Virtual wind tunnel?

[MCarlton]

Basically, I have a few ideas for wing sections which I would like to investigate. However, I'm not smart enough to draw up coordinates for them, camber lines and the like, I just draw what looks right with a pencil.

Now I know thats a hopeless way to design, and I'm not claiming any different, but what I'd like is to be able to scan my hand drawn wing sections into a programme which will run a "virtual wind tunnel" and give me an idea of the section's performance in comparison to other sections.

I have an idea which I scribbled up last night, which looks promising (to my untutored eye) for high lift/STOL applications. As well as some general sport model sections and sailplane sections I'd like to "try out".

Yes I could build wings with them, but that wouldn't give me an empirical performance comparison.

Any thoughts?

[nmasters]

There's an "import from raster image" option on the <import> menu in Profili that dose that.

--Norm

[BMatthews]

... and from there you run it through the Xfoil plug in that comes with Profili 2.


Thought I'd drop the other shoe just in case you didn't know that Norm's reference was intended to let you input your scan into something that would generate the sort of curves you want.

In particular look at the pressure gradient curves. One aspect for producing successful airfoils is to generate smooth gradient curves. If you look at the curves for the MB 253515 and MB303515 and then compare these to the Selig 4233 you'll see what I mean. In his original work report Selig tested the Micheal Bame airfoils and found some undesireable issues. His solution was the 4233 which corrects these and should produce an airfoil with better overall performance. The sort of bumps in the pressure gradients you'll see in the MB airfoils is just one aspect that you will want to avoid so you don't get any unexpected stalls or other odd effects.

Keep in mind too that how you build the wing makes a huge difference. The best airfoil you can find will be reduced to no better than a mediocer one if the building method uses open bay construction that allows the critical areas of the airfoil to sag between the ribs and destroy the control over the airflow. Things like the old D box design when appllied to the Eppler 205 and Selig 3021 failed with gusto due to the sudden covering sag at the airfoil's highpoint. However extending the upper sheeting back to around the 50 to 55% point of the wing supports the covering until it's past this critical zone and lets the airfoils perform up to or nearly up to what the wind tunnel and Xfoil both promise.

For your high lift airfoil you'll want to compare the Xfoil results to the Selig 1223. And keep in mind that airfoils with extremely strong curves such as the trailing edge of the 1223 require fully sheeted or vacuum bagged styles of construction in order for the curves to come out properly.

[nmasters]

Oh, sure, it's easy to anticipate future questions when your head isn't full of snot. :mad: :confused: :p

--Norm

[MCarlton]

Excellent, thanks folks.
I'm not aiming to set the world on fire with something amazing, I just have "ideas" and would like to see if they actually work.

The best airfoil you can find will be reduced to no better than a mediocer one if the building method uses open bay construction that allows the critical areas of the airfoil to sag between the ribs and destroy the control over the airflow.

Amen to that, and its one of my little bugbears, where people say they are going to use xyz section because it looks good in the polars, but I wonder how many people can actually build accurately enough to truly represent that section.

I'll see if I can get hold of the programmes and play around for a while.

The high lift section is a little "out of the box" and I just want to see if the theory works. I might scan a drawing in so someone can give me a heads up if I am onto a complete loser!

[BMatthews]

Get well soon Norm.... :D

[nmasters]

I just draw what looks right with a pencil.

Now I know thats a hopeless way to design, and I'm not claiming any different
That may not be the "correct" way to do it but you have to start somewhere and that has worked as well as anything for me. For instance a couple of years ago, before I had Xfoil/Profili, I was trying to produce an airfoil section with Cm0~0 that fit into (or rather around) a specific structure. My method wasn't exactly analytical. I picked a Wortmann section with the thickness and leading edge shape I wanted and raised the trailing edge to a position that looked about right. Then I just drew a straight line from the TE to a tangent on the lower surface. The upper surface was defined by simply copying the thickness of the original airfoil. The amusing thing is that last summer I finally met (via an e-mail list) somebody who was interested in the same retrofit. He had been going about modifying the same seed by reflexing the mean line because Jim Marske told him that that was the correct way to do it. He produced several versions by this method and non of them had as good glide as mine. Here's a graph from Profili showing its L/D and Cm compared to some symmetrical sections (http://www.rcgroups.com/forums/attachment.php?attachmentid=2251357). Farther on in that thread there's a chart showing the affect that tripping the boundary layer has on pitching moment. Very interesting stuff that wouldn't be obvious by building models and observing there flight characteristics.

--Norm