Wing-drive

Apart from recent mention of Reynolds number by JT! (which was for ~6kt wind, by the way), It is pretty sad to see all the misguided argument from amateur mathematicians. Please spare us argument about fluid dynamics, unless you present all the relevant variables for our consideration.

ALL of you should read this (which only took me 5 minutes to find on google, notwithstanding 1.5 years of debate in this forum) :

http://www.soe.ucsc.edu/~elkaim/Documents/GabrielElkaimThesis01.pdf

For those of you with short attention span, see the extract published in the AYRS’s Journal ‘Catalyst’ which can be found at:

http://www.soe.ucsc.edu/~elkaim/Documents/Catalyst_BoatArticle.pdf

 

I find it very interesting that the author claims a maximum practical CL for a main and jib is .6 -.8. There is no cite to support this statement. The paper then goes on to state that the wing has CL of 1.8 and a L/D of 10-30 and no support for that statement either.

Various wind tunnels have the max CL of sloop rigs somewhere between 1.6 and 2.0.

This wingsail differs from the WingDrive in that the wing has a 13% flap. The Atlantis uses the flap to effectively camber the foil on each tack.

The wing may well operate at CL = 1.8 as stated, but the area of the control foil is acting opposite so the overall lift / total area is less than 1.8.

The paper is very interesting. Thanks for the link.

 

What's drag got to do with it?

I'm not a "sailer", let alone a sailor, although I have studied aerodynamics extensively as it pertains to aviation. I have what I suspect is a pretty dumb question and I regret having to ask it, but until I understand this one thing, I just won't quite be able to get my head around this topic...

What does drag have to do with anything on a sailboat or saildrive? In simple terms, here's how it appears to me and I need someone to explain something simple about boats to me so I can understand:

Since a saildrive/boat is sideways relative to an aircraft, "Lift", be it from a sail or a wing, creates thrust.

But what does "drag" do? In an airplane, drag acts against the thrust of the engine, limiting top speed. But a sailboat isn't being propelled. I tried thinking of it like a sailplane, but sailplanes also have "thrust" in the form of potential energy (in the form of altitude). So no matter how you slice it, an airplane or glider must counteract drag in order to save its energy to travel faster or farther. So low drag is good. It means you can go forward faster for your given level of available potential energy. This is the same regardless of whether your P.E. is in the form of altitude or gallons of gas.

Now, anybody who's taken geometry knows that you can break a vector down into two components, a horizontal and vertical (e.g. lift and drag). And it is indisputable that the "drag" vector acts 90 degrees to the "thrust" vector. This means that the drag induced by the wing of a saildrive/boat pushes the boat sideways. In an aircraft, the induced drag directly robs the plane of potential energy. It acts against thrust. However, in a saildrive/boat, the wing or sail's drag vector acts perpendicular to the ship's own drag vector- this is *very* different from an aircraft where the fuselage's drag vector and the wing's drag vector are parallel to one another.

Therefore, in a boat, the wing/sail's drag vector doesn't rob a boat of any potential energy directly in the forward direction, it only robs potential energy indirectly by threatening to capsize the boat... which then needs to be offset by the keel using counterweighting or "underwater lift" on the keel which seeks to upright the craft against the "drag vector" of the wing/sail.



So what's all this talk about lift/drag ratios? Seems to me that just so long as your keel can keep you from a capsize, it doesn't matter one whit how much drag the wing/sail is producing... at least not until the gains in "lift" generated by the wing/sail is offset by the drag of the boat being nearly sideways in the water. What's wrong with my misunderstanding of this subject?

 

Venetian Blinds?

The more I think about it, the more convinced I am that my assessment is more or less accurate. L/D ratio, as it pertains to a wing, is irrelevant to boats so long as you have enough keel and freeboard to counteract the rotation along the longitudinal axis.

Now, having spent extensive time studying flap arrangements for slow aircraft. And having nothing except intuition to tell me that a Cl of 2 sounds reasonable for a sail and being in agreement that 1.5-1.8 is a reasonable range of Clmax for a typical airfoil, I have a thought...

Why not create a venetian blind sail?

If I recall correctly, some studies were done many years ago that indicated that a series of "venetian blind flaps" can attain a Clmax of about 3. Of course, they've never been fitted to an aircraft because the retraction mechanism would be a nightmare... but sails don't have to extend and retract at the flip of a switch...

For application on a sailboat, imagine a vertical set of venetian blinds... these are just totally wild guesses, but imagine a series of 12 "blinds" with each successive blind having a greater AOA than the previous blind. Maybe give each of them a chord length of about 6". Make them all maybe 12 feet tall. They'll form an arc, rather than a straight line at the bottom.

Yes, it will create a lot of drag, but the amount of lift could be stupendous!


Edit to add...

Wow. The more I think about this, the more intrigued I am with this idea. Yes, it would be heavier, however, with a greater amount of "lift", you could use a smaller sail area. Or, better yet, you could make the sail shorter and wider. This would reduce the ability of the sail to capsize the boat, thereby allowing you to use *more* sail area. I certainly don't know enough about boat design to really go into depth with this idea but, to me, on its face, it seems quite interesting.

 

I dont have aerodynamic qualifications but have read a fair bit, and sailed many different craft for a long time, and some of your statements/assumptions appear wrong to me.
(1) " But a sailboat isn't being propelled" If not, then it has no possibility of moving forwards.
(2) You are resolving the sail lift vector into one that acts along the centreline of the hull directed forwards and then resolving the drag vector to act at 90 degrees, that is directly sideways. This ignores the component of the lift vector that acts sideways contributing to heeling. It also ignores the component of the sail drag vector that acts along the centreline of the hull directed backwards.
(3) You assume that a large heeling force as long as it is counteracted by a sufficiently large keel does not influence boat speed. By this argument there is little upper limit to the speed of a keelboat planing hull as a hugely oversize sail will be balanced by a hugely oversize keel, without adding to drag forces, that would slow the hull, and in practice attainable speeds by planing keelboats built to take full advantage of your theory should easily exceed those of the fastest dinghies in most conditions. This is not borne out in practice. Maybe someone will explain all the maths involved here.

 

Thanks for the help... Allow me to go off on your points and seek further understanding.

 

Check on the vectors.

Toot, I am not getting into esoteric factors here as this is not my field. You need to check the basic vectors. Roughly speaking if you have a curved thin sail with maximum camber around 33% back from the luff, and sheet it so the boom lies at around 20 degrees to the centreline.
From the the maximum camber point, draw a line perpendicular to the sail surface at that point, it will be roughly heading towards the front of the hull at around 45 degrees. This is the lift vector. The drag vector is more complex to show in diagrams but is related to loss of laminar flow off the leeward sail surface, turbulence just behind the mast, skin drag, on sail surface. Lift force developed near the leech which acts approx. at right angles to the centreline of hull but sometimes even slightly rearward.
The resultant sideways forces then produce forces on the keel which in turn produces hydrodynamic drag; the greater the keel load, the higher the drag. As I said before, I am not an expert on this, but have a general working knowledge which gets me by in most situations.

 

I don't want to get into esoteric factors either. I'm just trying to get a grip on all this.

So am I correct in understanding your post that modern convention when designing a sailboat is to treat the forward (thrust) vector as being something other than directly straight ahead?

For example, an aircraft wing is angled upward along the longitudinal axis- in other words, the tip is higher than the root. However, it is common to resolve this vector- Instead of treating it like a vector on each side acting at an angle, you resolve the vector and treat it as two separate forces- one acting exactly perpendicular to the thrust vector, and one acting exactly inward towards the fuselage- basic trigonometry. Mathematically, it works out to the same thing, but as a practical matter, it's easier to deal with the resolved vectors than to constantly apply a correction factor to a long series of calculations.

Boat sails aren't treated in the same fashion?


I suspect they are. I also suspect we are saying the same thing, I'm just breaking it down a tiny step farther. Seriously, thanks for the help so far.

 

The simple analogy is a glider.

A glider uses a portion of the lift to provide thrust. The lower the drag of the complete aircraft the less thrust is needed, so the glide slope is flatter.

Sailboats use a portion of the sail's lift to move the boat. The rest of the lift is wasted as heeling force. The limit to how close to the wind a boat can sail is the L/D of the sails and rig.

Higher L/D = sailing closer to the wind in exactly the same way that higher L/D = flatter glide slope for an aircraft.

 

good start ... but ...

Sailboats don't always trim for CLmax. Sailing upwind is at L/Dmax not CLmax.

When reaching, Max CL = higher boat speed with the limit being the RM (Righting Moment).

For displacement hulls, a high CLmax is only needed until the hull is at S/L 1.2:1 or better. In light air you have a choice of large sail area operating at moderate CL and high (relatively) L/D, or moderate sail area operating at higher CL and lower L/D. The SA/D ratio (sail area/displacement) is an indicator of power to weight. Higher SA/D should be faster and point higher in light air.

As far as the venetian blind concept goes, there are a couple of things working against it. In an airfoil array each trailing foil has to be trimmed at a different angle to get the same AOA due to the influence of the leading foil. Sailboats must tack, so the system must be able to be trimmed with the wind coming from either side.

The total induced drag of the system will be equal to the induced drag of the total area and total CL. The A:R is total area and maximum span. To use the venetian blind concept on a high aspect ratio rig (to keep L/CDi high) the foils would have very small cord and low Re. This makes structure and airfoil choice more difficult.

A real life demonstration of the venetian blind concept is the old multi masted schooners. Schooners are Very fast sailing off the wind, since they can generate high CL. Upwind they suck because the A:R and L/D are low.

 

Everything.

When you draw out the typical vector diagram of the lift and drag of the hull and rig, it looks like this:

As long as the apparent wind is ahead of the beam, aerodynamic drag slows you down. But it's more than that.

The angle, beta, between the apparent wind and the boat's course through the water is crucial to sailing performance. The smaller beta is, the faster you go on any point of sail. The fundamental sailing performance relationship is:

Vb = Vt * sin(gamma - beta) / sin(beta)
Vb = boat speed
Vt = true wind speed
beta = apparent wind angle
gamma = course relative to the true wind (point of sail; 0 = head to wind)

For pure speed, you can always choose gamma so sin(gamma - beta) = 1. You just sail below a beam reach by the amount of beta. As beta becomes very small, the speed becomes very great. All this is geometry - it comes purely from the wind triangle.

If you do a little more geometry, you'll find that beta is the sum of the "drag angles":

beta_aero = arctan(aero_drag / aero_lift)
beta_hydro = arctan(hydro_drag / hydro_lift)
beta = beta_aero + beta_hydro

Since the drag angles depend on 1/(L/D), if you know the L/D's, you know the boat's performance. Making beta small means making the L/D's large.

Notice that the magnitude of lift or drag per se doesn't matter. It's the lift-drag ratio that counts. And that's why people are so concerned about L/D's. For nearly all sailing craft, reducing drag has a far bigger effect on performance than increasing the lift.

 

Wow wow wow! I have a woody right now and I'm not talking about a boat! This is exactly what I was trying to sort out in my head. You all have made it MUCH more clear. Thanks a lot!

RHough- you provided a great "practical theoretical sailing" tutorial that explains to me actual mathematical practicalities of theoretical sailing and your partial "shooting-down" of my venetian blind proposal in a problem-solving manner helps because that's pretty much how I come to understand *everything* around me... you know, asking "why?" and investigating it until I finally getting frustrated with every other possibility and come to the point of throwing my hands in the air and saying, "every other option sucks! And that's why they do it the way they do!". Well, either that, or else come up with alternatives that are worth investigating. And tspeer- you provided the basic theory that allows me to make sense of what RHough is saying. I was actually about to draw out that diagram from my own head (incorrectly, no doubt), scan it, and have you all correct it for me. What a great wealth of information! You two guys saved me a lot of frustration there with it. Thanks. And you both added some words to my vocabulary.

I've now got my old trusty scientific calculator here.... I'm going to go play around a bit with this and then pester you all with more questions. In the mean time (and at the risk of going slight on-topic), I dare say I can see how a flexible wing (e.g. sail) is much more efficient at handling a relative wind with a variable angle, as encountered in sailing, as opposed to that encountered while flying. Which, I suppose, is why my airplane's wings don't look like your boat's sails. But my words aren't to disaparage Mr Kjell either. I also see how his idea is a lot less complicated in practice and can work to a significant degree.

One thing really quick though... The angle between Leeway and the boat's longitudinal axis could be described as the "slip angle", right?

 

Leeway angle

Thank you for the kind words.

The way I think of it is that the leeway angle is the glide slope of the underwater foils.

The hydro lift must equal the athwartship vector of the sailplan. In glider terms the sailplan lift at 90 deg to the hull is gravity, so hyrdo lift must equal weight in aero terms.

The higher the L/D of the underwater foils the lower the leeway angle.

The aero L/D gives the angle that the boat can point into the wind. The hydro L/D gives the course over ground angle into the wind.

If a boat can point to a 30 deg angle to the wind, and it takes a hydro AOA of 5 deg to balance the sail force, the course over ground will be 35 deg off the wind. The boat's heading will be 5 deg higher than the course sailed to provide the AOA needed for the keel to work.

Improving the aero L/D allows the boat to head closer to the wind, improving the hydro L/D reduces the leeway angle and allows the course made good to be closer to the wind. With all else equal, the boat with the higher hydro L/D will make more distance into the wind than a boat with lower hydro L/D, the apparent wind angle on deck will be the same for both boats.

 

maybe this...

 

Actually, beta is the analog of the glideslope.

The leeway angle is the angle of attack of the hull. Notice that it doesn't show up at all in the performance relationships. Neither does angle of attack show up in the glider performance sketched above. That 's why the total fluid dynamic forces are typically resolved into lift and drag. You don't have to know the boat's orientation to estimate the performance, and you can work out the leeway angle later.

The figure showing sideforce and resistance is not quite right. What's labeled side force is actually lift. Sideforce and resistance are the components perpendicular and parallel to the hull centerline. For small angles, they are nearly equal to lift and drag, but they are not the same.

You can resolve the lift and drag into sideforce and resistance (or thrust). The leeway angle, lambda, adjusts itself so the sideforce from hull and rig balance. The speed adjusts itself so the thrust and resistance balance.

From a design point of view, as opposed to the sailor's point of view, up until the board/keel stalls, increasing the leeway angle is good because it means the wetted area is decreasing. To the sailor, however, leeway angle is bad because it means the speed is low or the board is stalled.