Wing Sails on Proas!
I think the drawings as you have them are correct - with the tail angled the right way to give lift on the main wing in the right direction. If you don’t want to cross over the control rods (cables?) for the tail then you could put the pivot for the square stirrup towards the leading edge of the wing. This would also help the counter-balancing a (very wee) bit.
As far as I can see, the only, trivial, drawback to the design is that you can’t get underway if the wind is dead ahead or astern since moving the cam doesn’t move the stirrup. For all the times that is going to happen you can carry a paddle… :o)
Peter
In other news, Peter shared this drawing with me (see attached). It’s possible to have both a main wing AND a flap/second element, BOTH controlled by a single tail vane. With this, on offwind courses you can camber up the wing and get much higher lift coefficient, or “fine” it out and minimize power. It’s not optimized, but it’s gonna be a bunch better than no second element, and hey, it’s all automatic, so what’s not to like?
Dave Culp
Hi Dave,
A Peter may well have shared the drawing but not the Peter you quoted…
My drawing looks like this - first attempt failed because I attached a pdf :o(
I’ll try again…
I may even be able to make a case for building hulls of the foam/heat-shrink though I’m not sure just yet. The engineering pencils out but the reinforcements, mast step, foil and ama mounts etc. labor and costs may push the final cost higher than “disposable” warrants.
A possible route for a foam vaka such as you propose is to rig the boat as a schooner and tie (in engineering terms rather than with rope) the masts, akas and lee boards/rudders together so the forces are transferred directly from one to the others. This leaves the vaka with little to do but provide buoyancy, hold stuff (including people perhaps) and be streamlined. It would probably need well distributed attachments with a large total area to minimise point loadings, though.
The schooner rig would also afford a lower heeling moment or (for the elite) more power. BTW I’d always imagined that the principal source of heeling moment was not so much the drag component of the aerodynamic force as the transverse component of the net wing force. Or do I have it backassward ?
I’ll try again…
Peter,
Sorry but I am a bit thick…. Could you please explain further?
When I look at your schematic and at your video (see below)
http://www.youtube.com/watch?v=jGo4bRHI22o
I have a hard time to correlate / link the 2 together.
So here is what I think is happening; please let me know if I am correct.
On your schematic, the boat is pointing towards the TOP of the drawing in all 3 cases. Right?
The boat is on a beam reach, port tack. Correct?
The main element of the wing pivots around the mast, symbolized by the big blue dot. The rectangular frame on your video is represented by the black oval. Right? That frame is tied to the bottom of the wing on a pivot point, the small black circle, at the back of it.
The red circle is equivalent to the ring that you can move fore and aft on the video. Did I get all of that right?
If I am correct so far; The top configuration represents “forward” (therefore the arrow on the left side, pointing up, in the direction of the front of the boat. Moving the circle forward, created a tilt of the black oval, pivoting around the small black circle. the side lever attached at the back of that black oval pulls the control rod, which “camber” the wing by acting on the rear flap in one direction and turn the tail in the opposite direction to induce an angle of incidence of the complete wing. I assume that we can consider that the tail (the profile at the extreme right of the schematic) is directly pointing into the wind.
In the middle schematic, we are in neutral; the tail, the wing and the flap are aligned and pointing directly into the wind. Yes?
In the bottom schematic, it is the opposite of the top schematic. If the front of the boat is still pointing to the top of the drawing, the wing will push the boat backwards.
Please confirm that I got all of that right.
2 more questions; on the schematic, the push-pull rod crosses over from the main element to the tail. The lever on the wing is on the left side, the lever on the tail is on the right side. On your video, they are on the same side… Why?
I suspect it does not change anything other than what direction the red circle needs to be moved to go forward or backward. Basically, if you reverse the attachment of the rod, you have to reverse as well the position of the circle to obtain the same effect. Right?
Last but not least, on your website, you show a control box which is not “forward-neutral-backward’ but instead “port tack-neutral-starboard tack”.
Has your control system evolved since then?
If I got all that right, your system is brilliant.
If I did not get that right, your system is most likely still brilliant; I just didn’t get it!!!
In the case where you have a flap and the tail attached to the rod, the amount of “off-centering” of the red circle will control both the camber (angle of the flap) and angle of attack. Correct? However, that angle of attack and camber will also be dependent on the wind direction. Yes?
Even with the red circle “in full tilt forward”, when sailing close hauled, the red circle can only generate a moderate off centering of the black oval frame, and therefore only a small angle for the tail, compared to the wing. You will get small camber and small angle of attack.
The opposite is on a beam reach, where the same off-centering of the red circle will have maximum impact on the black oval frame.
I am not saying it is a bad thing, I am just trying to make sure that I got it all correct.
I am looking forward to hearing from you,
Laurent
Peter,
Sorry but I am a bit thick…. Could you please explain further?
When I look at your schematic and at your video (see below)
http://www.youtube.com/watch?v=jGo4bRHI22o
Hi Laurent,
Firstly, I am not Peter Worsley so the video is not actually mine. You’re right - on the schematic in your post the boat is pointing up and down the page and the wind is coming from the left. The mast is blue and the cam is red so you have got all of that right.
The top configuration is for the boat moving up the page. The cam ring is positioned towards the direction of the travel and the tail is tilted to give the main wing the appropriate angle of attack to drive the boat up the page. The flap at the rear of the wing is tilted, to produce camber, by a short arm linking it to the control rod. The tail has a small angle of attack because the flap will produce a small moment of the wing about the mast and that has to be countered.
The middle configuration is feathered with no angle of attack for the wing, flap and tail.
In the bottom schematic, it is the exact opposite of the top schematic – the cam is ‘down’ and the boat is travelling down the page.
Even with the red circle “in full tilt forward”, when sailing close hauled, the red circle can only generate a moderate off centering of the black oval frame, and therefore only a small angle for the tail, compared to the wing. You will get small camber and small angle of attack. The opposite is on a beam reach, where the same off-centering of the red circle will have maximum impact on the black oval frame.
Yes, that is the consequence of linking the flap angle to the angle of the oval. That said, when close-hauled most of the lift is trying to lift the ama rather than moving the vaka forward. On a broad reach more of the power is moving the boat in the desired direction and so a full-flap, maximum-lift configuration is appropriate. If anyone has a bright idea as to how to flip the flap from full-positive to full-negative as the wing passes through neutral, please share it here…
In short, Laurent, you have understood my diagram perfectly and rightly upbraided me for not explaining it myself. As to why the control rod crosses over in this layout and not in Peter Worsley’s – having looked again at the video, it seems that the cam moves in the opposite direction to the boat. The cam moves forward to move the boat backward, that’s all.
hope this helps
Peter (not Worsley)
Nope, I think Skip’s backwards, at least according to Peter’s YouTube vid here: http://www.youtube.com/watch?v=jGo4bRHI22o
Dave Culp
You are correct, and not the first to comment so 😉
Either will work, Peter W’s is better, particularly the geometry. My problem was thinking cam forward to go forward, works much better the other way. That is, cam backwards to go forward.
Cheers,
Skip
Either will work, Peter W’s is better, particularly the geometry. My problem was thinking cam forward to go forward, works much better the other way.
There has to be crossed-over geometry in either the tail control rod or the flap control rod (see
<a href="https://proafile.com/?URL=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DAArllt2TRZo">http://www.youtube.com/watch?v=AArllt2TRZo</a>at 2’14”). I’m mystified why one should be so much better than the other - can you elaborate ?
thanks
Peter
... snip…In short, Laurent, you have understood my diagram perfectly and rightly upbraided me for not explaining it myself. As to why the control rod crosses over in this layout and not in Peter Worsley’s – having looked again at the video, it seems that the cam moves in the opposite direction to the boat. The cam moves forward to move the boat backward, that’s all.
Peter (not Worsley)
Precisely Laurent. I think we are being mislead, a bit, by Peter Worsley’s 2 videos cited in this thread. In this one: http://www.youtube.com/watch?v=jGo4bRHI22o There are only 2 elements, not three. There is only the single-element wing and the tail. The tail appears to be a flap because it is large and close to the wing. Peter did this to try to limit the rig’s radius to the boat’s beam. In fact he’s a little bit wrong (per Tom Speer); a much smaller tail on a much longer stalk will be both more efficient (“rob” less thrust from the main wing) it will also be more effective at damping oscillations or flutter. It’s not too far-fetched to imagine a folding (perhaps to near vertical?) or telescoping tail boom which might fit in a berth—and also have sufficient damping not to get the rig fluttering at the dock.
In this vid: http://www.youtube.com/watch?v=AArllt2TRZo Peter shows a rig with 3 elements, Main, flap and tail.
There is much more about this setup—including input from Peter W—over on the AYRS yahoo group.
Dave Culp
A possible route for a foam vaka such as you propose is to rig the boat as a schooner and tie (in engineering terms rather than with rope) the masts, akas and lee boards/rudders together so the forces are transferred directly from one to the others. This leaves the vaka with little to do but provide buoyancy, hold stuff (including people perhaps) and be streamlined. It would probably need well distributed attachments with a large total area to minimise point loadings, though.
The schooner rig would also afford a lower heeling moment or (for the elite) more power. BTW I’d always imagined that the principal source of heeling moment was not so much the drag component of the aerodynamic force as the transverse component of the net wing force. Or do I have it backassward ?
Wandering kinda off-topic, but I’ll try to bring it back: Peter first sent this to me as a PM yesterday. In responding, I was reminded of boats build in the long past much like he describes. The most germane was a proa (of course!) with inflatable hulls. These were sewn of 4 and 6 oz sailcloth and we re-purposed inflatable kayak bladders for airtightness. I think today I’d consider heaviweight blue-tarp and stacked WalMart air mattresses… 😉
At any rate, we formed longitudinal “pockets” in the hulls’ decks, just right to take a piece of wood (in this case 1/4” x 2.5” oak flooring, on a 14’ X 1’ hull. Easy to insert uninflated, locks into place on inflation. Overall, very light and stiff. We then had “hard points” to attach the hulls to the crossbeam structure which would carry the forces efficiently throughout the inflatable hull.
In a different boat, we built the entire crossbeam/rudders/winches/benches/etc into a metal frame—a weldment got out of 2” muffler tubing, though there are other ways—and just set this whole thing onto the rather fragile hulls (not inflatables in that case, but similar engineering) This worked a treat and lasted many hours (we just barely dumpstered it, obsolete, before it rusted out.)
I agree with others here that an unstayed wing sail has advantages, which means a heavily reinforced stub mast out there somewhere between hulls needs to be built. Incorporating this into the remainder of the strength needs, as Peter “not Worsley” suggests would be a good idea, I think. Cleaner and simpler on a proa, though, as the mast can go into the main hull’s bilge.
Dave Culp
Some notes on the (im)Practicality of wing sails:
Others have related most of the old saws about capsize danger, raising/lowering, reefing, storage/transport, cost and weight issues (did I miss any?) I’ll touch each if I may. You’ll note a recurring theme; let’s look at comparable challenges between wings and masted sails, not absolutes, and let’s look at an evolutionary route, not a fully-formed pandemic solution to every potential problem.
Capsize danger. Yup, it’s there. Same as in your garden variety multihull. Jam a winch and you can’t release the sheet and over you go. Let the main all the way out to the stays on a run and you can’t get the sail down—and over you go. Sail into the “broad reach multihull trap”, where you’re over-canvassed but holding your own by bearing further and further away—until you realize there’s no more to give ‘cause you’re on the edge of pitchpole—and you also cannot come up or the fast-shifting apparent wind will capsize you—and over you go. So sure, a jammed wingmast could spell catastrophe at sea. Be a good idea to keep those bearings in good repair, wouldn’t it? $10 million AC boats with 13-story wings might want to leave someone aboard on watch, but I don’t think a 16’ proa will have the same issue. It is quite common for the AC45 fleets to leave their boats on moorings—interestingly they use large bags of fresh water as mooring buoys-these are hard to drag around, but *can* be, so make good shock absorbers to keep the boat from “hunting” around its mooring (fresh water floats on salt)
Raising and lowering a wing mast is very similar to raising and lowering an aluminum mast. Best to do it facing upwind (ashore or afloat), use the same tabernacle, gin-pole, trailer winch and a pair of stays, precisely the same as raising a mast, and up it comes. There are numerous vids on YouTube showing this—and these are massive 300 sq ft wings more than 45’ tall. I’ve personally seen it done in 30kt gales, with smaller wings.
Reefing. I think I mentioned that SailDrone has more than 6k offshore miles on it without ever reefing—and their current mission is to try for bigger and bigger storms to drive it through, as further proof of concept. Richard Jenkins—the head of that effort—has some experience in wingsails—he designed and built GreenBird, the current record holder for fastest landyacht, ever. At one point, Richard had GreenBird’s wingsail disconnectable into two parts—upper and lower. He could then feather the upper half and only bring on the lower (using a tail plane, remember), thus only half the rig.
Storage/transport. It’s a good idea to split the wing into 2, even 4 major pieces, especially if it’s of any size. Richard’s is a good approach for a single element wing—you could almost as easily cut the wing horizontally into 3 as into 2 and have even more “reefing” choices. Second, it’s a good, practical idea to split the wing into 2 elements. The resultant pieces are much smaller, thinner and lighter than the single element. The connectives aren’t tricky or fragile; the C class cat guys have had it all figured out for many years and have fully published it for free—including a remarkably simple and precise twist control system, if that interests you (it doesn’t me—if a wing gives me 20% more power and speed, I believe I can forgo the additional 2-3% I’d get by precisely controlling twist. Plus the C-cat guys offer even simpler twist systems which aren’t adjustable. YMMV)
Weight and cost. Again, yup, they’re issues. Design ‘em light and strong, then re-design them light, strong and of cheap materials. I’m working on it…
Dave Culp
Either will work, Peter W’s is better, particularly the geometry. My problem was thinking cam forward to go forward, works much better the other way.
There has to be crossed-over geometry in either the tail control rod or the flap control rod (see
<a href="https://proafile.com/?URL=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DAArllt2TRZo">http://www.youtube.com/watch?v=AArllt2TRZo</a>at 2’14”). I’m mystified why one should be so much better than the other - can you elaborate ?
thanks
Peter
If the circular cam actuates the follower by sliding “sternward” the distance from the follower pivot to the tangent point on the cam is much shorter so you have more angular movement of the follower when the cam actuates by moving closer to the follower pivot.
There may be more to consider that I haven’t thought of yet, will wait til I’m closer to implementing the deal on the Broomstick, still have some to do finishing the boat and trialing the first rig.
Cheers,
Skip
If the circular cam actuates the follower by sliding “sternward” the distance from the follower pivot to the tangent point on the cam is much shorter so you have more angular movement of the follower when the cam actuates by moving closer to the follower pivot.
Skip
Ah, of course. Brilliant. You could get even more angle with a larger cam (and follower), no? Plus, you still have the ability to reduce the angular movement any time you want, via adjustments on the horn(s). You can also make the tailplane angle much more, or much less, than the flap angle. It’s all in adjusting the effective horn lengths (row of holes for attaching the push/pull rod)
Whatever you set it for limits what you get for today’s sailing—unless you incorporate some more complex scheme for changing the horns’ lengths so you can alter it while sailing—but I expect that, for many conditions, approximately is good enough. As you bias the whole system (via the Bowden wire) you’re setting camber and at the same time angle of attack You are then free to sail the boat on any course you like—so long as you maintain positive thrust—or even if you don’t. The rig does not care; it will drive you in reverse as easily as forward.
As you set the tail to go closer to wind, the wing automatically flattens. As you set the tail for higher angle of attack, you automatically increase wing camber thus power (and therefore limit how close to wind you can sail, but at much increased lift coefficient). Speer points out, as does the Editor’s link to wing building earlier in this thread, that if you set the hinge point between the wing elements *ahead of* the trailing edge of the forward element, you automatically and progressively open the slot at higher camber, and progressively and automatically close it up as you flatten the wing. Precisely as the aerodynamics want. Last, Peter’s system is auto-tacking. When you tack the boat under the rig, the whole thing automatically reverses for the other tack. The simplicity is compelling.
Dave Culp
Somebody mentioned that, from practical considerations, wingsails offer little or no advantage over soft sails for proas. I disagree. Wing sails can deliver far greater lift coefficients than soft sails (2-3 for a simple 2-element wing without twist control, versus 1-1.5 for a typical mainsail or jib).
Hi Dave,
I agree with much of what you say with regard to wings with tailplanes being a good option for proas (or any other boat for that matter) if you want good control without the need for reefing. However (as usual 😊 ), I have to disagree with a few of your points.
Firstly, while slotted multi element wings can deliver high lift coefficients, there is still a high drag penalty, so the high lift coefficients cannot be used when sailing upwind and you have to throttle the wing back by reducing the flap deflections to get good upwind performance (as you state in your next paragraph). Secondly, while a slotted wings lift sounds good compared to a main OR jib in isolation, a main/jib combination is in fact a slotted wing, and the total lift coefficient is greater than that of either component. A suitably configured ballestron rig could potentially deliver lift coefficients approaching those of slotted wings for downwind courses.
Second, the greatest requirement for righting moment for any sailcraft is when hard on the wind, therefore at the rig’s highest L/D. At this condition there isn’t as great a difference in the wing’s versus soft sail’s lift coefficient, but as the wing’s drag is much less (half, even a third), net overturning moment is much reduced. For a given amount of thrust delivered into the boat, the wing sail has significantly less overturning moment—the boat heels less. Put another way, for a given overturning moment, the wing can deliver significantly more thrust before the boat capsizes than the soft sail.
Unfortunately, simple trigonometry will show that the overturning moment is not greatly reduced. It may be marginally reduced, but not significantly. Regarding your claims of drag reduction, a wing will have significantly lower drag at low angles of attack and low lift coefficients, where parasitic drag is the predominant factor. However, at the lift coefficients at which sails are commonly operated (1 to 1.5), the predominant drag fraction is induced drag, which is the same regardless of whether the wing is a soft sail or a solid wing, so claims of half or even a third of the drag may be misleading. The best L/D ratio for any wing/sail is usually achieved at around the CL 1 to 1.5 range, hence why it is better to reef a sail and keep it operating in this range (with the added benefit of lowering the COE) than it is to feather it (operate it at a lower CL).
These two features—higher lift coefficient and reduced heeling moment per unit thrust are additive—for the same “horsepower” you can have a shorter rig, delivering more power per unit area, with much-reduced overturning moment. What the designer does with this has many options—you can build a lighter, less rigid boat if you wish. Or a free-standing rig with significantly less reinforcing. Or a narrower boat. Or you can pile on all the wing you can carry and sail very fast. Or you can build a smaller rig (20%?) and spend less money on your boat. I’m not making this stuff up. Every study performed in the last 70 years has decided in favour of wings over cloth and every class of sailing yacht which is allowed to carry them do horizon jobs on their soft-sailed competitors.
Having mentioned above that the high lift coefficients obtainable by slotted wings can only be utilised downwind, if you want to maintain upwind performance comparable to a soft sail, you need to carry a similar sail area, so the option of reducing the rig size doesn’t really apply.
As far as I am aware, the only long standing class that uses solid wings is the C-Class cat. The use of wings on the recent AC boats doesn’t prove anything because no-one sailed without one. They didn’t evolve within the class and I suspect that their main use is to hype them up for the TV audience. Recently, solid wings have been tried on A-Class cats and Moths. On the Moth there was no detectable benefit and the slotted wing was subsequently banned. Single element wings are allowed but we have yet to see one. Some are rumoured to be in development. The A-Class wing of a few years ago proved problematic, with no significant benefit and there has been no further development of it as far as I know
As I stated in my first paragraph, I can see the benefit of using a simple solid wing with a tailplane and I think it would be a good rig for a proa for ease of shunting, mitigation of backwards capsize, no need to reef etc. but currently there is a cost to these benefits of overall reduction in performance compared to a standard soft sail rig. It would be good to see the solid wing developed to the point where those benefits could be realised without significant performance, practical or cost penalties. Until then, I would prefer if people were given realistic expectations for wingsails rather than hype 😊 .
This is getting good.
This is the way I see wingsails on real boats. Unlike the glamour of the AC and Little AC cats, wingsails are NOT for high performance. They are for solid, easy, dependable, predictable thrust, come what may, hurricane or zephyr and anything in between. They are perfect for a wind powered drone which has limited ability to make mechanical changes to the rig “on the fly”. This sounds good in theory, but if you are the “drone” sailing across an ocean, you are far more capable than a few servos and a few strings of programming to make changes that benefit the ship. Adjust sail area, tweak the rig, fine tune. What the hell else are you going to do out there? And if it gains an extra knot of or two of traversing the leagues over the sea, then why not?
However, there are plenty of uses for relatively “mindless” wind power. Shipping being the big one. A short handed, limited skill crew could sail a wingsailed vessel from here to there pretty easily and safely, at least a lot easier than a clipper ship.
Combine the wing with an efficient platform (proa), and you have a going shipping concern. Solid, dependable, predictable.
Now, combine that with the Dave Culp “Kite Tug®” concept, which is essentially a highly skilled band/union of mercenary wind mechanics/pirates who can hook up to your vessel and double the speed (for a price) and we have a viable wind powered shipping economy, or at the very least the start of an excellent graphic novel!
Yes, that is the consequence of linking the flap angle to the angle of the oval. That said, when close-hauled most of the lift is trying to lift the ama rather than moving the vaka forward. On a broad reach more of the power is moving the boat in the desired direction and so a full-flap, maximum-lift configuration is appropriate. If anyone has a bright idea as to how to flip the flap from full-positive to full-negative as the wing passes through neutral, please share it here…In short, Laurent, you have understood my diagram perfectly and rightly upbraided me for not explaining it myself. As to why the control rod crosses over in this layout and not in Peter Worsley’s – having looked again at the video, it seems that the cam moves in the opposite direction to the boat. The cam moves forward to move the boat backward, that’s all.
hope this helps
Peter (not Worsley)
Well, I almost understood it all. I did not get that the bottom schematic was with the boat sailing to the bottom of the page (as indicated by the big arrow on the left… duh). But it makes sense now. The cam (red circle) is still pointing forward, and the boat is again on a beam reach, but on the opposite tack of the top schematic.
Regarding the issue of the flap (and tail?) going from full negative to full positive, as pointed out by Dace Culp, I don’t think it is a problem if you tack, right? As the boat turns “underneath the wing”, even with the cam kept in the forward position, when the boat is point dead into the wind, the wing will be pointing into the wind as well, with no camber, all 3 elements (wing, flap and tail) into the wind. It is basically the same schematic that the middle drawing in your attachment, except that the cam would be off centered… to the left.
Rigth?
This is another story for jibes. As it can be seen on the videos from Peter Worsley, in that case there is the tail caught backward, until the wind angle is such that every thing flips around, like a normal jibe….
I think that there is a simple “cure” for this, even if it defeats partially the “set it and forget it” approach of the cam system. If you want to jibe, let’s say from a deep broad-reach on starboard to port; why not do the following?
- you start with the tail on the port side of the boat (you are on starboard tack)
- bring the cam to neutral; the wing/flap/tail flattens out and weathercock. The tail is now above the front/port side of the boat, the whole wing is pointing towards the aft starboard side of the boat
- steer the boat on the new tack; with the wing still weathercocking, now pointing to the aft/port side of the boat
- reset the cam to the forward position; the sail should build up camber and angle of attack on the new tack.
Thinking about it a bit further, I do not know if it would be better to time the repositioning of the cam to “forward” position when you are dead downwind, in the middle of the jibe. Maybe yes…
So it negates the “fully automatic/mechanical/no adjustment while sailing” aspect, but it still seems pretty simple and mundane as a jibe, with no drama, whole sails and boom swooshing by above your hear.
If we have an unstayed rig, with NO control lines going back to deck, etc, which would impair true 360 capability, let’s take advantage of it!
Cheers,
Laurent