WingSail Design

It’s too cold and damp here to be doing much building at present so the 2M cat is on hold, however, I’ve turned my attention to designing its rig. I’ve decided that I am going to go the solid wingsail route and have been sifting all of the information that I’ve gathered and have started to sketch a design in my CAD package.

The exercise is going to be exerimental; I know I need to build light but also strong and I suspect I’ll overbuild first time.

But some specific questions if I may:

  1. How much area. A soft-sail equipped 2metre multi will have in excess of 2.5sq metres of sail area. Solid wingsails are more efficient so my thinking is for something around 1.25 - 1.5sq metres as a start?

  2. Twist. A conventional soft-sail rig has twist to accomodate the windspeed gradient as you go up the mast. Multihuls create their own wind by virtue of their speed and mini40 sails are cut flatter and generally have less twist. Full size wingsails appear to have no twist and my thinking is that I don’t need to build any in on a wingsail for a 2metre model?

Grateful for ant info or thoughts on these topics.

Ray

Sorry if this appears more than once. The web site is giving me a hard time, as it sometimes does if I take very long replying.

I am not entirely sure, but I suspect the right way to size the sail is by the maximum lift coefficient compared to that of the soft sail, assuming the soft sail is sized right. Maximum lift coefficient will depend on the configuration of sail that you pick. That’s based on the assumption that sail size is limited by stablility. Another consideration is the lift distribution on the sail. If the rigid sail is wider at the top, it may cause more heeling moment (or whatever the term is). Any scheme you do to simulate twist is also relevant, of course.

For twist, if you’re doing flaps you could make the flap narrower at the top, and that will act like a bit of twist.

There’s some neat stuff, including work on rigid wingsails, to be found at the following web site:
http://www.tspeer.com/

I think a couple of viable approaches are:
-conventional symmetrical airfoil with a trailing edge flap. If you’re really good at the mechanical side, perhaps it would be advantageous to do a two segment flap. I suspect this configuration has less drag and therefore better L/D when the wind gets too strong and you have to depower it a bit. I don’t know if a leading edge flap would be worth it at your Reynolds numbers.
-Multiple element, such as two symmetrical sections, one in front of the other, and the aft one pivots around to form a slotted flap. Check out the web site above for some of those. I don’t think the potential L/D of the isolated sail is quite as high, but a somewhat greater max. lift coefficient(I think) may be helpful when you include the drag of the rest of the boat.

An unflapped section will have less maximum lift and therefore a worse L/D, though you could of course make it wider and at least get the same amount of lift.

Obviously if you could get a smooth change in shape instead of using flaps, you’d pick up a bit of performance. But I’m not sure if that improvement would be a lot.

Given how gusty winds are near the surface, I think the boat may be speeded up if an automatic sheeting control, such as the tail on the land speed version of the Greenbird, is used. That way the sail will be trimmed right more of the time. This could also be done electronically without too much trouble, although I suspect it might not respond as quickly in most cases. (One way might be to turn the mast assembly with a gear and servo motor. The servo electronics would probably stay low, but you’d put the pot up on top of the foil with a vane on it.)

If you can, you probably want zero gap between the wing sail and the deck of the boat.

P.S. Above is somewhat theoretical. I’m an engineering type, and I fly model planes, but I haven’t built my wing sail model yet. Obviously I’ve been considering it!

Ray -

I will offer/solicit one of the members of the last Canada team that took the championship away from the U.S. in the “C” Class catamarans last (2008) fall. I know they are currently at work on a new version of their winged boat, he probably will not be able to provide the “most current” design issues, but if he is willing to visit and I think he will (was a former subscriber here) my guess is he would be happy to answer questions.

I will send him a post to see if he would consider the invite. What better person to ask, other than Duncan McLane and he is busy with the BMWO / Alinghi effort.

Here’s a photo from a different builder (Landyacht) to whet your appetite.

John's wing.jpg
You will notice the wing is designed to handle a limited amount of camber, being controlled by the aluminum plate on the bottom of the wing (near the flap-gap). I have an idea to use a variable trim for camber, but mine must be done before leaving shore. I’m guessing that without much problem, it could be adjusted on the water using a separate servo.

Keep in mind that the Reynolds number on that c-class cat is likely to be several times higher, which may mean that some other solution would be optimum. At the size I’ve been thinking about, it appears that a flat plate with a single leading edge flap is pretty good and a lot simpler than some other options. I’m sure that for a two meter cat the optimum solution will be something else.

That looks like a two element wing that’s set up so the rear element works like a slotted flap?

Thanks guys, some usefl information. My thinking thus far includes:

The wing will be a pivoted aerofoil, rather than a flapped wing. I’ll use something loke a NACA 0012 or 0015 foli, pivoted at the point of maximum chord; when the back half is pivoted relative to the fromt by around 12-15 degrees you get a good approximation to a Clarke Y aerofoil that is a good low speed lift generator. This approach appears to offer more than a flapped wing. Mind you, its easy to write about, much harder to build!

The wing will be rotated by a direct drive gearbox arrangement. Initially, at least, this will be controllled by me using proportional stick movement but it does leave things open to experimentation with with automatic ‘sheeting’.
I do share similar concerns to Lincolns though about how fast this could be made, too slow and it might be worse than useless.

I’ll probably embed a servo in the wing to control the pivoting of the rear section.

Ray

Don’t forget to drop a line to Jean Margail (WaterResist web site) - he had a 2 meter with unflapped wing so he may also provide you with direct experience comments. If you get in contact, wish him well from me. I owe him an email.

Dick

Using Profili (which is an easier way of using Xfoil), I looked at a NACA 0012 pivoted at 35 percent of chord as you say. Looks ok if turbulated, but it’s only better than a conventional trailing edge flap at high lift, say Cl of 0.9 to 1.1 or so. I also looked at the 0015, and it doesn’t do so well on max. lift at low (100k) Reynolds numbers, i.e. light air. My impression was you might be better off a little thinner, like 10 percent or so. Particularly if the rig doesn’t have a super high aspect ratio or span efficiency. I assumed a Reynolds number range of 100k to 300k. Of course a lot depends on your conditions and what the aspect ratio of the rig is.

Thanks Lincoln, it looks like I might have to look into the tools you mention, what does a NACA 0010 look like? By the way, NACA 00 profiles are thickest at 30% of chord as I recall.

Initial sketch of the wing has the bottom metre (40ins) with a constant chord of 50cms (24ins) and the top metre (40ins) tapering from a chord of 50cms (24ins) to 30cms (12ins). All sections have their point of maximum thickness vertically aligned. Thats a totla area of just ove 1 sq metre.

Ray

As I recall, there are Yahoo groups for Xfoil and Profili. Google Javafoil and you can find that program, which is handy for coming up with coordinates, such as for a section with a leading edge flap.

At Re of 100k, a 10 percent thinned version of the 0012 (probably very close to the 0010) outperforms the 0012 slightly, everywhere except at 0 lift. At least with the assumptions I used. For additional complication, a 15 percent trailing edge flap at 8 degrees deflection increases the max lift quite a bit. (These scenarios with forced transition (i.e. turbulated) at 20 percent of chord.

That’s less than a square meter, isn’t it? Also, at an aspect ratio like that, high lift is going to cost you some extra induced drag. I had assumed you were going higher. But I don’t know what the optimum is.

BTW, half a meter is about 20 inches.

My typo, should have been 60cms

Ray

Nice thread !
I’ve made" plans for RG65 wings but never went on to building…

calou

Magnus here, c-class guy.

I just got back from San Diego so I am catching up on a lot of stuff.

I am presently building a model wing for a one meter trimaran, which now has a slightly longer center hull. The boat is a few days from being completed.

I will post some photos of the boat, when I catch up to it in January.

In the mean time I will try to answer some questions etc, and throw some ideas up on the site over the next few days and weeks, when I can spare some time.

I will also try to post some design drawings for some model wings. you guys have questions, shoot away, I’ll do my best to answer them.

Wings really work, GO FOR IT!

MC

Welcome back - good to see you here again, and with just a “few” more years of experience with the wing stuff.
Sent you an email as well.
Thanks for coming back - I am positive your insight will be helpful, useful, and of great interest to the guys who can’t use a wing in thier class but are eager for “education”.

Dick

I feel a resurgence in wingsail ideas , maybe mostly from my quarter, but I am certainly interested in how the BMWO ideas would translate in RC multi’s,

I lifted the interview below from another site as I thought it answered alot of questions for me…maybe you to!, enjoy!!.

From Dailysail

Flight controller

Dirk de Ridder, the man who operates BMW Oracle Racing’s solid wing, tells us his secrets

Friday February 12th 2010, Author: James Boyd, Location: Spain
Dirk de Ridder holds considerable responsibility for BMW Oracle Racing’s result in this 33rd America’s Cup. While on normal softsail boats trimming the rig is the responsibility of a small troupe of trimmers, grinders, etc with the solid wing sail all these responsibilities fall on his shoulders, via just a piece of rope to operate the traveller (there is no mainsheet) and a rebuilt garage door remote control to drive the wingsail’s other hydraulically-operated functions.

As a sailor De Ridder has done the rounds. Nephew of the Checkmate Peter de Ridder, rather than the Mean Machine one, our man has done three Volvos on board Merit Cup, then winning with fellow BMW Oracle crewman John Kostecki on illbruck and then with Paul Cayard on Pirates of the Caribbean. He previously spent two years with BMW Oracle just before he left to join Pirates. He did a Star campaign with Roy Heiner in 2000. And to confuse the ‘de Ridder’ issue further in between all this he was a regular part of the Mean Machine crew with the ‘other’ Peter de Ridder. “It is a bit complicated,” he admits.

From his job doing traveller with BMW Oracle before he went to Pirates, when he returned to the American challenger under the new Coutts regime, his job on board the trimaran initially with the softsail rig was mainsail trim. Even this was very different to the mainsheet trimmer’s role on a monohull he says: “The traveller - you move it a little bit, but the problem is that we cant the rig so far to weather [up to 15deg] that all the load is on the leeward traveller, not the weather traveller - so you had to pull the traveller to leeward, rather than ease it. So we always ended up sailing more on the mainsheet than the traveller. It took some getting used to.”

In addition, with such a phenomenal apparent wind machine that the black and white trimaran is, the apparent wind, even when technically downwind in any sort of breeze is never more than 30degs aft of where it is upwind. De Ridder gives the example of their genniker, the shape of which he says is flatter even than an overlapping genoa on a keelboat. So the traveller would never get dumped far.

Another less obvious responsibility of the mainsheet trimmer with their softsail rig was that of managing the loadings of the boat. "When you do the mainsail, most of the loads going into the boat come from the main sheet, so that has all disappeared now [with the solid wing] because we don’t have any mainsheet load.” From more than 20 tonnes, de Ridder says that with the solid wing sheet loads rarely hit 3 tonnes. They have gone from operating the mainsheet from a top of the range primary with a 7:1 purchase, to a 2:1 on the equivalent of a pit winch. Meanwhile the responsibility for the overseeing loadings has now shifted to navigator Matteo Plazzi, who presses the buttons that operate the rig rake and rig cant hydraulics.

For someone with such a long background trimming conventional rigs, the move to the solid wing was a far from natural one, admits de Ridder. To help them get used to it, the team acquired Ben Hall’s solid wingsail A-Class cat and travelled up to Toronto on several occasions to sail Fred Eaton’s C-Class cats.

“When it first got introduced to us, it was very much all the boffins saying it was ‘perfect’,” recounts de Ridder. “Then they show you a very simple presentation of the size of it compared to say the mainsail. And you go ‘that’s bullshit - there’s no way…’ [the area of the solid wing is 50% of the softsail rig’supwind sail area]. Then you go sailing on the A-Class and the C-Class and you start going ‘that’s pretty impressive’. Then we got it in San Diego and once you get your head around it and you start believing in it, it is very impressive.”

The biggest problem de Ridder says he has found with both the softsail as well as the solid wing is simply their towering height - the solid wing is 68.5m high, almost exactly twice the LOA - and the huge wind shear than can occur between deck level and masthead. “You can have two knots on the water and 12 knots on the top. It is the same with Alinghi as us – you see pictures where it is a glass-off, but you are flying a hull, because there’s still seven or eight knots at the top of the wing. That is a beautiful thing, but it is hard with the wing - you have to have really good sensors to measure that. If you don’t have those, it makes it pretty hard.”

The solid wing also seems to defy all he has learned as a mainsail trimmer. “Especially the bottom camber is something you have to really get used to and get it explained by the experts why it has to be like that, because it doesn’t look right. If you are used to looking at a sail, it is the opposite - you have a huge amount of camber in the bottom flap and that angle goes up as it gets windier, because you twist more. So you can’t look at it like a sail.”

De Ridder reckons he has got to the stage where he could trim the wing visually, but he almost never does. “The best piece of advice I got – because we have MANY experts – was form Mick Kermarec and he said ‘just put it to the target camber, close your eyes and the sail the boat like you were doing mainsheet.’ So just trim it with the traveller and the twist controls and just do what you normally do with a keelboat with the rudder angle and boat speed.”

Having developed their targets for different wind speeds and angles during their training session with the solid wing sail while they were in San Diego, de Ridder says that Kermarec’s advice proved right: "Keep the hull out, minimise rudder angle [to reduce drag] and just sail to boat speed and then the wind angle will come on its own and its been pretty accurate actually.

“On the softsail you always look up to make sure it looks right. With this, unless a cable breaks you don’t really look up. The bottom camber doesn’t change that much and if the rake is right you just go back to how you would normally sail a boat.”

Interview continued…to long for one page it seems?.

Button punching…

As mentioned in last week’s article, the solid wingsail has three basic controls: traveller, camber and twist. The traveller is simply used to trim the whole angle of the wing to the wind. The camber control, in unison, angles all nine flaps on the trailing edge of the wing to the front element of the wing and the ‘twist’ control varies the amount of camber relatively between the nine flaps. “The whole system is designed around the even twist profile,” states de Ridder.

In addition there are two more functions. One allows the top three flaps alone to be opened. Then there is another adjuster for the mid-camber. “So you can keep the mid-leech in and open the head, if you want. Those are features we put in after sailing on the small boats. You can get it to invert as well. Let’s say you have 20 degrees of camber but you can get 40 degrees of twist. So -20 degrees at the top.”
Apart from the traveller, all the controls are operated hydraulically via eight on-off buttons on the ‘garage door’ remote. The other hydraulics on board are for mast cant and rake, the daggerboard cant and the down&^%&$ for the genoa.

The rig is completely self-tacking as the control lines are on a closed loop, so they set up automatically on the opposite tack. The ‘gearbox’ arrangement that allows this was designed by Thiha Win and as a result is known on board as the ‘Winny system’.

The rake of the solid wing is one of the most important aspects of the performance as this affects the balance of the boat and the amount of rudder that must be used, says de Ridder. “That is a lot more crucial than with the softsail, because with that you always have a front sail and sometime we do or we don’t have a front sail. So that is where the babystay gets a lot of use with getting the rake right. It is essential to the whole performance of the boat.” They can change rake while they are sailing, but generally they have fixed settings for it for different wind strengths and whether or not the genoa or genniker is being used.

Another very cool feature of the solid wingsail is how they measure the wind flow around it. To do this they have pressure sensors (mounted behind the black circles on the rig). "We used them a lot to refine the targets, because they can also calculate angle of attack on the wingmast and that is very crucial to the whole performance of the wing - the way the breeze bends around the front,” says de Ridder.

With data from the sensors, they have found that the settings for the solid wing are considerably more repeatable than they were for the softsail. “It is incredible, now that we have a good way of measuring it, how accurately you can trim it - like the hull flying wind speed, they can calculate it pretty accurately. With the soft sail that was a lot harder to do.”

Hull flying speed can be as little as 6 knots, but this varies greatly according to the wind sheer and conditions at the masthead. Sometimes they don’t fly until 8. “Especially in San Diego where there is a big different between the water and air temperature– you can have really weird days when it is hard to test.”

Another weird aspect of the rig is just how long flow can remain attached. One gets the impression from de Ridder that aspects of how the wind flows over the rig remains something from the black side. "The amount of camber doesn’t look right and it never will look right to a normal sailor. For a while we had a lot more telltales to make sure you didn’t stall it. It is amazing how high you can pull the traveller up and the flow still remains attached. From off the boat you think ‘no way - we should be going backwards now’. But it is amazing it still wants to pull forwards.

"Last week it was quite windy and we had to stop because there was a little problem. We went head to wind and I held both traveller lines tight so that the thing didn’t blow across and take people out and the wing flopped on the other tack because the camber arm was still loose and all of a sudden we were doing 7 or 8 knots almost head to wind. James [Spithill] was impressed! I was going – ‘I don’t know if I can repeat that again’! It was purely accidental that the flow reconnected. So that was pretty cool. But it was really windy. There are things like that where you go ‘wow, this thing is really cool’.”

In 20+ knot winds they stopped for lunch on one occasion with the camber fully off and the slot closed and the traveller lines loose and they were still trundling along at 8-9 knots at 20deg true!

“The thing just waves itself to weather. There is no load in anything and the thing just propels you forward. If this is the future of sailing then I think there are huge amounts of gains to be made.”

In this respect, how future solid wing sails might work, de Ridder reckons he could easily be made redundant by a computer, although the wing would need a different control mechanism. “A plane gets flown by computers.”

In terms of how the wing behaves on the race course, de Ridder says that it does improve tacking being powered up most of the way through. However significantly it also makes the whole danger moment of the top mark bear away much safer. “The problem with a mainsail is that when you ease it in more wind, it gets fuller and the forestay goes slack so the jib gets deeper so when you bear away it loads the whole thing up and you have to go through that ‘point of no return’. With this if you twist it off and pull the camber out around the top mark, it actually gets flatter and more twisty, so as you bear away it just happily sits there and nothing happens.” He says with the softsail rig they were still able to bear away in 25-26 knots, but “you have to put your balls on the table and send it! You have to commit to it.”

Performance numbers are not really forthcoming from either team, but de Ridder says that getting into the 40s is easy and on a reach they can comfortably sit at 35 knots.

Upwind they usually sail at twice wind speed. “I feel this boat is at its most scary when it has two hulls in the water when you are about to get going. Once it is up - it is a beautiful boat.”

Indeed it is.

What a great article. Thanks Morspeed!

Well, thanks to some offline dialogue with Magnus, I’m fairly well along with a desgn for a wingsail for a 2metre multi, reflecting some of the earlier input on this thread too. I’m not looking to build a pivoted foil as I outlined previously but to go for a wing/flap arrangement as per the amazing BMW Oracle Racing Americas Cup winning boat. Borrowing from Magnus I think I have the mechanism for twisting the wing sorted - remember, as per the Dirk de Ritter commentary, there is less twist the higher you go. Currently I’m working on a wing around 2.2metres long. Trying to work out how to turn a design into a build that will be strong but light at the moment.

Ray

Hey Ray,

I would be fascinated to see some pics or drawings of what you have in mind, I can kinda get my head around some ideas for controlling wing cambers but controlling the twists will take some special thinking and ideas, just curious also about the size of wing at 2.2 mt you are building, it seems very big( but then you are building a 2 mt tri) also I guess it depends on the wing area you are planning for, but i was interested in the comment by Dirk ‘(the area of the solid wing is 50% of the softsail rig’supwind sail area)’ which i translate as saying “take the sail area you would require for soft sails and halve the area needed for a wing”.

cheers…Gary