Conclusions

Answers to the questions:

• Does the separation arise in the leading edge region or move in from the wing's trailing edge region? It arises locally for thin airfoils.
• Can separation be avoided for all times with suction? Yes.
• Can separation be avoided for all angles of attack? Yes
• At what location should suction be applied? Near the nose, before separation starts and continuing sufficiently far downstream.
• Can suction be applied at a single location or should it be applied over a region? Except for fairly small angles of attack, or excessive suction volumes, suction over a region is needed.
• If the latter, over what sort of a region? Fairly long, according to the derived minimum sizes.
• What is the minimum amount that needs to be sucked? Is the minimum suction volume indeed inversely proportional to the Reynolds number? Presumably locally minimum suction gives the minimum amount. For an increase in stall angle by a fixed factor, the suction volume is small proportional to the Reynolds number based on the nose radius, and additionally small since the region of suction is proportional to the nose radius.
• What is the effect of unsteadyness? Suction can be started late and initially only over a narrow region.
• Can we apply suction in a better way? Maybe.
• 2D computations are typically accurate at 128² mesh points and can be done for 4096² points in a small fraction of an hour on a fast workstation. A factor 1,000 in computational time and accuracy. No fun.

10/16/01 0:38:14 10/16/01 0:39:18