Creating controlled turbulence in a home-built wind tunnel

[gsjbjt]

Hi, I have built an 8' long wind tunnel for use in an experiment to test the effects of turbulence on several variables. However, seeing as this is an educational-level/novice wind tunnel with a fan rating of ~1800 cfm the wind speeds only really go up to ~10 mph. From my understanding, this speed is not sufficient to really produce significant turbulence, and as my experiment is highly dependent on measuring the observable effects of turbulence, I was thinking that it would be necessary to "artificially" create turbulent flow by some sort of obstruction in the test section (opening measures 1' by 1', 2' long). However, I don't have a lot of time or money to build something elaborate-- does anyone know what sort of setup would get at least somewhat turbulent flow for testing?

 

[Simon Bridge]

You get a lot of turbulence around just plane surface turned face-on to the normal flow so just string a bunch of polystyrene shapes onto wires strung across the from of the fan. Anything that interrupts the flow should do - rough things better than smooth things. Think the opposite of aerodynamic.

You'll have to play around a bit to get something that shows what you need.

 

[boneh3ad]

Before I overthink this, I'd like to ask a clarifying question. Just to be certain, are you trying to increase the turbulence intensity in the free stream or are you trying to measure turbulence in the boundary layer over your test article?

 

[bigfooted]

Most wind tunnels I've seen have a converging section to increase the velocity in the test region. They basically look like big Venturi's. You just need to be sure that the walls of your test region are far enough away from your test object (or scale down your test object) so the measurements are not influenced by wall effects.
Also, in a large number of experiments you want the flow to be laminar before arriving at the test object, the flow will become turbulent mainly in the wake of your test object. Already creating turbulence well before your test object might be interesting, but if it is not really necessary then don't do it because it will only result in a large flow resistance and a lower maximum flow velocity that your fan can deliver.

Note that your Reynolds number depends on the size of your object. If you don't put anything inside your wind tunnel, then your characteristic length scale is the hydraulic diameter of your wind tunnel. If you place a pingpong ball in it, the characteristic length scale is the diameter of your pingpong ball and you need an even higher velocity to get proper turbulence.

What kind of fan do you have? Some domestic appliances have quite good fans that can probably beat your current fan.

 

[SteamKing]

In ship model testing in model tanks, the ship models have a line of 'turbulence stimulators attached at the bow of the model which run vertically below the waterline. Each stimulator looks like a raised button. The flow of the water around the bow of the model is disturbed by the stimulator and ensures that any laminar effects downstream have been interrupted.