|
The "settling chamber" or "stilling section" is the largest cross section, and contains a honeycomb and/or screen. A honeycomb with its cells aligned in the flow direction will reduce mean or fluctuating variations in transverse velocity (flow direction), with little effect on streamwise velocity because the pressure drop through a honeycomb is small. Woven-wire screens mainly reduce streamwise velocity fluctuations, with little effect on flow direction because the "refractive index" of a screen is small. The settling chamber itself is never long enough for pre-existing tunnel turbulence to decay significantly, so the name is misleading.
The honeycomb, always mounted upstream of the main anti-turbulence screens, should be far enough downstream of the start of the parallel-sided settling chamber for the flow to be roughly parallel to the axis. If a wide-angle diffuser with screens is mounted upstream of the settling chamber it will usually have a screen at exit where the wall angle changes suddenly, and this screen will help to reduce the flow angle. Also, a screen may be mounted ahead of the honeycomb in an open-circuit tunnel, to reduce large deviations in flow angle which might "stall" the honeycomb.
|
The distance between screens is not critical and a minimum is usually set by the need to mount the screens firmly, in removable frames or otherwise. Screens can be clamped, without adhesive, between two wooden frames. Tight pre-tensioning of the screens is not necessary: as long as there are no wrinkles, the airflow will pull the screen into a smooth shape (but note the comments in the Screens and Honeycombs section on the effects of large screen delfections on the boundary layer). The last screen should be far enough upstream of the contraction that the flow speed is still constant over the cross-section: near the entry to the contraction, the flow starts to slow down near the walls and accelerate nearer the centerline, so the pressure drop through a screen placed too near the entry will vary over the cross-section, leading to velocity variations in the test section.
|