Originally posted by gripen
The good point in the parabolic drag polar is that it has a theoretical explanation in the lifting line theory. I could form a formula, say the form Cd=Cd0+X1*Cl^3+X2*Cl^2 (X1 and X2 being constants) or something, which might work better at low and high Cl values, but such formula has no theoretical background.
Yep, but take a look at these three drag polars.
You can see by inspection, that only the NACA 0006 can be modeled with a parabolic polar. The NACA 4412 might yield to curve fitting, and because it has two distinct curved portions with a portion of high curvature between, there is a good chance that the slope of dCd/dCl^2 would have two distinct straight line portions, so it could be treated with two e values, one for each part of the envelope (check your copy of Pope’s book). However, the laminar flow wing with its classic drag bucket won’t yield to either approach. In those cases, a parabolic model is only good for one out of three of them, so a parabolic model may have a good basis in theory, but it is more useful in the classroom than it is in practical aerodynamics. The good news is that all three of those polars can be modeled in exactly the same way by aero engineers working directly from the raw data, and each can be represented with almost any degree of fidelity required, despite the discontinuities.
Originally posted by gripen
Anyway, I've been under impression that in the AH the drag rise is modeled with AoA ie not by making some kind of Cl based model.
I agree, AH has an excellent flight model, that responds well to classical analysis.
Hope that helps...
Badboy
