Thin laminar flow sections for centerboard

[windhagen]

I am trying to design a centerboard
with a limited span of 1350 mm due to flat lakes
and limited thickness of 35 mm due to
construction. Large Centerboard-Class with
angle of attack approx. 2-3°.

To keep the area at approx. 0.45 m², sections
can only be around 8-10%.

I compared 63010, 63012 and similar sections
with Profili and found more robust sections for low
reynold numbers for example Eppler E836. However,
12.5% (E836) thickness doesn´t give me enough area.
I didn´t find laminar sections for thinner profiles.

Who has access to sections similar to 63010 that still
provide large buckets at low reynold numbers?

I will probably start at the root with a 0008 section and
go progressively to laminar sections as soon as the chord
comes into the 63010 region. Good idea?

Thanks

Henning Windhagen

[Raggi_Thor]

Do you lift the board when you go downwind?
In that case maybe 0008 all the way is best.

Do the boat roll a little?
Then the angle of attac changes all the time.
maybe a simple 0008 is the best anyway?

[windhagen]

<<Do you lift the board when you go downwind?
<<In that case maybe 0008 all the way is best.

Yes, only 20% area will still be wetted.

Do the boat roll a little?

<< Heavy boat with large sail area, they start healing
at 2BFt. Not too much healing

<<Then the angle of attac changes all the time.
<<maybe a simple 0008 is the best anyway?

May be, may be not. The angle of attack seems to
be low, but I fear the tacks with poor acceleration.

Thanks
Henning

[Raggi_Thor]

I am no expert at this, but my impression is that laminar profiles often is a disappointment. I think the "bucket" is narrower in the real world than in calculations, or that motion like rolling makes the flow turbulent anyway.

[Tim B]

You need to look at the likely leeway angles you'll be sailing at. You'll find they are rarely in the drag bucket. Most sections are available on the web, and there is a paper describing how to draw NACA sections around.

If you're going to put the effort into building this, you want to get the basics right.

Primarily you need to know two things for each design case: Speed and Angle of Attack.

Just to complicate matters, the hull will affect angle of attack, but not by a lot for a modern racing yacht with U-Sections. More classic yachts with finer V sections will have a lot of effect on the angle of attack. Also, because of the presence of the hull the speed will change with distance down the centerboard.

You must be at least aware of these effects, and also the fact that 2D theory (X-Foil and Profili) are idealisations and so will probably present optimistic results.

3D CFD is the other approach to this problem, but it is easy to kid yourself about the answers.

Good Luck,

Tim B.

[gggGuest]

Originally Posted by windhagen

I didn´t find laminar sections for thinner profiles.

I have heard, though it might easily be nonsense, that the thinner the section the narrower the drag bucket.

[Raggi_Thor]

Sounds reasonable.
INtersting link here, http://www.mboats.no/Hardware/hardware.htm

[Karsten]

I think it's more important to get the planform of the centreboard right and try to achieve an elliptical lift distribution. This will reduce inducted drag which is much larger than friction drag upwind. Downwind you pull the centreboard out anyway.

For example if you find out that you can make the centreboard 10% smaller because you have a lightwight crew or you are out of the low drag bucket anyway you save about 10% of friction drag. If you decrease the chord length and keep the span the same you also increase your aspect ratio which in turn will reduce your inducted drag.

I think the low drag bucket is a bit of a mistery. You have to have all sorts of conditions in place to make it work (exact profile, polished surface, low turbulence, low cl). If in doubt I would go with a regular profile and operate it at high cls (small area) where you get about the same efficiency as with a larger laminar centreboard in the low drag bucket. I designed a high aspect ratio, small surface area foil for a lightwight 505 crew and it absolutely worked a treat. Once they got going (which was difficlut because of stall at low speeds) they were as fast as the heavy guys in heavy wind. It also damped gusts better and could be left down going downwind. No fiddling with the foil at the top and bottom mark.

[Raggi_Thor]

Just remember that there is absolutely no good changin g the planform if you can't keep the shape (profile) perfect all the way!

[boogie]

hi henning,

i'm mostly with tim b above.
find out the loads on the foil and then go from there.
with the load on the foil and the range of speeds you are rying to achieve you can come up with a range of lift coeffcients or angles of attack which your foil has to work at.

you can modify sections like scaling the thickness and analyse them quite easily with a program like XFLR5 [GUI frontend to Xfoil and a bit more].
http://www.geocities.com/xflr5/xflr5.htm

the E836 section scales quite well down to 7% depending on what you are trying to achieve [i use a modified version for speed windsurfing fins for it's high cavitation inception speed], but my guess is that for a centreboard they will be too fine and the "groove" they work in too narrow, might not even suit your loads with the span and thickness restrictions at all.

a good allround but still laminar section which scales well too is the PCK310
http://www.esotec.co.nz/condorkeel/Html/geometry_F.html

another very nice section is tom speer's H105 hydrofoil section after removing the camber and scaling the thickness with XFLR5.
http://www.nasg.com/afdb/show-airfoil-e.phtml?id=1187

i found for the purposes i looked into the PCK310 and the H105 zero camber both quite a bit better than the old NACA63 series.

good luck with your project
boogie

[frosh]

Without even referring to the analysis suggested by other posters I would personally use the 010 NACA section, which is probably the most proven one for centreboard boats. If you go to 35mm thickness this allows a chord of
350mm and using the maximum depth of 1350mm allows you to achieve an area of 0.45 sq m fairly easily. A thinner section than 10% requires a small angle of attack all the time, which will not occur during tacking. The speed difference between 008 NACA and 010 NACA will probably not be measurable on a fairly heavy boat with large sail area when travelling a straight line.

[Tim B]

I'm not sure you're quite on the right track there, frosh. While I will agree that generally thinner sections stall earlier, it is not always the case. and one should be careful when generalising.

Here is a nice picture of separation at the trailing edge. This was done with an Euler solution, so the flow has probably stayed attached a bit longer than it would in reality.

Tim B.

[frosh]

Tim, with respect what is the need for a thinner section than 10%.
I don't understand what it is that you are trying to achieve?
The simulation shown in your posting is of an asymmetrical section. If you were to fill in the flat surface with a curved body making the foil symmetrical it would be quite a bit fatter than 10%.
Also stiff foils underwater generally work better, and thickness equals greater stiffness. Do you agree?
BTW I have built around 20 centreplates/rudder blades to approx. 010 NACA for the centreplates and 012 NACA for the rudder blades and have always been satisfied with the performance allround.

[Tim B]

Ok, the pretty picture above was just an illustration, but at 15 degrees you see pretty much the same pattern for most aerofoils, some have more seperation, some have less.

What we are trying to achieve when we optimise a keel or rudder is reduce the onset of seperation and stall, with the minimum possible drag. And whilst I agree that the NACA 0010 and 0012 are adequate foils, they are by no meaans the best.

The idea in going to thinner foils is that there are better lift/drag ratios, however, the stall problems are likely to get worse, and that is where the detailed analysis comes in, so we can get them to work at high angles of attack.

Stiff foils generally work better, but I think (with a bit of clever design) that in certain cases flex may be made to work to your advantage. There is nothing to say though, that with modern materials a thin foil could not be adequately stiff.

nb. frosh, please use the full NACA designation, otherwise you'll confuse everyone when we are discussing 4,5 & 6 series aerofoils. Also, sailing in a straight line does not infer a low angle of attack. Upwind the leeway might be quite considerable.

Tim B.

[tspeer]

Originally Posted by gggGuest

I have heard, though it might easily be nonsense, that the thinner the section the narrower the drag bucket.

Quite true. See Summary of airfoil data.

The reason is because the thinner sections form a more pronounced pressure peak at the leading edge as angle of attack increases. The adverse pressure gradient on the downstream side of the peak causes laminar separation, followed by transition to turbulent and (hopefully) reattachment, forming a laminar separation bubble. With a thicker section, you can shape the leading edge so as to delay the angle of attack at which the pressure peak occurs. With a thin section, you've basically got nothing to work with - it's going to act like a flat plate.