Sunday, May 22, 2011

I Think it'd Be Cool Build a Wind Tunnel.

You know what would be fun to have? A wind tunnel. Something small that I could run tenth scale cars in. I have some projects that need aero development. With my 3D printer, I could make models of the cars  and then test front wings, rear wings, flat bottoms, rake, ride height, splitters, diffusers and ducts.

Last summer while taking some classes at ASU I saw their wind tunnel and I don't think it would be hard to build. I roughed out a sketch based on what I remember, and it looks like this.

After doing some snooping around the Net I found a company called Aerolab that sells wind tunnels. I'd guess ASU has one of their designs.

This type of wind tunnel is called an open circuit wind tunnel. It is less expensive to build than the closed circuit wind tunnels and can be built as large or small as you can afford.  The difference between open circuit and closed circuit should be pretty self explanatory based on the name. With the closed circuit design the same air is recirculated, which has obvious efficiencies, and the open circuit design sucks in air from outside and accelerates it.  One of my favorite open circuit design is A2 at AeroDyn in North Carolina.

Open circuit designs are typically made up of five sections, the settling chamber, contraction section, test chamber, diffuser and fan section.

The settling section usually has some screens and honeycomb to equalize and align the air flow as it comes in from outside.

The contraction chamber which accelerates the air and reduces variations in pressure.

The test section is where the model being tested resides. Here things like smoke nozzles, and test devices are also inserted. 

The diffuser section gradually transitions from the test chamber the the fan section.

The fan section of course has the fan.

Any volunteers to help build this? You can use it when I'm not. 


  1. Sure I'll help out! Now let's see, where are those 12' fans I had laying around...?

  2. Only problem I see, is that air doesn't scale that well from what I've read, but you would have baselines compared to other models I guess, and you could still tell which one was the most efficient.

  3. Even though it doesn't scale perfectly I think the visualization would have value. And I already have a 36 inch fan.