I need help designing what i guess is basically a wind tunnel.. long story here, but i need to make a tunnel that is as tubulence free as possible.. nearly any material, shape, etc you can think of.. but i need to channel a large volume of air asd completely smoothly as possibly through a passage... can be any size really too.. being very vague i know, wish i knew more so i could ask better questions... if anyone can try and direct me in the right direction i would be very grateful

thanks!

Hope you got lots of money for this as this kind of technology doesn't come cheap.

Years ago I came across a book on windtunnel design. There was some interesting stuff but it soon showed my that I was out of my league pretty quickly.

The 25 cent explanation. Circular is best but square or rectangular with large corner fillets is not too bad either and it lends itself to a decent work area. To move the air you need a fan. The fan needs a powerful motor if you're moving much air at any sort of decent speed. I have no idea how much power you need but for single pass tunnels (open ended) it takes more power than for a closed loop. If you go for the loop idea you need a sets of airfoiled vanes in the corners to prevent turbulence. The wake off the impeller blades needs to be evened up and straightened. Sets of honey combs with the holes about 4 to 6 times as long as the hole diameter in a series were suggested for this. And for very low velocities with very little turbulence they even used a fabric barrier to help even up the airflow across the whole face of the tunnel and to eliminate the last of the turbulence from the honeycombs.

If you're looking for something to do airfoil testing then a fairly small tunnel would fill a good size room. If you need a test area larger than 3 feet x 3 feet x 5 feet then you'd better buy a large lot. You need that much room for the air to run around and settle before it hits' the test area.

You've certainly got my curiosity up. What are you planning on doing with this tunnel that you can afford the 100 grand or so to build it?

Regal1975,

The word 'turbulence' is used to mean different things. What you need to do to get turbulence-free flow depends a little bit on what you consider to be turbulence. Generally, turbulent flow has some kind of disturbances that cause it to be unsteady ( changing from one moment to the next ). The main distinction to make is the size of the disturbances. If you want to get rid of the large disturbances, which essentially vortices, then turning vanes, as mentioned by BMatthews, might be all you need. If you want to get rid of all disturbances, large and small, you need something more elaborate. First, you need a fairly fine screen, to make sure that you get small vortices throughout the flow ( this is the role of the fabric BMatthews mentions ). This is because the small vortices damp themselves out very quickly. Immediately downstream from the screen you need some kind of flow straighteners. Some kind of honeycomb-like passage is a good way to go. I don't know the ideal dimensions for the screen holes or the honeycomb. Downstream of the honeycomb, you need some open area, to allow the small vortices ( turbulence to damp out ). I don't know how long this section has to be either, but I don't think it is much more than one diameter of the duct.

As BMatthews says, wind tunnels take a lot of juice to run. If you have some kind of flow that just needs to be really smooth, you don't need much more power than you currently use. The only increase would be due to the pressure drop of the screen & straighteners. I don't know what the setup is, bit it sounds like you might be able to get away with adding the screen & honeycomb upstream of the area where you want smooth flow, with no other changes.

I may be able to find more info. in some old textbooks next week when I return from a business trip. PM me if you like.

banktoturn

You don't say what you expect to be able to achieve with the wind tunnel. All design efforts must start with a set of design objectives. Otherwise there are no rational criteria for judging alternatives in the design process.

The bigger the tunnel, the easier it will be to build test models of sufficient precision. The smaller the wind tunnel and its test models, the higher the velocities needed to achieve the reynolds numbers of interest.

At low Reynolds numbers, turbulence reduction is critical to accurate results. Noise and vibration abatement have to be considered.

Why don't you break down and say what you are trying to do? I have used tunnels of all sizes and velocities. I was the aero guy on the design of a tunnel for testing SCRAM jet engines at Mach 5 and 8 flight conditions, designed a transonic tunnel that we didn't get to build, and spent a month running tests in the 16' transonic tunnel at Tullahoma at a cost of $70k per day for tunnel time. If you can narrow the problem we can see if we know what can be done about it.

Hello Regal...

The University of Minnesota has a wind tunnel that is available for rent. I'm sure the cost per day is a lot less than the cost of building one from scratch. Air velocities can be varied from 10 mph to over 80 mph.

Test space is about 6' square.

Les Garber and I used it for our autogyro rotor tests with satisfactory results. During a 10 hour day we managed over 35 separate tests with different rotors, blades, etc. See RCM August 2001 for a summary of our results and some photos of the tunnel itself.

Bill

Hello

Here is a small wind tunnel.

http://home.earthlink.net/~savaiaa/wt.html#top

Hope this helps.

Alan

P.S. for flow straighting, a bank of soda straws works very well

WOW! thanks guys, you have helped greatly.. here it is, ill try to appease your curiousity...

I am working on designing a car stereo enclosure design for competition use.. the basica stereo consists of a small car with a reinforced cabin, and a large enclosure with a large port venting the pressure from inside the enclosure into the cabin.. currently im working with port velocities in excess of mach 0.15 .. not too fast, but with current port designs, the airflow tends to simply go all over instead of being directed where it is needed and can be most effective at pressurizing the air around the microphone being used to measure how loud the vehicle is..

the general port design being used right now is simply a wood rectangular tunnel, the length of which determines the tuning frequency..

I have heard someone suggest using Straws before, but havent tried it out yet... and the cloth idea does intrigue me... would something like a very fine mesh.. almost like panty-hose i guess work for this??

so basically what im seeing is that the best bet is to use a honeycomb or other small tubes in a 4-6:1 ratio of diameter to length.. would using this idea as well as the cloth be advantageous as well?

one other question.. im assuming that i would want to use as slick of a material as possible.. perhaps i could paint fiberglass resin over my material...

you can see im full of questions here..

thanks a ton guys, i really found the right forum! you have been a great help!

Well, I guess your doing a little cross-training from Sounddomain.com and Carsound.com! You're obviously an SPL addict! Directing the port at the mic and just "blowing" at it probably isn't going to help much. The "straws" thing is interesting...but it sounds like some sort of "band aid deal", and the change in port tuning freq from something like that would seem to lead to lots of trial and error...don't you think a smooth walled flared port of optimum dimension would better pressurize the cabin? The keyword is COUPLING. You want to acoustically couple the enclosure to the cabin. Most guys use ungodly cone area and power and a simple mathematical port volume/frequency tuning to get the cabin to resonate..(assuming you've determined the resonant frequency of the cabin itself). It's sound pressure that moves the diaphram of the mic, not port wind. You need to think about a low restriction port that will effortlessly allow the enclosure to work best...eliminating aerodynamic inefficiencies of the port itself. Usually it's the SQ guys that want to eliminate port noise, but eliminating port noise usually increases SPL also. Also, you mention a port velocity of mach .15..!? The actual pressure waves (I would assume) are actually mach 1.0...since it is sound pressure. Interesting stuff! ---Dennis

hehe.. well mostly www.caraudioforum.com and www.termpro.com .. a little bit on carsound though

well i would go the high-displacement route, but it simply isnt allowed in my class... not sure how familiar youwould be with DbDrag rules, but i wil be in Super Street 1-2 most likely.. and only 2 drivers are allowed


yeah i guess it is partialy pressurizing the cabin and partially getting the wave itself to aim directly at the mic in front of the glass in order to have it reflect directly at the mic... as far as im concerned it could be silent in the rest of the cabin, all i care about is pressurizing or "exciting" the mic..

i am curious about how you are describing coupling.. i have only heard of coupling in reference to 2 or more drivers coupling with each other... i am working with the resonant freq of the cabin already also.

thanks! looks like there is a fellow car audio nut!

Regal1975,

I don't think you need to worry about reducing turbulence or vorticity in the port flow, at least not for the reason you are concerned about. You do not need to direct the air toward the microphone, because it is the pressure waves that need to get to the microphone, and that is going to happen whether you get a stream of air going that way or not. The potential problem I see is generating undesirable noise in the flow, and this could indeed be a turbulence problem. Avoid sharp corners and you should be OK.

banktoturn

been working with a current world record holder and several other people at very high levels of competition and as of now, it seems that all of the people excelling at spl are the ones who are managing to direct the airflow directly at the mic... you guys have given me tons of great ideas so far!! :surprised