Increasing Centerboard lift

Doesn't everbody?

For an unballasted board, yes. You only have to deal with the gap at the hull that gives the range of adjustment you need.

For a fixed keel with ballast or ballast bulb, I think the structural requirements to keep keel and boat together make a trim tab or flap the best option.

All that is needed is to deflect the flap enough to effect a 0-3 deg trim change. It does not take much camber to create Cl = .3 (although and Tom will soon point out the .1 Cl per degree of AoA is a 2D solution and does not take (dare I say it?) Aspect Ratio into account. You don't get Cl = .3 at 3deg AoA on a real wing or foil. The 3D lift line is not that steep. In effect what a cambered foil does is to move the zero lift angle off the centreline of the vessel.

Sorry, you missed a move or three. Instead of moving the ballast to windward and having one or more extra foils to produce lift... why not move a buoyant bulb to leeward to produce the righting moment? Once the righting force is to leeward, why not get the supporting strut out of the water to reduce drag? The righting bulb would then be on the surface. Why not make the bulb big enough to provide an extra berth or two? ... bingo a catamaran! No more lead to haul, more righting moment and less weight ... if we build a second pod on the other side of the boat we get a trimaran ... storage in the floats, no lead to haul and back to a single foil for lift.

And BTW, those canting keel boats have to run engines to move the bulb, they are not sailboats, they are motor boats.

Now that we have a sensible trimaran platform, we no longer have to worry about the effect of heel on lift, there will be very little. Since the boat is much faster, the AoA needed to produce the needed lift is very low and a laminar flow foil can be kept within its drag bucket. The trim tab is then used to balance the helm rather than alter the AoA very much.

I'm quite enjoying it Ivor,

Randy

 

Reply to RHough

Randy,
I see that we, and Tom, have attracted the attention of Brian Eiland. I think that we all march to the sounds of our own drums and that if from time to time there is discordance so be it.

Yesterday you suggested that everyone has Abbott and von Doenhoff. I do not know anyone that I can talk to face-to-face who has any interest at all in the physics and mechanics of sailing although I know a fair few people who sail including my son and grandson. It is interesting that the few who contribute to this forum are spread all over the world.

So, back to this thread. I agree with your case for the multi-hull but the sheer cussedness of people leads them to chase round the world in classes and the mono-hull is one of the classes. I do not take any special interest in the world of sailing and buy magazines and so on to keep up with what is being done but I do like thinking about the evolution of these boats. The boating world have taken a long, long time to get to grips with aerofoils under their boats, mostly sailors just talk about depth and their archaic ideas of centres of lateral resistance which is no more than the centroid of the underwater profile and of little value. Now that hydrofoils are being used we can all think about the possibilities.

If you do then, if aligning the hull with the course is the goal, it seems to me that you have to think quite separately about what you might like and what you can have. I started by suggesting a continuous control of cant and turn of the keel. You object because an engine is needed to supply such a system yet it is used on round the world sailing so presumably power is derived from the wind both to drive the boat and to trim its keel. Whatever, I shall take it that control is possible just to develop an argument.

If the whole fin can be rotated about its vertical axis then there is no doubt that it can lead to the realignment of the hull relative to the course. Once the fin is moved, in order to get things back to sailing in proper balance, the rig will have to be re-trimmed and the rudder moved to get the keel back to the required angle of attack. None of which is impossible. The question that it raises is; is it mechanically possible? Then we move to what we can have rather than what we would like.

I do not think that I would want to cross an ocean in a boat with this system but I may be a faint heart. If canting keels are sufficiently reliable to take round the world it may be possible to use the existing mechanism on a new swivelling mount to give two axis movement. I note that fairing the hull round this moving keel has caused problem in the past just to make the fairings robust enough to withstand the water flow at speed.

If mounting the fin on a shaft to let it swivel is regarded as an option then the design becomes a nightmare because the diameter of the shaft is limited by the profile of the fin and we run into the fact of life that materials that can operate at very high stresses are unable to take a permanent set and snap without warning.

All in all there is a very strong incentive to look at alternatives. Your moveable control surface is the obvious candidate. If we are trying to align the hull with the course then I think that we must accept than the whole of the transverse force needed to balance the transverse force from the rig is now produced by the fin. (Centres of lateral resistance become irrelevant.). This means that we need to know the angle of leeway. You may know this off hand but when I question sailors of many years experience they never seem to know or even be able to offer a figure. I think that 7° might turn out to be reasonable for a modern yacht but not for a cruising yacht. Perhaps you have a better figure. Are you sure that the moveable surface will shift the lift curve far enough to off set this leeway? I have grave doubts after looking at A and vD.

I must wind up but the dialogue is interesting and I am forced to look at things that might not come my way normally.

 

Foil Induced Drag

RHough:

Where did you ever get these numbers? The percentages depend on the type of boat, but I've never heard of a boat with such high foil induced drag! Are you sure you don't have the numbers reversed?

Doug Halsey

 

I think there can be very large benefit to changing the AOA on the centerboard, but you have to optimize the design to the new configuration.

You have to have an foil section that is optimized for a higher lift loading (lift over area), which means it can be much smaller that a fixed one, and you have to design the whole centerboard to operate at best L/D. With a smaller more effective centerboard you also get less healing moment since the center of lift will not be as deep on the centerboard. Consider that if with a fixed centerboard the effective AOA is only 2 deg, but with it at 6 deg AOA, it would only need to be only one third the size. You will have less wetted area and less induced drag of the centerboard.

The other benefit is that having the hull center line moving closer to the direction of travel can reduce the hull drag considerably depending on the design of the hull. On a multi-hull, the long skinny hulls when at an AOA to the direction of travel will generate a small amount of lateral load, assisting the centerboard, but at a large cost in terms of drag. Those very low aspect ratio hulls do not generate lift very well as compared with the high aspect ratio center boads. By putting an AOA on the centerboards, the induced and form drag on the hulls will go way down I would think.

It is a great idea, but you need to know how to design centerboard to take advantage of the concept. You also have to have a way of sealing the larger centerboard slot (a sliding plate perhaps?) since this would also increase the drag as well.

 

LOL ... I used the same measurements and formulas to get those numbers as I've used when designing RC aircraft. I've used wetted surface and lateral (projected) area and corrected the 3D lift line for the AR of the keel or board. To estimate the numbers. I used a hull wave drag estimate from one of the software packages to check the drag curves.

It has been awhile since I've looked at the data, but it seemed logical at the time.

I'm pretty sure that the drag force curve I came up for the hull and foils in the zero lift condition was similar to those produced for the same shape in Freeship or what ever I was using at the time. I then added the lateral loading needed to balance the sail plan and added the induced drag to get the totals.

I could be *very* wrong. Since that time I've collected more data. I should probably plug the numbers from a boat I have a good VPP for into my model and see where the errors are. I doubt that I'm off by an order of magnitude or two. The drive requirements for a slender hull are very low, the induced drag of stubby little foils is very high. I had to keep the drive/heel forces within the RM limits of the design I was working on and the resulting polars made intuitive sense.

I'm no professional, but the RC sailplane designs I've built have mostly performed as expected. I have no reason to doubt that the sailboat will too. However, I freely admit that it is mostly an anti-Alzheimer's exercise for me, better than crossword puzzles and sudoku.

Randy

 

Lift on the board = lateral force from the sails.

There is a requirement for high lift at low speed from a symetrical section when turning and when accerating from low speed after a tack (lateral force from sails is the same as at full speed, but the boat is perhaps at 70% of full speed).

Altering the AoA of the board might align the hull with the course, but it also increased projected area of the hull/rig to the wind, increasing drag. The waves are more or less in line with the true wind, aligning the hull with the course also increases the projected area of the hull to wave motion. The leeway angle of the hull with a fixed board, reduces both of these. On would have to know the magnitude of each force, plot the curves and find the angle that produces the lowest total drag.

If the conclsion is that a different AoA on the board compared to the hull centrline will reduce total drag, it is my opinion that a variable camber foil is a better solution than a canting foil. There are no major structural changes needed to the trunk and no large gap to seal. A faired hinge line on the trim tab is easy and known technology.

 

Actually, I was trying to sail consistently!

Where I want to go eventually is to build a VPP from first principles, and then calibrate/validate it with data. Then I'll have a baseline from which I can evaluate potential changes to the boat. It's definitely a long-term project.

I agree. I plan to develop a Kalman filter algorithm to process the data. The Kalman filter allows you to combine everything you know about the problem - how you expect things to work, the fact that things don't work the way you expect, real-world measurements, and the fact that the measurements are noisy and uncertain.

To advance from one time to the next, the Kalman filter has five steps. First you predict the measurements you expect to see, based on your dynamic math model of the motion. In a navigation filter, this could be as simple as dead reckoning - you were traveling at a certain speed and direction, and in the time since the last nav fix, your position should be the speed times the time interval plus the previous position. For sailing, the math model could be the whole VPP itself.

Next you predict the accuracy of your estimated measurements. The longer you go, the more uncertain the estimates become. In the third step, you use the predicted accuracy along with the accuracy of your measurements to calculate a weighting factor, called the Kalman gain, that will be used to combine the predicted values with the measured values to get an improved estimate. Finally, the accuracy of the estimate is updated, based on the reduced uncertainty due to the measurements. Then you're ready to do it all over again for the next time step.

In principle, the estimates become more accurate than either the model or the measurements because each one provides some information that the other one lacks. It's also possible to estimate errors, like position error coefficients or sensor misalignments, if the effects of those things are included in the model. And you can estimate things that aren't measured, like leeway angle or current.

You're right that if the current isn't changing, you can use maneuvers to distinguish between leeway and current. For example, when tacking, the leeway will change sign, but the current still comes from the same direction. Still, I'd like to be able to measure the leeway!

 

Well.

That'll teach me to ask stupid questions.

 

Centerboard's AoA

Hi Everybody,
I just discover this thread today, and would like to share my experience with this issue:
In 1996 at the A-Cat World in Spain, I put epoxy filler in my Bim centerboards' case, in order to move the centerboard leadind edge by 1.5 or 2 mm "inside" the boat.
At this time centerboard had 250 mm chord and 660mm depth, and hulls were a bit V shape along the front part of the "keel line".

The purpose of this experience was to reduce overall induce drag.
As already mentionned in this thread, the angle of the hull will decrease and therefore its "induced drag".
In fact it is like if you swap a part of hull's lifting and induced drag to centerboard's lifting and induced drag. Because centerboard is supposed to provide a better L/D ratio than the hull.
Moreover, when you sail with the same mast rake you feel you have to maintain a little force on the rudders to give them a slight positive AoA, and again the purpose of transfering some of the lifting task from centerboard to rudder enables to decrease overall "Induced drag"
because induced drag is a square function of lift.

Anyway, on the water windward with strong wind and waves it was very striking how it was effective compare to similar boats. Obviously I had a little competitive advantage.

Today A Cat's hull shape have so flat underwater sections, that I doubt you can improve hull's induced drag significantly with this solution.
But sharing the lifting task between centerboard and rudder should provide the same advantage on any cat or I guess so.

Sorry for my poor English and for this not very academic presentation, but I hope it can provide little insights.

Regards

Erwan

 

Foil Induced Drag

Hi Randy,

I'm hoping to go sailing today, so I'll have to keep this very brief.

The induced drag of the foils should be something like 1/2 (very roughly) of the total drag of the foils alone, and a much smaller percentage of the drag of the hull + foils.

If you have Marchaj's old book (Sailing Theory & Practice), there are curves of measured drag vs. speed (with & without sideforce) for several boats (including full-scale measurements for an International 10**2 Meter Canoe) that you can use to convince yourself of this.

I don't mean to be critical of you or the software you've used, but sometimes it's good to look at some of the old available data ,to get a reality check.

Regards,
Doug

 

John Shuttleworth has a good breakdown of the drag components of a performance-cruising catamaran in his Dogstar50 article. The breakdown is:
WIND
Sail lift - lbs 3610
Total air drag 1013
Sail and rig drag 361
Air drag of hull 652
WATER
Keel lift 3610
Total drag 687
Keel only drag 176
Hull drag (from tank testing) 511

His keel (actually daggerboard) drag includes both the induced drag and the board's profile drag. Note the amazing coincidence between the sail lift and the keel lift. And the small proportion of the keel drag to the total drag. [Now, he paid exceptional attention to reducing the aerodynamic drag of the hull in this design. Nevertheless, if you apply the philosophy that, "If you're hunting elephants, you have to go where the elephants are," which of these elements would you attack if you wanted to improve the L/Ds of the boat?]

 

That is only the max L/D condition for the foil. It is far from the theoretical max L/D for the hull and foil combination.

To reach L/D max for the combination, either the parasite drag of the foil/hull has to be reduced to equal the total induced drag of the foils (fin & rudder), or the Lift must be increased until total induced drag equals the total parasite drag. L/D max for the combination is always at a higher Cl than for the foil alone.

I didn't think you were being critical. As it turns out I had gone to the wrong part of my model ... an old man's brain fart.

I was trying to estimate the effect of leeway angle. From the data that Tom posted:

Keel L/D is 3610/176 or 20:1
If we assume the Keel is at L/D max, then induced drag is 1/2 of 176 or 88
88 is 12.8% of the total drag. About 3 times my estimate based on looking at the wrong section of my notes.

When I go back to my model and look at how I calculated I see that I did indeed add my induced drag from leeway angle to the hullform drag curve and at a S/L ratio of 1.05, the model reaches the point where Di is 12.79% of the total. Shuttleworth's numbers are for a a cat sailing at a S/L of 1.76 ... well beyond what I was trying to model ... displacement monos.

At S/L 1.34 ("hull speed") my model predicts the Di to be 2.15 % of the total.

This was all part of a design study for a Schooner ...

I got to a point that it no longer looked like I wanted to build it so I stopped working on the mathematical model.

Still, the indications from my research and from the numbers that Shuttleworth has apparently verified, the fact remains that there is only a small gain to be had by changing the heading/course relationship with a canted board.

As Tom hints, the aero parasite drag is the first place to go looking. Altering the heading to match the course increases the projected area of the hull(s) to the relative wind and will probably increase what is already the greatest source of drag.

Hope you had a great sail!

 

Max L/D

Hi Randy,

It looks like we're on the same wavelength about this now, but I would like to add a couple of things.

As you pointed out, in the max L/D condition for an isolated wing (or centerboard) the induced drag is half of the total drag. This is a classical result, probably taught in lots of Aerodynamics courses. However, the optimum depends on exactly what problem you're trying to solve & on the constraints of the problem. In this example, you're given the lift & the shape of the wing; you want to determine the size of the wing that will minimize the drag. You get this result only if you vary the wing by scaling equally in both directions (span & chord). In other words, aspect raio is constant.

In a sailboat, the centerboard's sideforce (or the daggerboard's or the keel's) is far from constant, since it has to balance whatever sideforce the sails are producing. So the designer has to make some compromises in determining the size. While sailing, however, the skipper gets to do some design work, by deciding how much to raise or lower the board (Keelboat sailors can ignore this.) In this case, the aspect ratio is not constant.

If we assume a rectangular daggerboard, when it is raised or lowered the span is varying & the chord is constant. Aspect ratio is not constant. For a given set of conditions (ie- boatspeed & sideforce), an optimum span can be found & it turns out that the induced drag is 1/3 the total drag of the board (not 1/2). Like the first problem, this assumes that the profile drag coefficient is constant, but in addition it assumes that the induced drag efficiency factor is also constant.

I've never seen this result in print, so I'm inviting the technically-inclined to try to prove or disprove it for me.

I hope nobody thinks I'm being a nitpicker to mention this. The subject is interesting to me because it's closely related to the problem of how high to fly my foiler Broomstick. In any surface-piercing foiler, the span of the foils decreases as the boat goes faster & the boat flies higher, but it doesn't necessarily vary in an optimum manner. Fly too high & the induced drag gets too big. Fly too low & the profile drag gets too big. On Broomstick, I can vary this by adjusting the aft-foil incidence.

One other point about jibing boards: Bethwaite (in his book) cautions against using them because they make the wake from the board impinge on the rudder. I'm wondering : has anyone with a jibing board ever noticed problems with rudder stall or ventilation, that went away when the board was prevented from jibing?

Regards,
Doug

 

I am now wondering if I induce 3-4 degree of "jibe" from stock (ie no AoA) CB will that have more negative influence my dual rudders (NACA ???? design) which presently are situated 12 feet behind CB and spaced 28 " apart. Or, if I decide to push them an additional 18 ins. rearwards onto a yet to be built jack extension with a new spacing of 18 ins..

 

Thanks, I agree most people are saying about the same thing but from different points of view.

Is your foiler a single track boat like a Moth? If so I suspect that you have found that the best performance is when heeled to weather. On boats with this configuration, when the daggerboard T foil is angled in relation to the surface, a part of it's lift is acting in the same direction as the dagger board, acting to reduce the lateral force requirement. This should reduce ventilation of the daggerboard. Another side effect of the windward heel is an increase in span.

I think you are equating the immersed span of the daggerboard while foiling to raising the board in a displacement hull. I'm not sure if the two are alike.

The total lift requirement does not change (much) at different foiling heights, the lateral lift should actually go down with increased foiling height (the heeling arm gets longer). For constant foiling height, windward heel reduces the load on the daggerboard and increases the load on the T foil. The added span in this attitude should reduce induced drag.

Foil wake and rudder interaction is a whole other topic ... Rudders always operate in downwash from the foil. Wake implies turbulence to me, if a canted board causes turbulence under the hull ... it would be a pretty poor design.

I did a little vector drawing to try to show what I'm talking about.