Hi all,

For the last 6 months, we need a wind tunnel to calibrate our pitot tubes, but since we are under budget, this dream was never realized. We are now attempting to build one from scratch, and since I have a little background on CFX 5.5, I voulenteered to construct a model just to see how it works/not before we build it.

After looking up at several wind tunnel design, I decide to model a rectangular wind tunnel with straighteners. The source of wind comes from a triple blade fan 30 inch in diameter rotating at max 1000 rpm (adjustable speed). The goal is to have a relatively steady state and of straight flows about 30 ft/s in the testing section, where we place our pitot tubes to be calibrated. I still need to determine the converging section angle, and how long the test section should be.

So, I built my geometry using CFX build in CFX 5.5 package. I encountered a problem of trying to model my wind tunnel inlet (wind coming out of the fan), I didnt know what best to model my inlet. Can anyone please help, and should I model the whole as transient flow or steady state?

This is really a challenging project, and I would like to have it realized. Thank you for any help you can offer.

Pandu Sattvika

Hi Pandu, I should imagine the important aspect is the amount of swirl you get from the fan and then how well the straightening vanes remove this from the flow. In our wind tunnels, one of which is a large atmospheric boundary layer tunnel, we have 2 fans then upstream of the contraction we have a 2 foot honney combe straightening section. How you model the straigtening section and the inlet conditions will be fundamental to your results. You could use CEL to impose a swirl onto the incoming flow (one of the tutorials details how to do this). I'm not sure how well this would predict the flow phenomenon, but it would be a good starting point. Turbulence models will also be an issue for you. We did a brief study where we were looking at a new wind tunnel design and weather or not we needed fillets in the corners to reduce the vorticies in the working section after the contraction. Unfortunately we couldn't complete the work due to other project requirements, but my mesh sizes were soon getting rather large as I was trying to capture these effects.

Hi Pandu

Not as such a CFD reply, so please no flame mail! ;-)

If you haven't started constructing your tunnel yet, I'd strongly urge you to look at a suction type tunnel (ie fan behind the test section) rather than a blowdown type (fan ahead of the test section). Having the inlet of a suction tunnel in a relatively still area would eliminate much of your flow rotation problems, and by adding a relatively fine grid would help break up residual turbulence. Especially for the speeds you are looking at, I would think a suction type tunnel is very suited to your needs. We've got a big one that goes up to 100 m/s with 0.9 x 1.2 m test section and it works like a charm ;-)

Suggested reading: Low speed wind tunnel testing by Alan Pope... I know a new edition recently came out and I found it very helpful in some experimental projects I did. It covers quite a lot of wind tunnel design theory.

Back to the CFD then: having a suction type tunnel would elimanate your problem of specifying inlet conditions...

Best of luck

Louwrens

Hi,

My sincerest gratitude for your opinions, Mr. Bob and Mr. Louwrens. I would try the suctions wind tunnel, it seems like a very good idea.

Pandu