canting keels and all that

[dionysis]

hi all,

I am really impressed by the potential of canting keels - all that righting moment, means you can design rather more slender hulls, less resistance, both wave and friction, hence lighter and more efficient.

My question is, at what point and for what reasons does the keel become inefficient as a leeway preventer?

At maximum cant, say 45 degrees, is it just a matter of loss of projected area, and an adverse vertical component, or is there some kind of loss in terms of lift and drag of the foill itself as well?

cheers guys, dionysis.

[Tim B]

Dionysis,

This is a tricky question. The horizontal lift vector (ie. normal to the keel when the boat is stationary with keel at 0 degrees) will change with the Cosine of the angle (measured from the vertical), but it also introduces a cross-flow. This is similar to flying a wing in air sideslipped. ie. The flow is no longer normal to the leading edge. This will cause a change in Cl and Cd of the keel.

However, I would think that the benefit of reducing the hull wetted area (and possibly wave drag) would far outweigh in increase in drag, and thus, even if you had to point slightly lower, you would still get better VMG than before. If you can, build a model and stick bits of wool over it, then blow air past it, at different angles of heel and keel cant. Watgh the pieces of wool, and this will give an idea of the flow in the boundary layer. A small handle with wire and wool tufts would show stream-lines around the keel and hull.

Good Luck,

Tim B.

[dionysis]

Tim,

I thought as much. The loss of 40% horizontal side force, and an adverse vertical lift component adding to heel, together with loss of efficiency due to slip, and its induced drag penalty is a high price to pay. No wonder the open 60 yachts have separate centerboards to counter leeway and minimise any contribution from the canting keel.

But then again, they have proportionately larger wetted surface than the kind of vessel I had in mind. After all, maximum righing moment is only needed beating into strong winds.

I agree VMG may be increased for a very slim hull, and your idea of modelling the flow is well worth looking into.

Where is a sensitive VPP when you need one!!

Thanks and cheers, dionysis.

[SailDesign]

Tim,
I'm kinda curious as to where your crossflow comes from when the keel is canted. I've looked, but can't find any ;-)
The effectiveness of the keel certainly drops off dramatically with cant and heel, but due to regular stuff lke angle, not crossflow.
The Black Sea 40 used a canting fin (not designed for lift, just low drag) with a pair of asymm daggerboards for leeway. It _feels_ as though she crabs to windward, but I think that's just wishful thinking.
Steve

[gonzo]

The canting keels I've seen are either a regular IOR style or struts with a bulb. How about a wing instead of a bulb? It should produce more lift than a normal foil keel. I think it would eliminate some of the problems with fixed wing keels. For example optimal angle of the wings to the keel.

[Tim B]

I've just reasoned through this question, and I was right about the leeway. Note that righting moment is just F*D, leeward force is F*D*Cos(theta). As the boat heels, the bow will tend to sink faster than the stern, and thus the boat will be put at a pitch. The pitch and heel, combine to put the keel at an angle of attack, but also an effective angle of sideslip, causing spanwise (correct term for crossflow) flow towards the tip. What is complicated is that as the keel is canted to windward, the sideslip angle changes, as does the maximum pressure coefficient on the keel.

Gonzo,
May I suggest that we add the wings on later, when we actuallly know what's going on on the keel itself. I like the idea though, and we will address it.

I think this question is going to take most of our fluid dynamics and hydrostatics skills!! As well as Cad and VPP, possibly even with some models examined. Anyone got a 1metre class model yacht I can borrow?

Cheers,

Tim B.

[Tim B]

Check this link out, might be some good stuff:
ftp://ftp.fh-kiel.de/maschinenwesen/schiffbau/pub/Graf/Publications/

Cheers,

Tim B.

[SailDesign]

Tim,
I still have to take issue on the spanwise flow. Why "will" the bow sink faster than the stern? If you are designing a boat to have a canting keel, then this is the kind of thing you "design out" of it. Since it doesn't have to fit a restrictive rating rule that favours fat sterns, you can make the boat heel "flat".

Gonzo - Let me think about the wings. I'm not saying they shouldn't work, but they _feel_ wrong ;-)
Mind you, some things feel wrong until you realise that it's because they do work so well.

Steve

[SailDesign]

Tim,
While hitting the send button on the above, I realised where you are coming from. The extra leeway from having an inefficient keel in "leeway" mode _would_ create a spanwise flow.
Steve - begging the question of why you would _ever_ want a boat that made that much leeway.........

[dionysis]

Tom Speer in an email suggests there is no added loss due to spanwise flow or 'slip'. He also suggested incorporating 45 degree anhedral wings to the bottom of the keel to gain some more side force, vis:


anhedral endplate

What made me start thinking of keeping the canting keel as the hydrodynamic lift device, was the idea of using a multiple foil keel: a la the condor keel; very efficient upwind.

Check out Philip Carter's site:

Innovations in IACC Keel Design (http://www.esotec.co.nz/condorkeel/)


Cheers, dionysis.

[dionysis]

:confused:

anhedral.gif

[dionysis]

:cool:

[Dim]

Hi All,

I beforehand am sorry for the fact that has come in discussion of a problem. Your discussion very much has interested me with certain the parties. On my sight, the problems of a flow around a hull - are difficult enough in solution. The problem of a flow of any superposition of bodies - is even more difficult. To some extent of accuracy these problems can be decided with the help of the complex and expensive software. But purchase and operation of such software can to itself allow not all. The exactest answers can be given only by a model experiment. But such luxury can afford too not all.
Question. Whether there are any methods of solution of similar problems, which can give acceptable accuracy at the acceptable price?
I understand, that there are many empirical datas. But what to do if you on any design stage understand, that they cannot be applied to your construction?

Thanks, Dim.

[dionysis]

Hi Dim,

I agree with you it is a problem. Even if you had the money for the expensive software, you have to learn to use the software confidently and this can take a long time.

One way out is to go back to what Tim B says up the top, about making smaller models and looking at the flow from a qualitative point of view using wool tufts etc.

Sometimes you got no choice.

Cheers, dionysis

[Dim]

Thanks for your answer Dionysis.

I with you completely agree. I was decided so many to write only because the problems of a flow's calculation are very interesting to me. I wanted know your judgement on existence simple (in practical calculations), but reliable (from the view point of outcome's quality) of methods. I think, that them do not exist for today. Sorry, for bad translation.

Thanks, Dim.

[Tim B]

Dionysis,

In reply to Tom Speer's e-mail, I was merely pointing out that span-wise flow existed. I obviously cannot suggest whether it is a problem or not.

Anyway, anyone up for a model test, we'll need a 1m model and a towing tank (or wind tunnel).

Cheers,

Tim B.

[Dim]

Tim,

I have two question:
Question first. Why the model should have length in 1 meter? The size of model, basically, depends on the size of original hull.
The scale factor the is less, than more model. And, therefore, we shall receive exacter datas of experiment.
Question second. What quotations on running of model in towing tank? How much it is necessary to make of runnings of model?
It already third question.

Good luck, Dim.

[Dim]

Dionisys,

The solution of a problem nevertheless is. I once again have addressed to firm "Tesis" with the request to announce their remarkable program.
For now I have gone on their site and has downloaded for all of you the documentation in English. But it is one large file. I can not post it.

Good luck, Dim.

[Dim]

I have recollected as this file it is possible to place on a forum.
The truth two diskettes on 1.4 mB will be necessary for you.
Write each part on a separate diskette and make dearching from diskettes.
At first insert into the drive a diskette with the first part. Then WinZIP itself will request you to insert the second diskette.
We practically do not use such old method of archiving. We use other program, which divides the large file into a parts.
But the format of this program is not supported on a forum.

First part in this zip-file.

See further.

Dim.

[Dim]

Second part in this zip-file.

See further.

Dim.

[dionysis]

Hi Dim

my unzipper software cannot open the files. I will wait till the TESIS site is back up again and download from there.

CFD is interesting but it looks really complicated to me. Good luck if you can get the software to work for you.

I say go to a towing tank facility if you have one near you and check it out.

cheers, dionysis.

[Dim]

Dionisys,

I have not the price-list from "Tesis" and consequently I do not want to name the price of their program. I want to wait of their probable answer on this forum. But previously I can say, that the price of the "Flowvision" program for the order is less than any "western" program ( for example Flow3D ).
In the other case I would not begin to speak, that all of us have basic solution on a flow's calculation.
Sorry for your zip-problem.

Dim.

[Tim B]

There is a one-meter class of model yacht, which sounds like it is a similar hull shape, and there are 1ms all over the place.

The last time I looked at any CFD, it took me three (solid) days work to learn the basics, and a further weeks (rather intermittent) work to get any good results out of it. I just felt a model test might be cheaper and quicker than a CFD test.

Cheers,

Tim B.

[tspeer]

I think dionysis may have misinterpreted what I wrote when I said the lift was the same for the canted and conventional keel.

Simple sweep theory basically says the spanwise velocity component doesn't count. So what you have to do is resolve the velocity into components that are normal to the plane of the fin, spanwise, and chordwise (normal to the other two). The angle of attack is the arctangent of the normal component divided by the chordwise component. The effective dynamic pressure is density/2 times the sum of the chordwise component squared and the normal component squared.

Sweeping a keel on the centerline doesn't change the normal component, but it does change the chordwise component. So the effective speed drops and the angle of attack actually goes up some.

Canting a keel shifts some of the normal component into the spanwise component, which both lowers the effective speed and the angle of attack. At 90 degrees of cant, the angle of attack would be zero and the effective sweep angle would be the same as the leeway angle. This means the amount of lift on the keel from a given leeway angle is less.

However, the total sideforce on the canted keel is the same as for the conventional keel. This is because the sideforce from the keel has to match the sideforce applied by the sail rig. As a result, the canted keel boat will have a greater leeway angle, as measured between the bow and the velocity vector, than the conventional keel. Whether or not this is bad for performance depends on the drag. If the drag is higher, then the canted keel skipper will be able to brag at the bar about the small angle he tacks through, but it'll be the conventional keel boat that brings home the pickle dishes.

So the two really important questions to ask about a keel design are, "What is the righting moment?" and "What is the drag?"

[Dim]

I have removed this parts of archive. As I have twice tried to restore the document from these zip-files and at me it has turned out nothing. I bring the apologies to those people, which loaded these zip-files.
For 9 years, that I work with a computer, I probably only two times performed such operation. In what the reason of a mistake I do not know.

Dim.

[Tohbi]

so, in layman's terms, is a winged keel advantageous for the average sailor? i'm asking this because i'm modifying a 21' model of a 12-meter design and i'm debating keeping the wing.

the problem with the existing winged keel is twofold: draft and vulnerability to grounding. the boat is, simply, too deep for my kind of sailing and, since i'm going to a shorter mast, i think i can reduce the depth of the ballast and shorten the keel.

secondly, the wing on the bottom of the keel is vulerable to damage from a grounding that would allow the weight of the boat to list to one side or the other, creating a lever arm that would be translated to the wing/keel junction itself and to the hull/keel connection.

the advantage to the wing, as i see it, is in allowing a shorter keel but retaining pointing ability. so the question is about retaining the wing and, if so, in what configuration to offer the most protection.

[tspeer]

Originally posted by Tohbi
...the problem with the existing winged keel is twofold: draft and vulnerability to grounding. the boat is, simply, too deep for my kind of sailing and, since i'm going to a shorter mast, i think i can reduce the depth of the ballast and shorten the keel.

Shortening the keel would be viable from the standpoint of stability. However, shortening the keel or eliminating the wings will make the keel less efficient and affect pointing ability.

You might want to consider a combination of fixed keel and centerboard to provide shoal draft without as much of a degradation in performance.

secondly, the wing on the bottom of the keel is vulerable to damage from a grounding that would allow the weight of the boat to list to one side or the other, creating a lever arm that would be translated to the wing/keel junction itself and to the hull/keel connection.

I can see this might be a problem at the wing/keel junction, but not at the keel/hull junction. If the wing can take it, the keel/hull junction is already designed to take the moment of the whole weight of the keel at 90 degrees, so there should be no problem there. And since this is a 12 metre model, I suspect the wings are designed for the boat to sit on the wing tips and front of the keel.

the advantage to the wing, as i see it, is in allowing a shorter keel but retaining pointing ability. so the question is about retaining the wing and, if so, in what configuration to offer the most protection.

In effect, the wings add hyrodynamic depth without as much physical depth. If you shorten the keel physically and delete the wings, you're making a huge reduction in the effective span of the keel.

You may be starting down a road that leads to a completely different boat that's better suited for the sailing you do.

[Tohbi]

thanks tom? so which is the best remedy? shorten the keel and keep the wing? shorten the keel, keep the wing and add a centerboard? or shorten the keel, eliminate the wing and add a centerboard?

i would prefer to avoid the centerboard because of the complexity and intrusion into an already small interior. but, as you mention, the boat must be able to point well.

older designs used bigger keels with keel hung rudders. i wonder how effective a wing is as compared to a bigger keel? since we want the least amount of wetted area, this is a consideration. is the wing worth retaining?

[tspeer]

To really answer the question, you'd need a velocity prediction program (VPP). After all, that's how the 12 metre's are designed.

Without knowing anything about how your keel is shaped or how much you propose to shorten it, my guess is you're going to chop the keel off above the wings. So the wings will be gone and you'd have to recast a new keel to put them back.

Since the 12 metre has a reasonably long keel to begin with, I'd be inclined toward housing the centerboard entirely within the keel by cutting a slot in it. That way it doesn't intrude too much into the rest of the boat. But with the shorter ballast arm and lower center of lateral resistance afforded by the centerboard, you really need a hull with more form stability. A metre boat is a carefully balanced integration of a lot of factors that combine to provide good performance. When you start making significant changes to one area, that balanced design starts to unravel and performance will suffer.

This all sounds like you're making pretty drastic changes to a boat that's supposed to be a scaled down 12 metre. Even if you're successful, it's not going to be like a 12 metre anymore. Whatever resale value it has will be shot. I really think the most successful course is to sell the boat as it is and get a used boat that is designed from the outset for shallower water. You might even be able to find someone that would be willng to swap boats without any money exchanging hands. That way you save the time and expense of the modifications and get a boat that will make you happier in the end.

[Tohbi]

ha! good advice, tom, but, of course i'm not taking it. redesigning a boat that has been developed at a million dollars in engineering is too exciting to pass up. surely, we can make it better.

seriously, this is a tank test model with only a single skin of c-flex over the frames. lotsa bondo must be sanded off before new glass can be laid and the keel is kinda stubby and probably made of foam with a glass coating. at least, the wing is foam covered with bondo and paint. truthfully, the boat isn't worth much as it is.

so the boat must be rebuilt, anyway. the question is about using the design as it is, using an older keel design, or inventing some combination. i was told the design is from 'stars & stripes.' i'm thinking it is the '87 version, although, i haven't seen a profile of the '83 or '86. i have so little knowledge of winged keels that i'm not sure of their value.

so, we're experimenting.

[tspeer]

You weren't kidding when you said it was a model of a 12 metre!

I guess you'll just have to take the lines off her, work out the hydrostatics, and see how the stability works out for the rig you plan to use.

A retracting bulb keel might be a good option. Perhaps it could come up just behind the mast so there's not too much interruption to the interior volume.

[Tohbi]

i've been thinking of a daggerboard type keel but this boat is only 4' wide and any kind of intrusion into the cabin will be very invasive. moreover, i've been wanting to avoid a compression post altogether and just beef up the scantlings across the frames to support the mast.

what about shortening the existing keel [it is several inches thick beamwise] and welding a facsimile of the existing wing to a shank that will fit inside the keel in a kind of "T" shape? the keel could, then, be filled with weighted material: epoxy-saturated lead or fortified concrete, for instance. hopefully, the shank encapsuled within this mixture will be strong.

you mentioned that the boat is probably designed to sit on the wing so, maybe, it can be made strong enough to take the weight through this process.

i'm thinking that some surface area that is lost through shortening can be regained by adding a bit of blade behind the shortened keel. do these mods sound unradical enough to work without compromising handling too much?

[Tohbi]

some of the keel dimensions may be affected by the change in sailplan. as you know, ac yachts have very tall masts but i'm shortening this to fit the trailersailer application.

i've designed a standing mast pulpit that will allow the mast to be hinged and, simply, folded back. this reduces the mast to 15', necessitating a gaff mains'l. with a furling staysail, the total sail area should be around 170' with approx -2% lead.

the '87 ac boat that is the progeny [i think] of this hull is aprox 66' long with a 45' lwl. so this 21' model should have corresponding dimensions.

so i'm changing the rig, as well as the keel, and trying to get them to balance.

[tspeer]

Why not go with a gunter rig? Short mast, and taller rig when sailing.

[Tohbi]

tom, i've never seen a gunter rig except on paper, much less sailed one. they look kind of fragile and inefficient, but i know looks can be deceiving.

what's the advantage over the gaff rig?

[tspeer]

I've not sailed one, either, but mechanically it's like a gaff rig with the peak halyard all the way in. Aerodynamically, it's like a jib-headed rig, so the center of effort's in the same place and it's taller for better efficiency. It just seemed like a better fit to the hull you have, while still meeting your need for short spars.

[Doug Lord]

Without a doubt the best systen so far for canting keels is CBTF. Upwind with the collective feature you can dial out leeway by turning the rudders both the same direction; manouvering is done by turning both rudders opposite directions.
I've just received a completed design for a One Meter model from Graham Bantck for a F100 CBTF. He used a proprietary VPP in doing the design and "raced" the F100CBTF against a fixed keel F100: no contest: the delta's ranged from 100-to 500 seconds per mile!
We were concerned that the wetted surface penalty in a model(greater than fullsize) would cause design problems that might not allow CBTF to work at this scale: wrong! Now to the building....

[tspeer]

Just for comparison, it would be really interesting to see how the CBTF compares to a 1 meter multihull of the same sail area!

I'm curious as to how the VPP handled trimming of the foils for zero leeway and what effect that was assumed to have. Or did the VPP not take the heading of the hull and yawing moments into account?

[Doug Lord]

The current VPP was not able to model the foils for zero leeway; So the results are correct not just for a CBTF model but for a model using a fixed daggerboard as well; another model sailor and designer, Craig Smith is working on a VPP for canting keels. Bantock consulted with the Wolfson Unit on modelling this boat correctly.Contact me by e-mail and I can give you his e-mail if you have any technical questions-nice guy and one of the best model sailors and designers in the world!
So what that means probably is that the CBTF boat using collective steering upwind will have a better VMG upwind than a fixed daggerboard version though downwind the daggerboard model could pull up the board.But CBTF offers one further advantage: reduction in wavemaking drag with the proper positioning of the twin rudders... So I'm going to build two 100% identical models with the same SA : one with cbtf; one with a retractable gybing daggerboard and race them.As a related note: the Formula 100 class ,originally started in France, and just getting started here is the ONLY one meter monohull class to allow movable ballast!! Except, I guess ,for the new just getting started maxZ86 model(yep-86" LOA with canting keel and spinnaker legal) just given the go ahead by Bill Lee.....
There is a new model class called the multiONE(www.rcmultihulls.com) which is a one meter multihull class; max SA 1100sq.inches(The F100=1300 sq.inches). The problem with small multihulls that have high beam to length ratio hulls is that they pitchpole in the blink of an eye--so they need to be sailed with hydrofoils in my veiw. I'm doing a Formula 48 class(LOA 48" ) multihull using instantly retractable main foils and have already built a 56" model that is beyond fast. Foils on small multihulls add a great deal of speed but their biggest contribution is to control.Using a version of Dr. Bradfields altitude control sysem the boats develop their own righting moment and are extremely stable in pitch and roll. They foil in as little as 5mph of wind and the bigger model(F3) has been reported by a couple of owners to tack ON the foils. Never happened in testing(very, very close) but they swear it happens on a regular basis...
So a one meter multihull to sail in more than 5mph of wind MUST use foils in my opinion... In the old days there was even a recognized multihull class in the AMYA(www.amya.org) but they faded away because of the problems I mentioned plus being poor designs. The new models tack on a dime and are incredibly fast around any course.

[colin.goldrick]

Changing the subject a bit, I have been trying to track down some info on the bearings or whatever that are used to support the keel/

Are there any off the shelf items that are suitable? The keel I am designing is only in the order or 3-400 kg (6-800 lbs).

Any ideas?

[ClarkT]

Colin, if you are designing for extremely high bearing loads, I've found HyComp (www.hycompinc.com) materials to be very tough. This stuff is a real pain to machine, but can handle some really high loads.

[DORY]

Crossflow Reduces Speed In The Tangential Direction Intended Especially At Speed Directed To Lifting Craft Through Vertical (upward Pressures). Any Thoughts How To;
1. Maintain Maximum Speed In An Upward Motion (vertical Direction) While Reducing Crossflow And Lifting Stern To Prevent Deaccelaration

[Tactic]

Doug ,
Have you built and delivered any of the Bantock designed F100 boats yet?
Are there any pics on the web?

Cheers
Brett

[Doug Lord]

No F100's yet but three smaller prototypes test ing a new canting keel trunk and on deck Power Ballast System. The trunk works ok(no hump) and the PBS allows the battery to move to windward when the keel does-it is a small curved carbon rack on which the battery module slides-"solid state" instead of the carbon tube racks shown on other boats on the microsail site. One proto is 22" and two are 30" with identical weight and main only rigs but one is set up with CBTF and the other the kFOIL...All three proto's move the keel from center to 55 degrees in just over one second.
The CBTF setup of the one 30 was worked out with the approval/help of Bill Burns of CBTFco as an experiment and has the aft foil transom hung with a removable hydrofoil on the tip.
These boats aren't products so I don't see the value of posting pictures probably until the F100 is ready. But I could change my mind -they're pretty neat looking little boats..
....The model business has been extremely low priority this year and that won't change for a few weeks longer..

[yago]

Lots of canting keels in the current Vendee Globe race, check out pics and videos on http://www.vendeeglobe.org/, more videos on http://vendee-globe.vendee.fr/
On some of the videos it looks as if the keels are cranked up almost to the side of the hull ;-)
But then, almost all the boats now have big and deep lateral daggerboards in addition to the thin keels that are mopre and more reduced to a balllast-holders only.