Return to Winged Keels

Voiled

Member
390
284
#5,6  fins too short for any tip vortex reduction effect
Have to disagree on this one. The #5 example is certainly too fat. Vortex reduction fins can be quite small. Look at all the airplane examples and this, crappy, illustration from the 2003 AC:

image-28.jpg


 

IStream

Super Anarchist
10,788
2,972
FWIW, here's the keel on my Catalina 50. Draft is 5.6' / 1.7m, ballast is 13,250 lbs / 6,000 kg. Overall dry weight is 36,000 lbs / 16,330 kg. Full specs here:

https://sailboatdata.com/sailboat/catalina-morgan-50

Plenty of wetted area but she's a pretty stiff boat with a nice motion. Decent compromise for a Florida boat. Here in Seattle, a Max Prop and big asymmetric help recover some light air performance. 

/monthly_2017_06/large.keelAft.jpg.d24c8576183674588dba0b42e9715ba2.jpg

/monthly_2017_06/large.IMG_20151218_094943.jpg.edcb5f144bcbdd0982f761f66d3246c7.jpg

 

DDW

Super Anarchist
6,317
1,012
In aircraft, winglets have probably been more studied for sailplanes than anything else, as tip vortices (induced drag) are all important. Very well designed and modeled by a team of university students, you can get about 0.5% improvement in performance at some speeds, while sacrificing performance at other speeds. Not well designed - well, they can still look cool. On a Boeing, which flies at the same speed and conditions 95% of the time, you can tune them to do some good. I put the factory winglet kit on my sailplane, no measurable difference in performance before and after, I could circle a couple of knots slower so maybe they delay tip stall and spin departure a little. They do look cool on the ramp. Like PHRF you get a rating hit when you install them in handicapped competition. Like PHRF I don't think they make up for the rating hit. 

Dangerous to compare a sailplane (AR = >30) to a boat keel (AR ~ 2). In a boat keel the whole thing is a tip vortex. For most of 'em, I think it is a place to put more lead, makes blocking a little easier, and ungrounding more difficult. The cool factor is only important at boat shows (admittedly important for the builder) since you can't see the wings at the dock. 

 

Zonker

Super Anarchist
9,288
5,218
Canada
Have to disagree on this one. The #5 example is certainly too fat. Vortex reduction fins can be quite small. Look at all the airplane examples and this, crappy, illustration from the 2003 AC:
Hint - look at the chord of the fin. You can't compare the flow around a keel with and without a bulb. Totally different. And airplanes versus sailboat keels.

 

MFH125

Member
164
166
Something to remember about AUSTRALIA II, is that the major innovation in the minds of the designers was the inverted keel shape, not the wings. 

By having a longer root chord than tip chord, the center of lift on the keel was lowered.  Getting the low pressure region on the keel further below the surface apparently resulted in a noticeable reduction in wave drag. This keel arrangement appeared to be a significant improvement all on its own in both their panel code simulations and tank tests.  The one downside of the upside-down keel, was that the larger tip resulted in more induced drag.  The winglets were added to try and combat that.

In aircraft, winglets have probably been more studied for sailplanes than anything else, as tip vortices (induced drag) are all important. Very well designed and modeled by a team of university students, you can get about 0.5% improvement in performance at some speeds, while sacrificing performance at other speeds. Not well designed - well, they can still look cool. On a Boeing, which flies at the same speed and conditions 95% of the time, you can tune them to do some good. I put the factory winglet kit on my sailplane, no measurable difference in performance before and after, I could circle a couple of knots slower so maybe they delay tip stall and spin departure a little. They do look cool on the ramp. Like PHRF you get a rating hit when you install them in handicapped competition. Like PHRF I don't think they make up for the rating hit. 

Dangerous to compare a sailplane (AR = >30) to a boat keel (AR ~ 2). In a boat keel the whole thing is a tip vortex. For most of 'em, I think it is a place to put more lead, makes blocking a little easier, and ungrounding more difficult. The cool factor is only important at boat shows (admittedly important for the builder) since you can't see the wings at the dock. 
That's interesting, but not too surprising.  Gliders are really the wrong platform to see strong benefits from winglets -- they're already so optimized for low induced drag.  Winglets rely on spanwise flow induced by the tip vortices to generate thrust.  When your AR is greater than 30, the tip vortex is small, and you get very little AoA on the winglet, which means very little lift, which means very little thrust relative to their area... which means they're hard to justify.

The original NASA technical report by Whitcomb which kicked off all the winglet fuss in the late 70's demonstrated a ~20% reduction in induced drag for a wing with an AR between 6 and 7, if memory serves.  Modern airliners tend to have wing aspect ratios of around 7-9.  So, compared to your average glider with an AR of approx. 30, an airliner has 3.75x more induced drag.  Even on airliners, winglets only really make sense if you have a fixed span.  The new Boeing 777-9X went with folding wing tips for a higher AR (AR = 11, apparently), rather than winglets -- it will be interesting to see if that becomes a trend.

 

SemiSalt

Super Anarchist
7,752
269
WLIS
It's like having the bottom of your keel molded to the shape of a Bruce Anchor... who wouldn't want that?!?

Yes, the whole whole boat moves with the wave but unless the water is also moving up and down equally, horizontal surface will do something good for once, and dampen the motion thru the water. As Zonker says, it's almost certainly not worth the added drag of that surface area in the first place.

- DSK
As the boat heaves, the keel is pulled upward through the water, but the effect is lessened because the wave action also moves the water up. Also, the boat is moving forward. The combination means that the angle of attack on the wing can be large, at least briefly. I think it's a pretty common observation, noted above, that the wing is a greater handicap in waves, especially waves large enough to give the boat a pitching action. But still, the boats do get to the windward mark even if the deep keel boats get there first.

But how big is the overall effect compared to other compromises? Is the hit taken by a wing keel boat bigger or smaller than the hit taken by a boat with an all-purpose roller furler in place of suite of sails for every condition?

 

Rasputin22

Rasputin22
13,690
3,264
I was convinced that I could 3d model and print winglets for my AC-4 sailplane. I had seen a lot of the after market tips like DDW mentioned but didn't want to alter the original wing tip shape. I figured that I could get a close fit of the mating surface of the printed tip to the wing itself and use 3M VHB double sided tape to affix the winglet. That would also let me adjust the toe in as well as the dihedral within a narrow range. If a tip peeled off in flight no big deal I could just print a replacement. After a lot of research I came across one of the authorities in the glider world who said something like this, 'Adding winglets to a sailplane has a small chance of resulting in a small improvement but a big chance of screwing up what was a perfectly adequate design in the first place'!

oRrq-RfgPEV-5wKmeL-ihK_IoR8j-fMKgilqRO-3hHll94Dg_2RWCqHYofArPxrSdCiS9d1G_GFCAA


 

Panoramix

Super Anarchist
The twin keels are a great alternative either side of the English Channel for sitting down gracefully when the tides out without resorting to stilts, but they are not a match for performance for the same effective deeper single span (and sitting down can be a lot worse than graceful.)
I will take a well designed twin keel over a well designed wing keel every time.

A good twin keel will go upwind very well which IMHO matters on a cruising boat for obvious safety and comfort reasons. They will be a bit slower downwind but who cares if you get say an hour later on a 24h downwind crossing. Plus you can go in places that dry which is a real plus.

 

DDW

Super Anarchist
6,317
1,012
The original NASA technical report by Whitcomb which kicked off all the winglet fuss in the late 70's demonstrated a ~20% reduction in induced drag for a wing with an AR between 6 and 7, if memory serves.  Modern airliners tend to have wing aspect ratios of around 7-9.  So, compared to your average glider with an AR of approx. 30, an airliner has 3.75x more induced drag.  Even on airliners, winglets only really make sense if you have a fixed span.  The new Boeing 777-9X went with folding wing tips for a higher AR (AR = 11, apparently), rather than winglets -- it will be interesting to see if that becomes a trend.
Absent other constraints you are always better off with more span than winglets. The very first NACA papers on winglets state this, and it's been true every since. 

Probably bigger difference than AR is Cl. Induced drag is proportional to AR and Cl. Airliners fly at low Cl at cruise speeds, those speeds being about 4x what their stall speed has to be (so dynamic pressure 16x, and Cl & induced drag 1/16). Sailplanes on the other hand have two primary speeds, thermalling ( at min sink close to stall speed) and high speed cruise between thermals (around 2x - 3x stall speed). It is much easier to optimize the winglets for one condition than two. Add to that the same design sailplane has to be competitive in weak eastern conditions and running at 75 knot (Cl about 0.4) vs strong western conditions running at 130 knots (Cl about 0.1). We fill them with water for strong conditions but still a big difference. 

While you might get a 20% decrease in induced drag on an AR of 6, that will occur only at high Cl. At low Cl profile and parasitic drag dominate. The AoA of a sailboat keel is pretty close to its leeway angle, so likely Cl not very high. 

Actual tank tests on keel designs prove that it is very hard, sitting in an armchair, to guess what the performance will really be, especially at low AR. 

 

Rasputin22

Rasputin22
13,690
3,264
DDW, I still have a hard time wrapping my head around the practice of 'filling our sailplanes with water' and I know the theory and stats and all but it seems contradictory to me. That is until I think back to skiing down hill and how a heavier skier will reach a higher speed on the same skis and the same slope. I have never had the privilege of flying in a ballasted sailplane but would love for you to show me sometime just what that is all about. Do you have a photo of your aftermarket winglets? It seems like it would be a very hard thing to do to pick up a saw and shop off the wing tip of a 6 figure sailplane just to fit a unknown winglet. That is as ballsy as flying sailplanes in the first place!

 

MFH125

Member
164
166
Probably bigger difference than AR is Cl. Induced drag is proportional to AR and Cl. Airliners fly at low Cl at cruise speeds, those speeds being about 4x what their stall speed has to be (so dynamic pressure 16x, and Cl & induced drag 1/16).

While you might get a 20% decrease in induced drag on an AR of 6, that will occur only at high Cl. At low Cl profile and parasitic drag dominate. The AoA of a sailboat keel is pretty close to its leeway angle, so likely Cl not very high.
Yes, wing loading matters.  If you barely load your wing, there's less induced drag to minimize, and adding winglets is mostly adding parasitic drag.  The rate at which induced drag increases with additional wing loading, though, depends on the aspect ratio. The Whitcomb paper found that:

At the design Mach number of 0.78 and near the design wing lift coefficient of about 0.44, adding the winglets reduces the induced drag by about 20 percent and increases the wing lift-drag ratio by approximately 9 percent.

That's a fairly typical lift coefficient for airliners, business jets, small planes, etc. at cruise.  Values up to around 0.6 are not uncommon.  Whether those are "high" depends on your perspective.

For keels the CL is low, but so is the AR.  The induced drag of low aspect ratio keels is much more sensitive to changes in CL than a glider wing would be.  Basically, the gain is much higher.  A keel is lucky to have an AR > 2.  AUSTRALIA II's keel has an AR = 0.4.

 It is much easier to optimize the winglets for one condition than two. Add to that the same design sailplane has to be competitive in weak eastern conditions and running at 75 knot (Cl about 0.4) vs strong western conditions running at 130 knots (Cl about 0.1). We fill them with water for strong conditions but still a big difference.

Actual tank tests on keel designs prove that it is very hard, sitting in an armchair, to guess what the performance will really be, especially at low AR. 
100%.  Airliners are relatively easy in this regard. Sailplanes are harder. Keels are harder still.

I dug up the van Oossanen paper about the development of AUSTRALIA II's keel and it's clear that they spent a LOT of time fine tuning the wing geometry.  Different wing dimensions, wing profiles, wing angles, etc. were considered.  The optimum seems to depend a bit on heel angle. The final wings have twist to deal with the changing angles of incidence along the winglet span -- a feature you don't generally see in production wing keels.   The best case scenario seems to have resulted in a 25-30% improvement in lift/drag when going upwind.  A major issue is that if you optimize the wings for upwind work, they generate a lot of lift even when the leeway angle is zero -- which means you're lugging a lot of unnecesssary induced drag around in addition to the extra surface friction when going downwind.  Lots of details to get right/wrong and compromises to make.  It's unlikely that designers without AC budgets, access to panel codes, and extensive tank and wind tunnel testing were getting these kind of details dialed in when designing production keels in the 80's and 90's.

Practically, they seem to work fine.  Well designed ones probably have some modest benefits when going upwind, but most are probably not justified from an overall performance perspective.  But, the wide variety of keels out there makes it clear that there are plenty of ways to get satisfactory windward performance for cruising.

 

Whinging Pom

Super Anarchist
My (new to me) boat, a Dehler 35 CWS has a wing keel.  She was designed by Van De Stadt who know a thing or two, it's not obvious in the photos but the wings have a fair amount of downward deflection (diehedral?) the standard draft is 2m and in wing form it's 1.5m which will be useful when we get back to the Bahamas and the ICW.

 I haven't sailed her that much, but with utterly knackered sails she points very well and makes little leeway at good speed. On my home river, there are two other sister ships, both with full draft keels.  I'm looking forward to having a tussle with them later this year.  

A happy new year to you all.

2021-05-15 11.08.21.jpg

4464723.jpg

4464738.jpg

 
Last edited by a moderator:

Steam Flyer

Sophisticated Yet Humble
42,471
8,678
Eastern NC
That's actually pretty close to what I remember of Australia II's keel; I think her wings were angled slightly forward.

- DSK

 

SemiSalt

Super Anarchist
7,752
269
WLIS
Here is an addition to Zonkers catalog of wing keel types. This is an O'Day 272. Not a real good boat for a variety of reasons. Very shoal draft. IIRC, I read somewhere that O'Day, and that would probably mean the Hunt design firm, admitted they know not from wing keel design and brought in a consultant to help out. 

2022-01-03_11-01-19.png

2022-01-03_10-59-41.png

 

SemiSalt

Super Anarchist
7,752
269
WLIS
A boat that might be fun to compare is William Crealock's Columbia 28 design. As you can almost read in the text, there is a flare at the foot of the keel meant to work as an end plate. This was about 1967, shortly after Noah went aground on Ararat. 

2022-01-03_11-09-23.png

 
Last edited by a moderator:

European Bloke

Super Anarchist
3,407
826
I sailed an MG335 with the winged keel quite a bit. Up wind once you get up to speed she goes like a tortured crab, out of the tacks before you get some flow over the foil it's much worse. The full fin version is a nice yacht 

I agree with the statement that the wings on most cruising yachts are either marketing or misguided.

 

Remodel

Super Anarchist
10,126
801
None
without some pretty sophisticated tank/wind tunnel/CFD most winged keels are pretty draggy and useless. At best they keep the VCG low with shallow draft.
But they make for great anchors when you eventually run aground...

 

Ishmael

Yes, we have no bananas
50,865
11,210
Fuctifino
The venerable Shark 24 by George Hinterhoeller has a flat-bottom flare on the keel. I was interested to see that they are apparently are still being made in Germany.




 

Israel Hands

Super Anarchist
2,629
1,543
coastal NC
FWIW, here's the keel on my Catalina 50. Draft is 5.6' / 1.7m, ballast is 13,250 lbs / 6,000 kg. Overall dry weight is 36,000 lbs / 16,330 kg. Full specs here:

https://sailboatdata.com/sailboat/catalina-morgan-50

Plenty of wetted area but she's a pretty stiff boat with a nice motion. Decent compromise for a Florida boat. Here in Seattle, a Max Prop and big asymmetric help recover some light air performance. 

/monthly_2017_06/large.keelAft.jpg.d24c8576183674588dba0b42e9715ba2.jpg

/monthly_2017_06/large.IMG_20151218_094943.jpg.edcb5f144bcbdd0982f761f66d3246c7.jpg
I sailed with a keel much like that (winglets more aft) for years here on the US east coast, where the bottom is shallow and often mucky. The keel provided much less draft than the fin keel version, and since I wasn't racing, it was worth trading a bit of performance to increase range. When I would occasionally "find" the muddy bottom, it never stuck with suction as naysayers predicted. It's fun to pee on concepts that didn't take off, but most people posting here probably haven't sailed on a boat with a winged keel so don't know how they handle.  :)   (Okay, I admit I'm sailing a fin keel boat now.)

 
Top