Overview of the L9 Analysis of the Wind Data

To expand the coverage of the parameter space while using as few tests as possible, we structured 18 of the tests as two fully-saturated 'L9' arrays. This provided 4 levels for each of the experimental factors, but provided no information on interactions between the factors. There was also no noise estimate provided by the experiment. The advantage of these tests was that they demonstrate in more detail the effects of changing an individual factor. The factors for the tests were:

Group 1 (first night)

Elevation (El) at 30, 45, and 60 degrees (actually measured as Zenith angle, which is 90-El)
Azimuth Angle of Attack (AoA) at 0, 90, and 180 degrees
Upwind Vent Gate (UVG) Open, Half, and Closed
Downwind Vent Gate (DVG) Open, Half, and Closed

Group 2 (second night)

Elevation (El) at 30, 45, and 75 degrees (actually measured as Zenith angle, which is 90-El)
Azimuth Angle of Attack (AoA) at 45, 90, and 135 degrees
Upwind Vent Gate (UVG) Open, Half, and Closed
Downwind Vent Gate (DVG) Open, Half, and Closed
A complete list of the tests, by group is also available, as are some comments on the stability of the wind conditions.

Results

Generally, the results for each factor are as follows:

Elevation
The elevation angle is still not a significant contributor. The first group data suggest that the 60 degree elevation may be somewhat lower than the other angles. However, the L16 data and the second group of L9 data all indicate that the elevation is not a significant effect. The lower value in the first L9 may be an interaction effect between two of the other factors, or it may be a real effect, but that cannot be resolved from just these tests.
AoA
The first group reveals, not surprisingly, that the effects are generally higher in a cross wind a direct head- or tail-wind. The tail-wind is the least disturbing condition. The second night indicates that the quartering head-wind is worse than the direct cross wind, and that the quartering tail-wind is less of a disturbance than the direct cross wind.
UVG
For both nights, the results indicate that the 'Half' and 'Open' conditions are essentially identical. This suggests that for protection against wind, the half open position is an upper bound on the width of the gate.
DVG
For both tests, the results show only slight variability, but suggest that there is some increased response, particularly at higher frequencies, for the 'Half' position.
The largest effect, not surprisingly, was the UVG position.