Sail7: sail's performance

No no, we are talking about the vertical variation of the vertical component of the undisturbed (freestream) flow.

 

Errr... no, we don't.

I have understood the windshear in the aeronautical sense. It is not. Found the discussion about it: http://www.boatdesign.net/forums/hydrodynamics-aerodynamics/asymmetrical-speed-polars-45006.html
A vertical variation in the freestream wind direction (horizontal) due to some geographical features around the sailing field.

 

That's the way I've understood Mikko's comment (not shure if right). It is a common phenomenon in our lakes (surrounded by mountains) . The most visible consecuence is that a perfectly balanced, well trimed boat, will go to windward at slightly different angles/velocities when in different sides (and will require different trimming configurations also). This situation appears and disappears sometimes in the same regata. I never undestood it well...

 

Check the link in my previous post.

 

I'm reading, I'm reading...

 

- Is input truewind with a gradient + boat velocity? That would be nice.

- The mirror plane is necessary for meaningful results. The hull (deck) would help a little, as an endplate to the jib, but without the mirror image the CLalpha will be all wrong, and consequently lift & induced drag, too. In my MacSail I found that a mirror plane at about 50% freeboard height gave best results compared to windtunnel tests (this was for dinghys with low freeboard, Europe & 470, and without the hull modeled).

 

The wind shear you (and Tom) are speaking about is rarely a geographic feature, but rather a meteorological one. But forget about that for saildesign, just a combination of the true wind with a velocity gradient and the boat motion creates a twist in the apparent wind - that's interesting for saildesign. The asymmetrical twist Tom was referring to in his polar post is not relevant, since it's a special case, and you have to design the sails for both tacks anyway.

Still about the mirror image: You need to have the sailplan mirrored under the waterplane (or some other plane between the sail foot and the sea), if your boat with sails is just "floating in the air", the result will be wrong. It's like calculating an airplane with one wing only - the lift on the wing will be quite different when you add the other wing on the other side of the fuselage.

 

That's exactly what I was suspecting. Thanks for clarifying that one.

 

Hi all,
All in all, I agree with what is said above about the VLM and it's limitations. The method is valid essentially at low a.o.a., and does not account for flow separation.
For wings operating at low Re, this is compensated in the original xflr5 by interpolating viscous drag on pre-calculated values issued from XFoil which is a 2d viscous solver. This is clearly an approximation, and I have nothing similar to propose for sails.
I derived sail7 from xflr5 because it required limited development work, and because it may in the future be used as a basis for more complex models which could include simulation of flow separation. There is little chance that I will do this work myself because my primary interest is in (identified) flying objects, but if somebody is interested, they are welcome to it and I will help as I may.

Last, to answer more specific points : yes, there is an option in the polar definition dialog box to include symmetrical reflection about the z=0 plane, and yes the hull representation is clearly a bad approximation which is best left out of the model. Also, the wing gradient model is very crude, and could easily be improved.

André

 

André, thanks for posting!

Ok, that's the "ground effect" option, found at "Polars-Define an analysis" and "Polars-Current Polar-Edit", see attached image, it shows also the wind gradient definition menu.