wind Tunnel - RC Groups
Thread Tools
Nov 10, 2011, 08:13 PM
Registered User
Flying Penguins's Avatar

wind Tunnel

So I'm making a wind tunnel for my grade 11 science fair project at school this year, I figured it was the perfect project since I already like planes. It's gonna be small and as basic as possible. I've started the build probably too early without planning... so far I have, as the wind part, 2 5 inch computer fans pushing (not pulling I know) air. I'm testing 5 different airfoils at 0, 10, 20, 30 degree angles of attack, I have solved the about the project circular flow from the fans by cutting 1 cm diameter straw into 5 cm sections and glueing them together.
So what 5 airfoils do you reccomend I test (preferably real ones)?
What are the implications of pushing and not puling it?
What other advice do you have for me and what questionsdo you have about the project?
Thanks in advance!
Sign up now
to remove ads between posts
Nov 10, 2011, 10:07 PM
B for Bruce
BMatthews's Avatar
With a drinking straw flow evener and straightener you SHOULD be fine provided you have a little bit of a gap between the fans and the wall of straw segments. It will be far from perfect for truly sensitive scientific comparisons between airfoils but it'll be good for showing the basics.

Model size airfoils stall at a far smaller angle than full size airfoils. So you really need to narrow down the range of motion for angle of attack. At model size and speeds such as I would expect to find in a two fan tunnel like yours the airfoils will stall at around 6 to 7 degrees. So going much past 15 degrees is pretty much a waste of time. Instead you need to zero in on the -5 to +10 zone with more detail.

At the same time you'll want to arrange some form of scale that will show the amount of lift from the airfoil. Or perhaps a rod with some very light threads tied to it at even spacings that will serve as flow visualizers. Like smoke trails but less messy.

Good experiments would be to show how camber on it's own is able to produce lift compared to a symetrical zero camber airfoil. Another would be to show how flaps alter the airflow and lift. And yet another could be a flat plate vs 15% thick airfoil if you can get the thread flow visualizers to work to show how a flat plate stalls at a very low angle of attack while a thick airfoil resists a stall.

Without a lot of fancy measuring gear you really won't be able to detect much more than such rather coarse differences. So it's best to stick with such basics.
Nov 10, 2011, 10:14 PM
Registered User
Flying Penguins's Avatar
Sorry about that, I forgot to mention that I have already developed a working smoke system with mineral oil and nichrome wire that works really well. I have planned for the scale that you mentioned too, I need a way to measure by dependent variable quantitatively so I am connecting fishing line to the bottom of the wing which I will run through pulleys and attach to a lightweight tray, I can then add weight on the tray to see the lifting capacity of the airfoil.

Another question I had was, what wingspan segment to chord ratio should I go with.
Example a square shaped test airfoil when looking from the top, or a long rectangular one? which would produce more lift or make lift easier to measure?
Nov 10, 2011, 10:26 PM
Registered User
Hi Penguins, sounds like a fun project. We had a similar thing when I was in high school, but the school had a commercially built wind tunnel fo rour playing.

With your home made tunnel, are the computer fans fairly powerful?? Some of them do not push a lot of air, but others do. As long as you have nice strong fans it should work nicely.

As far as the size of the wing sections, I am not sure. As for one of your original questions regarding which airfoils to test, here are a couple of quick ones off the top of my head.

Clark Y
NACA 24xx series, say a 2410 perhaps.
A symmetrical NACA section, a NACA0010 perhaps
Maybe on of the Selig series, for example a S3021.

Just some suggestions. Here is a link for you with loads of airfoils to choose from.
Nov 10, 2011, 11:13 PM
Registered User
A wise man once told me, "An hour in the Lab will save you five minutes in the Library." Therein lays the true beauty of science and engineering, the ability to predict the behavior of natural occurrences. I would start with the fixed parameters and calculate values for the variables from there. Go online and find the rated volume flow rate of the computer fans, and then perform a back of the envelope velocity calculation using the following relation:
Velocity = Volume Flow Rate / Test Section Area
This isn’t absolute because you’re not accounting for the losses through your straw section, viscous losses through the section, boundary layer effects, or reduction in effective area due to the model, but it is close enough for your purposes.
Once you get an idea of what the free stream velocity is, you can estimate what magnitude of lift you’ll be trying to measure using the following equation.
Where L is the measured lift, CL is the lift coefficient, rho is the air density, U is the free stream velocity, and A is the cross sectional area. In this case you’re now limited by the resolution of your measurement, in other words, the minimum amount of weight you can accurately measure with your fishing line pulley set up. You could use something like small ball bearings as a means of determining your minimum resolution. So now the relation becomes:
L = Weight of Ball Bearing
Cl = Minimum measureable change in lift (resolution of measurement)
Rho = Density (FIXED)
U = Free Stream Velocity (FIXED)
A = Width (FIXED) X Maximum Chord Thickness, note that the Chord thickness is generally given as a percentage of the chord.
Now you have some Parameters to play around with. Another important quantity derived from free stream velocity and chord thickness is the Reynolds Number which is given by
Re = Rho*U*L/mu
Where Rho is the density of air, U is the free stream velocity, L is a distance in this case the chord thickness, and Mu is the viscosity of air. From my experience this is the single greatest dimensionless parameter in all fluid mechanics. It allows you to compare your results to all others at the same Reynolds number independent of size.
Once you get your airfoil sized, I would suggest the first model you test be the NACA 0015 section. This is a classic, symmetric airfoil from which you can learn a lot. Its behavior is well documented and it will allow you to verify the results you’re getting in your tunnel. You’ll learn where the other airfoils are likely to stall, and the maximum coefficient of lift you’re likely to encounter. The 15 percent chord thickness will hopefully allow the flow to remain attached and delay stall to higher angles of attack at the low Reynolds number of your tunnel.
For the other four sections I would recommend you steer clear of the literature initially and dream up your own wing profiles and see how they perform. Form a hypothesis, then test it in the tunnel, then report your findings back here on the forum. Make your own original contributions to the field of aerodynamics.
Here is a link to the lift curve for the NACA 0015 section:
Here is a great piece on small wind tunnel design:
Nov 11, 2011, 12:33 AM
Grad student in aeronautics
Before you can do any experiments you have to validate your setup. This includes the tunnel as well as the intruments used for taking measurements. I would recommend the NACA 0012, seeing as it's the most tested airfoil in history. Here is a link to a 1987 paper where the author examined wind tunnel data for the 0012 from over 40 different wind tunnels and compared them. It is on the NASA Technical Report Server so it is free to the public. This is critical before you do any experiments.

The key to accurate results is a smooth and uniform flow. This can be as much of an art as it is a science. I've never built a wind tunnel, but I've been told the key to achieving uniform flow in the test section is constriction. I'm not sure if I undestand the size of your wind tunnel, so this might not be feasible for you.
Nov 11, 2011, 12:49 AM
B for Bruce
BMatthews's Avatar
One thing that you need to consider for the size of the test airfoils is the height of the test area. To avoid interference from the floor or ceiling of the test area the airfoil should have one chord width of clearance both above and below the airfoil. Any less and you won't get readings that make any sense. And some would say that even as little as one chord width of clearance is too little.
Nov 11, 2011, 11:31 PM
Registered User
Flying Penguins's Avatar
Sefco, thanks for that great information, I'm currently working my way through martin Simon's model aircraft aerodynamics and this stuff is right out of the book but I didn't think of applying the way you did so thank you..
I can start post
Thanks for the height tip too BMatthews, I have some room there so that can be easily done.
I'll start posting pictures soon of my progress
Nov 12, 2011, 08:13 AM
ciurpita's Avatar
i think you'll find that the slope of the lift curve (change in lift per change in angle) is the same for all airfoils. There will be more variation in drag with different airfoils and airspeed (Reynolds number).

you might also consider a flat plate in addition to more conventional airfoils.

i'm not sure how much air a couple 5" computer fan can move and how much control you'll have. i think most wind tunnels use fairly large fans and funnel the space down to a smaller test area which also increases the airspeed. have you considered using a larger box fan. If the box fan produces too much air, you can just add a vent to let some escape.

You may want to measure the airspeed in the test space. Dwyer makes an inexpensive wind meter.

the flow straightening only needs to be in the smaller space and that space could be more rectangular if convenient. I think larger wings will be easier to build and tolerances can be larger.
Nov 12, 2011, 06:39 PM
Registered User
It really would be beneficial if you can relocate the fan(s) to be pulling rather than pushing. You won't need nearly as much flow straightening, and the air will be much less turbulent. Turbulence levels have substantial affect on the results.

Nov 12, 2011, 09:57 PM
B for Bruce
BMatthews's Avatar
Another straightener and pressure equalizer to avoid "fast spots" across the area of the test zone is to put something like a panel of fly screen just downstream after the straightener cell layer. It creates a small degree of back pressure but by doing so it makes the velocity coming out the screen more even from edge to edge and top to bottom.

Push or pull it's still a good idea to have the straightener cell "core" as well as something like the screen for evening. Also the smaller "micro turbulence" from the wires of the screed will damp out much sooner than tubulence from the drinking straw straightener core.

For the screen to do its job you want to ensure there's two or three inches of room between the drinking straw core and the screen.

The proof of the pudding will be found when you measure the air velocity you get in the tunnel. Until then there's not much point in panicing. If it turns out to be suitable for only testing indoor model slow flying sections THEN look at some hunkier fans.
Nov 15, 2011, 08:19 PM
Registered User
Flying Penguins's Avatar
ok, so i switched the motor/ fan units around on the weekend, took a lot of work and some destruction, but hot glue is my friend
I tested it then by basically quickly covering the sides and top with just some plastic and books and foam board, and I was quite successful! with just the straw filter south of the test airfoil (which was a flat bottom one that i just sketched) the wing segment lifted itself the entire height of the tunnel
i'll post pictures later tonight, my next question however is can you guys think of any systems that allow me to adjust the angle of attack without taking up much width of the wind tunnel
right now i have planned 4 dowels verticially, one in each corner of the wing segment that guide the wing vertically, i need a way to adjust the angle of attack with taking up minimal width in the tunnel because it is only 5 inches wide
Thank everyone!
Nov 16, 2011, 02:17 AM
B for Bruce
BMatthews's Avatar
Normally you don't want the wing to move. Instead you want it to stay in the middle of the test area and conduct the lift force to a measurment scale. With that in mind you should be able to come up with some sort of fairly flat "plates" which extend from a lift measurement scale below or above the test area done through the "floor" or "ceiling" and that have a round disc portion which the airfoil attaches to. This sort of arrangement would also avoid a gap between the end of the airfoil and the tunnel walls. A gap which would create a lot of leakage, drag and break the proper 2D flow that you want to maintain.
Nov 20, 2011, 10:55 PM
Registered User
Flying Penguins's Avatar
Here are some of the photos so far. I think I have decided on 3 airfoils, the Clark Y, NACA 0015 and a NACA from the 24xx series. Any other suggestions would be great
Nov 21, 2011, 07:51 AM
Registered User
Flying Penguins's Avatar
Now that I have switched the fans so they suck air through the tunnel, where should I put the fly screen?

Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion Wind Tunnel F/S, not mine, but someone may have a use Stalled Modeling Science 1 Dec 30, 2011 12:35 AM
Discussion Has anyone attempted to build a wind tunnel to measure drag coefficient of models? MRGTX Foamies (Kits) 1 Sep 27, 2011 10:23 AM
Discussion PA Shock & Wind tunnel future test. Ari33 Electric Plane Talk 0 Aug 31, 2011 01:52 PM
Discussion Propeller wind tunnel tests yomgui Power Systems 35 Jul 30, 2011 06:35 PM