|
The modeling requirements for validating viscous computational fluid dynamic simulations of a high-lift trapezoidal wing were investigated. This wing has a full-span slat and a full-span flap, and has been tested exten-sively in the Ames 12-Foot Pressure Wind Tunnel. Because of the size of the wing, there are significant facility effects in the data from the 12-foot wind tunnel. The objective of the work was to quantify the effect of wind-tunnel walls, and to develop a computational model that would effectively simulate this effect.
Two different computational models of the test facility--differing in fidelity--were developed and tested. The first of these is a complete viscous simulation of all surfaces within the 12-foot wind-tunnel test section. This computational model more than doubles the cost of computing the flow over the trapezoidal wing, as compared to simulating free-air conditions. The second model of the wind-tunnel test section was a highly simplified straight duct, whose wall was treated as an inviscid surface. The surfaces used in this second model are shown in figure 1. Simulations with the second model required the same computational time and memory as a free-air simulation.
The computed results were compared with experimental lift, drag, and surface pressure data. The results showed that the simplified inviscid model of the test section performed as well as the high-fidelity, viscous test-section model. Computed results generally compared very well with the experimental data at all but the highest angles of attack. Figure 2 is a plot of the lift coefficient versus angle of attack for the computational simulations and experimental results. The computed results include lift from both of the wind-tunnel models, as well as a free-air simulation. The experimental results include the original uncorrected data, and data which include theoretical corrections to exclude the effect of the wind-tunnel walls. A comparison of computational results from both free-air and wind-tunnel simulations at the same lift condition indicated that it was necessary to simulate the wind-tunnel in order to perform validation using the 12-foot wind-tunnel experimental data.
Point of Contact: Stuart E. Rogers
(650) 604-4481
rogers@nas.nasa.gov
Back To Top
Previous Paper
Return to Revolutionary Aviation
Next Paper |
|
Fig. 1. Surfaces of the trapezoidal wing and simplified wind-tunnel test section.
|
Fig. 2. Lift coefficient versus angle of attack.
|
|
