Increasing Centerboard lift

Here's a question that may or may not even make sense, and if it does, it may not have an anser-- but here goes.

Let's say one has a very efficient daggerboard and a very small rig. What happens when the forces on the foil overmatch the forces on the sail?

Ya folla?

 

Thats easy - the forces on the foil win!

 

From all I've been learning recently(especially on this thread) is that the force above the water will always match the force on the centerboard.................I think !!!

 

The boat then follows a curved path. If the lift from the foil& hull is greater than the load from the sails, the boat's course will curve to windward. If the lift is less, the course will curve to leeward. This is fundamentally how helm steers the boat. The rudder changes the heading of the boat, which changes the angle of attack (leeway) of the board & hull, which then generates a net side force on the hull, and the boat's course is accelerated to the side.

Say you had a perfectly balanced boat that is hit by a sharp gust, changing the load on the rig while the boat continues to maintain its heading. The unbalanced side force will accelerate the boat to leeward. As the acceleration persists, the sideways velocity will grow. When this sideways velocity is added to the original velocity, the leeway angle grows and the lift on the board increases. This reduces the mismatch between the sail force and the board sideforce, and the acceleration will drop off. The boat will establish a new equilibrium in which the lift from the board matches the load from the sail rig. This process is known as sway damping, and it happens pretty quickly.

Ultimately, it's the helm that ensures the equilbrium is maintained, steering a straight course as the aerodynamic loads change.

For the purpose of analyzing the performance, you can include the longitudinal and centripetal acceleration of the hull as apparent drag and lift components (d'Alembert forces). The same relationships then hold between the apparent wind direction, beta, and the equivalent lift/drag ratio of the hull and the aerodynamic lift/drag ratio.

 

Bingo! Except when things are not steady state ... as Tom explained.

 

Actually, I'd put it the other way around. The force on the centerboard will match the force above the water. The sail trim has to come first - just think of the difference in board force between sailing upwind vs DDW, or between luffing the sail and trimming it in.

 

Terrific stuff.........

But now, when on a nice and steady close reach, everything is then nicely balanced. A sudden gust picks up, heels the boat over more, which I guess spills the air from the sails, messes up the equilibrium and the boat points quickly up to windward not leeward.

Am I getting this right? If in fact the sails lost their power, but the CB for that moment didn't, then did the power of the CB push me into a weatherhelm situation which I ultimately corrected by applying more compensating rudder?

 

Probably not ...

In the situation you describe, when a gust hits, the apparent wind moves aft. This tends to heel the boat more and the boat starts to accelerate. If you look down from above the boat you will see that the sail has moved to leeward when the boat heeled. At the same time the foil moved to weather. This offset in drive (sails) and drag (foil) is what causes most of the weather helm, when a gust hits the lever arm is greater and the drive increases. The boat almost has no choice but to turn up. The boat does this before it has a chance to accelerate much and before the wind gets spilled from the sails.

After the gust passes, you might be sailing faster than the normal breeze, if you don't reduce the lift from the foil by steering away from the wind, the lift from the foil due to the higher than normal speed will produce a turn to weather as Tom describes.

 

In reply to RHough,
Quite right. I cannot see the change in course. I can certainly see the hull change its heading. No excuses, just sloppy wording that should not have happened. At least we got a clear explanation of it all. My thanks.

As a further word about models. It is extraordinarily difficult to evaluate any device that might be fitted to a model for the want of steady wind. I travel 40miles to get usable conditions occasionally. When conditions are right it is a joy to sail.

 

Ivor,

I sailed a 1M RC boat years ago, I completely understand the frustration of sail in the breeze that close to the water surface! Frustrating is a very mild word to use.

The lift, course, heading, "pointing higher" discussions that sailors have are a sore spot for me. The more I come to understand what is going on, the more frustrating some of the statements I hear become. It is obvious that the force from the sails is never really steady, thus the lift the foils produce is also not constant. Any fixed amount of centreboard cant will only be 'right' for a limited number of conditions. An adjustable trim tab or variable camber foil is what is needed to be able to adjust heading to match the course and reduce hullform drag. In RC sailplanes I used full span flaps that were 30-40% of chord. Zero flap for maximum L/D, negative flap for high speed runs, and positive flap for minimum sink ... we change the camber and AoA of our sails, it makes sense to have that same adaptability on the water foil.

One of the reasons I am focused on this is due to my efforts to make sense of instrument readings. To hit a design target speed, you have to be able to read true wind speed and direction. Sailboat instruments have to calculate the true wind numbers based on boat speed through the water and apparent wind angle and speed. The leeway angle makes the apparent wind reading wrong by an equal amount. A leeway correction must be added to get the real apparent wind angle to the course sailed, if this is not done, the true wind calculated is also wrong. On top of this, the position of the masthead wind sensor is effected to some degree by the upwash from the sails. For a sensor mounted a bit higher and forward of the mast, the upwash tends to cancel the leeway error ... the trick is to find out how large these errors are and to have a correction table to get accurate information ... much like a deviation table for a compass.

If you are finally able to get the knotmeter to read the correct speed through the water, the wind instrument to read the correct speed (corrected to nominal 30 ft off the water) and angle relative to the boat then you can use polar target speeds and angles.

In still water (no current), the instrument system will then see the leeway as current set and drift, it should be the same on both tacks ... (good luck)

When the hull has a 5-6 deg different heading than the course through the water, the knotmeter is no longer aligned with the course ... so there is another error to compensate for. Depending on knotmeter transducer placement another error can be created by the local pressure gradient on the hull ... another variable with speed and leeway angle ...

So when I hear a sailor say, "The boat points higher." I have to wonder what they think they mean. Pointing higher could be "Sailing at a higher leeway angle (while sailing a course father off the wind)" or it could be "Sailing a course closer to the wind (while sailing slower and at lower VMG)" or it could be "Sailing a course closer to the wind (at the same speed and higher VMG)".

After the "My boat points higher." comment, ask about how the VMG changed ... 9 of 10 times you get a blank stare or an answer like "My GPS says the VMG is xxx higher." GPS "VMG" is really "Waypoint closure velocity" ... GPS "VMG" is never constant when sailing to a mark or waypoint. In fact, the GPS "VMG" goes to zero when the waypoint is on the beam. The boat has been making the same VMG (Velocity Parallel to True Wind) for the leg.

Sorry for the rant ... not aimed at you or anyone in particular ... just the result of years of frustration ...

 

You're combining a lot of boat responses, here. When the gust hits, the boat is no longer in equilibrium before it ever heels or changes heading. Most of what we've been discussing in this thread has had to do with one of the boat's six degrees of freedom of motion, the sideways translation or sway. Now you're starting to talk about the roll and yaw axes, too.

Weather helm has to do with balancing the moments about the yaw (vertical) axis. There are several contributions to these moments. The centerboard or keel is typically close to the center of gravity, but carries a large force, so it is significant despite the small lever arm. The sails also provide a large sideways force. The side force from the sails acting at the sails' combined center of effort and the side force from the hull acting at its center of lateral resistance can just about be combined into a couple, with the distance between them being the lead. Naturally, there's a moment from the rudder, and in the steady state, this is varied to balance out all the other moments.

One of the biggest ways heel comes into the yaw balance equation is through the lateral displacement of the sails' center of effort. When the boat heels, the center of effort hangs out over the lee side of the boat, and the forward drive from the sail rig generates a yawing moment to weather. The shape of the hull in the water changes, too, but I suspect that much of the weather helm often attributed to hull shape is actually due to the sail rig instead.

Just as the heel changes the angle of attack between the sail rig and the apparent wind, it also changes the angle of attack between the board/keel and the direction of travel. So there are lots of things all changing at the same time.

I would expect the biggest contribution to weather helm when the boat heels is due to the fact that the center of effort of the board/keel moves to windward while the center of effort of the sail rig moves to leeward. As long as they were lined up vertically, only their side forces acting through the small lever arm of the lead generated any yawing moment. When heeled, the large vertical lever arm between the two centers of effort that causes the boat to heel under the influence of the side forces, picks up a horizontal component that contributes to the yawing moment under the influence of the sail drive and board drag forces.

John Letcher has as good discussion of all of this in his book on self-steering.

 

Good points. IMHO, the lack of a leeway measurement is the biggest deficiency in typical sailboat instrumentation. I don't know why there aren't leeway sensors on the market, except perhaps because it is hard to accurately measure leeway.

There are two ways I can think of to measure leeway. One is with a vane. You see vanes like this on the sides of aircraft to measure the angle of attack. Some use a sensor that looks like a wind vane. Others use a narrow cone with slots in it. An electronic pressure sensor measures the difference in pressure between the two slots and rotates the cone until the two slots have the same pressure, at which point the centerline between the slots is facing directly into the apparent wind.

Another method is by trailing a string, possibly with a bead on the end, off the stern like a taffrail log (possibly combined with a taffrail log), and measuring the angle between the string and the rail.

The upwash effect on the wind vane is known as position error, because it's due to the position of the sensor in the flowfield. Position error also affects the speedo, because the flow accelerates as it passes over the bottom when there's leeway. In flight testing, the first thing we do is to have the instruments calibrated in the lab to determine their calibration curves, then flight test to determine the position errors. Position error for an aircraft is pretty simple, and can be represented as a pressure coefficient vs Mach number curve.

For a wind vane, it's going to be more complex. It will be a function of the apparent wind angle (rig angle of attack) and apparent wind speed, but also of the sail trim and heel angle. To really nail it down, you'd need an independent measurement of apparent wind angle and speed. A separate set of instruments on a pole at the bow or from a long boom sticking out ahead of the mast would have a smaller position error that would help. GPS velocities will give you part of the answer, provided that you know the current and true wind.

One of the things I hope to look into in the not-too-distant future is an extended Kalman filter that can estimate position error corrections from a body of sailing data. Getting good data is an art in itself. The attached file is an example of typical data I collected on a beat in that didn't seem all that gusty at the time. The bottom figure is a polar plot of the boat-speed/true-wind-speed ratio. The scatter is tremendous. If you look closely at the data, you'll see the polar plot is actually in streaks, oriented radially. I suspect these are due to the boat accelerating (and decelerating) in gusts, but I haven't looked into in enough detail, yet.

 

LOL ... another great subject that has been debated to death.

Two simple experiments to try:

The "hull shape" myth.
Beg borrow or steal a kayak. Paddle way until you have some speed, then list the boat to one side and note how fast and it what direction it turns. Paddle back up to speed and then dip a paddle off to one side and note the rate of turn and the direction.

The kayaks I know turn s l o w l y in the direction of the list or heel, the hull shape myth says that hull should turn away from the heel ... doesn't work for me.

The "changing rake (lead) to correct weather helm myth".
Lead is set during design and construction and effects the leeward yaw moment built in. This is built in lee helm, a tendency for the bow to fall off. It should be obvious that to sail upwind the keel/board must have a positive AoA, achieved by heading the bow closer to the wind than the course sailed; a weather yaw angle. To get the required AoA the built in leeward yaw moment of lead must be countered and overcome. Using the rudder to do this requires a leeward lift vector at the rudder to turn the bow into the wind. The leeward force on the rudder acts opposite the lift needed to sail upwind.

If the boat has built in weather yaw moment, the rudder has to act in the same direction to hold the bow down and both foils share the side load from the sails.

How is it possible to get a weather yaw moment from a boat with built in lead that creates a leeward yaw moment?

Move the sail effort to leeward (heel the boat) so the lateral displacement of the sail effort overcomes the lead and the boat has a weather yaw moment. Simple, no?

Now listen to a sailor tell you how mast rake controls weather helm. For sake of argument, say the sail CE (Centre of Effort) is halfway up the mast. Say 20 feet up a 40 foot stick. Now rake the mast 5 deg ... the CE moves about 1.75 feet.

Now consider that the mast base is 5 feet above the Centre of Buoyancy and the Centre of lateral area is 3 feet below the CB. Now the arm is 28 feet from the Sail CE to the Hull/Foil CE. Now heel the boat 5 deg, the sail CE moves almost 2.5 feet!

5 deg is a huge change in rake, 5 deg of heel is next to nothing. There is simply no way to make large changes in weather or lee helm by changing rake is the boat is allowed to heel as much or more than the rake angle.

The heel angle in the 'normal' 0-25 deg range of a ballasted mono is the controlling factor of weather yaw moment or "weather helm".

To much weather helm? Sail the boat flatter. Lee helm, heel the boat more.

If you can manage to rake the mast aft far enough to create weather helm in light air with a big Genoa up, you are going to be screwed when the breeze fill in and the boat heels to 20 deg with a blade jib up (moves the CE even father back).

Oh, Yes the Letcher book is a must have item. One of the few books I was ever tempted to steal from the library ... lucky me, I found a copy at a used book sale.

 

My software allows me to enter a correction table for position error of the wind transducer. How to get the values needed for the table is an opportunity for creative thinking ...

Long term leeway can be derived from COG and Heading. Current obscures the data. One thing I've played with a bit is running a timed square under power to get some Set and Drift values, then sailing to collect data, then repeating the timed square. The data under power gives me an idea of the current field in the test area, to compare to the Set and Drift data in the same area under sail ... the difference should be due to leeway.

IIRC, you are not trying to hit targets from a VPP, correct? All you need is repeatability not accuracy, right? Sail around a bit collecting data and use the best performance as your targets. After a time you will have the information you need to evaluate changes in sails and foils.

As for the spiky wind when it didn't seem all that gusty ... welcome to the world! All my wind data looks much the same as yours and like the examples in Frank Bethwaite's "High Performance Sailing", ain't no such thing as a steady breeze close to the water. Frank goes into it in mind numbing detail ...

On page 164 in my copy he gets into the leeway measurement question ... He had a design he knew should require 3 deg AoA on the foil or a 3 deg leeway angle ... he tried the string method and a direct observer from shore and could not measure the leeway; he ends with "In no way do I suggest that the centreboard does no meet the water at about the 3 degrees neccessary for the force to be developed. All that we report is a measurement difficulty in that the measured leeway is less than theoretically possible. Something is happening that we do not yet understand."

I think an intellegent software application should be able to take real time Set and Drift and the relative wind angle to deduce the leeway, then apply it to give the 4 wind numbers; Ground Wind (anchored that the RC sees), True Wind (what a boat drifting in current sees), Relative Wind (wind angle to heading), and Apparent Wind (wind angle to course with leeway correction).

The chartplotter on my boat produces 3 of the 4, the leeway corrected Apparent Wind is the missing number, also the one needed to make the True Wind correct.

Are we having fun yet?

 

Reply to RHough,
I was interested to see that we share a history of model sailing and RC gliders but I have never sailed a full sized boat.

I discussed the problems of model sailing in the preamble to section 2 on the Thames sailing barge on my website. I think that it is worth a read. Following that I built a model of a Norfolk wherry with sail control just like the full size just to see whether it could handle confused winds. Two weeks ago I sailed it 13 times round a long course on a lake surrounded by trees and did not make one tack. It also truly gets closer to the wind than a model barge as is evident by the fact that it can beat from end to end of the public lake at Maldon in the UK in two tacks where model barges need four.

You bring your flying experience to bear on the centre-board problem and see some possibility of using a variable profile like one might do on the wing of a glider. I do not know whether you have ever had a copy of Theory of wing sections by Abbott and Von Doenhoff in which the outcome of the NACA testing of aerofoils is given. The data is impeccable. There are pages of graphs of coefficient of lift versus angle of attack for all sorts of sections. If you riffle the pages it is immediately obvious that the slope of every graph is the same, 0.1 per degree. What changes with camber is the position of the plot relative to the axes, the maximum value of the lift coefficient, the shape of the graph at the stall and, most significantly, the shape of the graph of the coefficient of drag against angle of attack. As you must be well aware, when you were operating your flaps you were effectively switching from a symmetrical section with low drag to a heavily cambered section with higher drag, if you go outside the drag bucket that is, via a section with less camber and you were optimising for the flight modes that were required for circling in lift or getting back to the landing site.

Now you want to do the same thing for a sailing boat and its fin. I do not know for certain what speed range is involved with model gliders but I would have thought that I would be 2 to 1 and I am sure that it is higher than this for full sized gliders that are capable of 130 knots. Yachts, when racing, do not have such changes in speed through the water and I think that their needs can be met with a symmetrical section. As I read the NACA data it looks to me as though the ordinary symmetrical section NACA 0012-64 is just about the best compromise available.

I am aware that this whole affair is so complicated that other solutions may come to mind but there must be some pay-off between complexity and reliability somewhere and you may see some more attractive answer. It may be that you see a very gentle stall as a first requirement and that is possible with a cambered section.

I said that I could foresee canting and twisting fins and my reason for that is that we are now looking at well over 400 miles in 24 hours in a mono-hull. Such boats must plane all the time and I can only think that they can plane so well because of the canting keel to keep the hull upright. The next obvious move is the get the hull in line with the course by turning the fin. It is not especially daunting mechanically but it might be if you want to change camber as well.

Treat this text as a sort of dialogue. I jotted it down quickly and did not take time to reflect on it just to keep the pot boiling. If it is in error somehow let me know and we shall all learn.

I will think about the rest of the note more carefully as I have no direct experience to get it into context.