rope for the parachute. the "parachute" version of the cart is on page 37.

I couldn't come up with a nice way of opening and closing the chutes, but the sailbarge idea fixes that. Instead of having teh chute lines all connect to a single point, its more like a sail where lines attach to two different tracks or chains. WHen the chains are togeher, the sail is "closed". When the chains move apart, the sail "opens" and catches the wind.

 

[GregLondon]

John,

If the screw thread is nicely greased

I think polished steel on polished steel with a light oil has a slim chance of working if it uses a big enough prop.

Otherwise, I'd say use magnetic levatation so the corkscrew and track don't actually touch.

maybe electrically insulate them from each other and then give both pieces the same electrostatic charge so they repel each other.

Oh my god.

I've got it!

You make the corkscrew bouyant and float it in WATER.

The propeller is pitched to engage the air. The corkscrew is pitched to push the water.

Then put the water in a trough so the corkscrew doesn't sideslip.

 

[Furcifer]

Take a piece of 8 guage wire (really fricken stiff wire). Now go to the imaginary hardware store and get a big fricken bolt that's a foot in diameter and has big threads that are something like an inch apart. Take the wire and wrap it around the bolt so that it follows the threads of the bolt. Then unscrew the bolt and throw it away. (Don't worry. They're imaginary, so they're cheap if you need another one.)

That wire is the "corkscrew".

Now, get yourself a metal four bladed propeller that is 11.5 inches in diameter. Get your soldering torch out and braze the tips of the prop to the inside of the corkscrew.

you now have a single moving part.

Now, go to the imaginary hardware store and get yourself a big metal nut that would fit the 1 foot diameter bolt you just used. You want a nut that is a few feet long. Take the nut back to your imaginary shop and put it on the bandsaw and cut going along the shaft of the nut so that you end up with a slice of exposed threads on one side. You just want a little bit of curvature to the threads so that it will hold the corkscrew in place and keep it from rolling off to either side. The rest you want to cut away so that the wind can hit the prop, and so you don't have any more friction than you need.

This sliver of a bolt is the "track".

Put the track on the ground, with the threads facing up. Put the corkscrew in the track. Aim the track so its facing directly downwind. and if friction is low enough, it'll move.

Then its a matter of adjusting the pitch of the prop blades so that you get the gearbox ratio going on and you can go faster than the wind.

There are two moving parts, the corkscrew and the track affixed to the ground.

If you ever need a reference for technical writing feel free to contact me. My real IKEA furniture instructions didn't make as much sense as your imaginary faster than wind bottle opener

 

[John Freestone]

John,



I think polished steel on polished steel with a light oil has a slim chance of working if it uses a big enough prop.

Otherwise, I'd say use magnetic levatation so the corkscrew and track don't actually touch.

maybe electrically insulate them from each other and then give both pieces the same electrostatic charge so they repel each other.

Oh my god.

I've got it!

You make the corkscrew bouyant and float it in WATER.

The propeller is pitched to engage the air. The corkscrew is pitched to push the water.

Then put the water in a trough so the corkscrew doesn't sideslip.

Wild! Thanks, Greg, I get the idea now. There seems to be something a lot simpler using the same technique, unless I'm being stoopid again. Take your prop and mill the shaft mount hole in the centre to fit over a bolt instead, no? I would have thought that would do the same thing a lot easier and without getting to the imaginary hardware shop and welding stuff. It means there's automatically a small surface area for friction, and the prop is perfectly balanced, instead of trying to make itself roll out of the track.

I haven't got my head round the mechanics fully yet. Should the track have a longer pitch than the aero pitch of the prop, or the other way round? I'm confused. I think it's the first, but sometimes my head screams that it's the latter. I think that is because I imagine the prop has to be forced to turn faster than its progress down-track to have any effect on the air, but I'm forgetting that there's a tailwind. The cart is still difficult for me, but these two different pitches do my head in. That just seems to fit with the general idea I've got that the pushing surface has to go slower than the vehicle, so the prop has to actually seem a bit feckless. It's like when the cart sets off in a tailwind, it's a bit like a plane in - I'm not sure what they call it - like a deliberate stall, nose up, but falling tail first. The cart prop has got almost the full tailwind trying to turn it the other way, but the wheels force it to create more thrust against the tailwind.

I think the Racing Association might put magneto-levitation tricks out of bounds, but we can just set up our own!

 

[GregLondon]

Take your prop and mill the shaft mount hole in the centre to fit over a bolt instead, no?

Yeah, but then you need a "ground" that is basically a nut elevated up off the ground on some kind of tower, so the prop doesn't ground strike.

The corkscrew idea at least kind of has a track which sort of looks like ground.

Same concept though. prop turning a "bolt" in some sort of "thread".

I wonder if the "floating corkscrew" idea could actually work, in which case you don't need a bolt or threads. The water acts as the threads and the corkscrew turns against the water pushing the screw forward.

not sure how efficient the corkscrew would be though. A wheel on teh ground is really efficient. not much power lost in slippage or friction. A corkscrew in water is a lot sloppier I think.

But if it did work, the corkscrew would look like a vehicle, and wouldn't need a threaded track. It would need to be in a trough or something to keep it from moving sideways, though.

 

[John Freestone]

Yeah, but then you need a "ground" that is basically a nut elevated up off the ground on some kind of tower, so the prop doesn't ground strike.

A bolt, I think you mean, not a nut. That was your nutty invention! Yes. I was thinking it could be like a cable, stretched between two pilons. If it were pliable enough, it could even be hung in a big loop from pulleys and made into a perpetual...sorry...

The corkscrew idea at least kind of has a track which sort of looks like ground.

True.

Same concept though. prop turning a "bolt" in some sort of "thread".

Oh good, I'm not completely barking then.

I wonder if the "floating corkscrew" idea could actually work, in which case you don't need a bolt or threads. The water acts as the threads and the corkscrew turns against the water pushing the screw forward.

not sure how efficient the corkscrew would be though. A wheel on teh ground is really efficient. not much power lost in slippage or friction. A corkscrew in water is a lot sloppier I think.

But if it did work, the corkscrew would look like a vehicle, and wouldn't need a threaded track. It would need to be in a trough or something to keep it from moving sideways, though.

What it must be like inside that head of yours, Greg! You're amazing. You come up with really creative ideas. Sounds like it could work. It's starting to have similarities with that boat design that uses the differential flow of water and air, a prop above and one below, and it all gets rather complicated and 'sloppy' for my little brain...and something about it not being obvious what its doing anymore, where the land cart is relatively clear in what it does. I suppose it's fairly clear in a trough of still water. I suppose it loses some efficiency from that 'sloppiness'. It would tend to drift downstream rather than provide the good traction of a land cart, but I guess you could allow for that in the gearing somehow.

I think I'll keep dreaming on the land cart with a cylinder of some sort, cousin of the tumbleweed and mender's monster man-crushing wheel.

 

[spork]

Oh my god.

I've got it!

You make the corkscrew bouyant and float it in WATER.

You still have the problem that the entire cork screw would quickly be blown downwind just the little bit that it takes for the cork screw to hit the edges of the trough. At that point you're in the same old boat trying to eliminate frictional losses.

We frequently use the idea of a prop moving down a threaded rod as an intermediate step of explaining how to get from an ice-boat on a 45 degree downwind tack, to the prop cart. With bearings I'm confident this could be made to work easily. But the "vehicle" is not very general at all. It's neither a land, air, or water vehicle. It a right-hand threaded metric shaft vehicle that operates only on a shaft of the right diameter and pitch.