Introduction: How to Make a Wind Tunnel

Picture of How to Make a Wind Tunnel

It's always been cool to see the car commercials depicting the clean, crisp, curvature of cars (namely the Infiniti commercials).   As an aerospace/aerodynamics enthusiast, I spend my free time looking at planes and watching videos about how they work and interact with the medium they travel through, air.  For a while, I've known that planes, rockets, and even cars are designed by subjecting small scale models of them through an instrument known as a wind tunnel.  However, the wind tunnels I know of are either very big, expensive and hard to access unless you have connections or small pitiful demonstrations the science museums use show the airflow around a very limited array of  objects, often times a static display.  I've always desired to see the effects of different objects in wind tunnels for myself so when my physics teacher gave us a end-of-the-year assignment to learn about anything physics related and use what I've learned to teach the class, I thought to myself, this is a perfect moment to make a wind tunnel.  I partnered up with a peer, Ian Kelley (also interested in testing out various objects and seeing their effects) and we set out to build a wind tunnel that was simple and cool at the same time that we would be able to see the effects of air clearly on various objects.   This is a guide documenting our process and how the problem was approached.  Included are images and tips on how to make your very own tunnel cheaply in order to learn about the aerodynamic properties of objects. Hope you enjoy.
UPDATE: Be sure to check out the updated version of this project, here.

Step 1: Design & Why?

Picture of Design & Why?

To start off, we knew the basics of wind tunnel.  A air source that is compressed, increasing the speed allowing for various objects to be tested for aerodynamics and air resistance. A source of inspration for making the tunnel was another Instructable that was made previously by user Goalieguy ( be sure to check it out!).  
We also used NASA as a resource into the design of the tunnel. 

Now that the tunnel design was set, we had to come up with a way to generate smoke.  Ian thought up of the idea of using a pump and a jar and that can be seen on the bottom right side of the diagram.  Pretty smart idea I gotta say.

Now how does a wind tunnel work? Well the physics behind it are based off the work of Daniel Bernoulli, a Swiss physicist.  Many might remember him from the equation that you learned in physics back in high school of constant= P + 1/2 p v^2 + p g h.  An equation he helped contribute to, the continuity expression in physics of A1V1=A2V2.  This provides the foundation of wind tunnels by showing how if you decrease the cross sectional area, your  speed subsequently increases.  In the case of this specific tunnel that was made, the speed increased by a factor of 3.5 from leaving the fan to being compressed by the contraction section. 

Step 2: Stuff Needed to Build It and Make It Happen

Picture of Stuff Needed to Build It and Make It Happen

In order to make the wind tunnel, you need the following materials and tools:

-Toilet Paper rolls/ Used paper towel rolls
-Sheet of Lexan (Might be a tad expensive but does not shatter when cutting with a jigsaw)
-Duct Tape
-Masking Tape

-Hobby Knife
-Measuring tape/long straight edge
-Sand Paper/ Dremmel
-Hot Glue/ Waterproof adhesive 
-90 degree angle tool
-Safety Glasses

These two lists are purely to make the tunnel itself without the smoke section to make the airflow visible.  Keep in mind that some of these materials and tools are optional and not necessary to complete the project. 

Step 3: Make the Contraction Section

Picture of Make the Contraction Section

Now it is time to make the part of the tunnel that helps magnify the force of the wind by using Bernoulli's principle to increase the speed of the air that is being pumped from the fan. 

Refer to diagram in step 1 on how to properly tape the pieces together.  Remember to tape in the inside and outside to have a smooth transition between two pieces and to reinforce the part.  

Step 4: Making the Flow Straightener

Picture of Making the Flow Straightener

Gather around 20-30 tubes and arrange them in such a way that the outer ones are the largest and the inner ones the smaller ones.  Tape this arrangement together and place into the three sides of the body of the tunnel.  Once you squeezed all of the tubes inside the tunnel, tape on the 4th side and bam you are done!

Step 5: Testing Section / Exhaust

Picture of Testing Section / Exhaust

This section can be of any length, enough to incorporate your flow straighter, the hole for the gas intake, and viewing screen. 

For this section, what we did was cut out 3 parts of the 4 and for the 4th part, use a shorter piece cut that in half and have the Lexan sheet in the middle of the two. Don't forget to peel off both sides of the Lexan protector! 

The last two images show how to make one of the two stands used to stand the tunnel up since it is at an angle without this part because of the contraction section.  Just cut slots into each piece and shove them together to make the stands.

Step 6: Smoke Generator (Important Tips Included)

Picture of Smoke Generator (Important Tips Included)

This step is essential to the tunnel since you need some kind of indicator to be able to see the approximate wind flow. As to what the smoke is out of, we first tried using incense for the smoke.  This was a horrible idea since we were stuck testing with this in a dark, closed garage with airflow from a fan for 3 hours.  Incense also clogs up the tubing over time and we constantly had to clear the tubing and reignite the incense if it ran out.  The single most important tip I would give is to use dry ice and warm water if you have access to it.  The purpose of the water is to speed up the process of creating smoke since the cooler the temperature of the system, the less likely the ice wants to sublimate into CO2 gas.  It is also cleaner and doesn't clog up the tubing as much.  Only downside is that you need a water catcher under the tubing when the cold air condenses inside the tubing creating water. 
The first image is what you want to make.  Refer to the bottom right corner of the design diagram on step 1 for details. 

Basically, you will need the following to make this
-Jar w/ lid
-Hand pump of some kind
-1+ ft of Airline tubing (fish tank) 
-Hot Glue/ Water tight sealant
-Incense / Dry Ice + warm water

-Drill w/ right bit for the tubing and pump head to fit into


Step 7: Final Touches

Picture of Final Touches

This makes everything stand out and look nicer and stand out especially those that are far from the tiny viewing screen.  Attach a string of Christmas lights into the tunnel.  Since they were a pretty long sting, we tried putting one part at the top and weaving it back around the bottom.  It looks great but has a reflection when taking images to we just opted to stay with the lights at top. 

Next, adding a back background to your testing section helps give a good contrast to the smoke/gas and helps with viewing the thin wisps of smoke if you happen to be using that. 

Lastly, in order to splay the tubing ends, you get a pair of scissors, put into the tube about 3/4'' and cut that multiple times around the tube to achieve the effect. 

Step 8: Experiment!

Picture of Experiment!

These are pictures of things I used to test and what it looked like with the fan on and the smoke blowing.  I did this testing with smoke, which did not look as good as dry ice + water but I did not get pictures of those.  Try out different things and see what effects you get with them. Personally, I thought the paint scraper sponge was the coolest since it was porous and it curved the smoke upwards slightly. 

Post your tests below in the comments!  I can't wait to see what others have to try!

Step 9: Finale

Picture of Finale

I would like to thank Ian Kelley for helping me with this project and Mrs. Hester for allowing us to bring it to school and teach kids about how wind tunnel testing affects the designs of everyday modes of transportation.

Thanks for viewing this Instructable! 


klombersa made it! (author)2017-07-06

I made this project this year for my part of science fair. it turned out pretty awesome, but it just took a really long time to make. I designed it around my needs for the project (which was testing the aerodynamics of different ice hockey helmets). I just changed the diminsons a bit and added a box on the bottom of the tunnel to make it slightly easier to interchange the helmets, but other than that i used this orignal idea. Thanks so much for the insirpation!!

NassamA (author)2016-06-10

Hello. I am planning to build and control a wind tunnel for my final year project. I am interested in your design and would like to design a similar one. Do you have an operating video?

dstmartinmaker (author)2015-05-14

I think every school should have a wind tunnel. I have built a simple one for our school, and it is immensely popular!

trogfield (author)dstmartinmaker2016-04-01

what grade are you in? there are thousands of experiments all schools should have.

Osama Mohammed (author)2015-11-14

whats the use of sand paper in this project

DeepakD1 (author)2014-12-30

thanx james you are very helpful for me

thanx a lot

Neeraj Techie (author)2014-10-27

I liked this project very very much. I wanted a project for my science exhibition and I have got it. thanks.

I wish you the best of luck with your science exhibition project. I'm sure you can make your wind tunnel even better than mine by integrating some sensors! :)

AAronvb (author)2013-12-08

Very NIce.
Now I have a headache. I want to build a wind tunnel to test my VAWT. 24"x24". My question is Scaling. How much larger should the test section be than the object being tested? With that I think I can figure out the math for the rest of the tunnel.


moraguezma (author)AAronvb2014-11-03

The rule of thumb is that the frontal area of the model seen by the flow should be less than 7.5% of the cross-sectional area of the test section. So if you want to test a full size VAWT you will need a very large wind tunnel. It would be more practical to test a scale model and relate the model results to the full scale prototype using similitude and pi groups.

jamesabt007 (author)AAronvb2014-01-09

I am not sure exactly of how large you should make it, my guess is that you should just make it big enough to be able to fit the test subject into the tunnel with enough clearance for instruments and taking in and out. As a high schooler, I am still limited on my knowledge of aerodynamics since they do not teach this stuff at school. You might have a better answer if you ask one of these professionals on this incredible website. Thanks. :)

AAronvb (author)2014-01-09

Thank you James. I do love the avatar.

High School? That is AWSUME!! You can think and learn own you own!

Much respect to you!

I found some where, (1' item = 2' test area) 2x size larger than the item being tested.

Thank you, AAron

DeusXMachina (author)2013-12-02

I think Coroplast (corrugated plastic, the stuff election candidate signs are made out of) would be a great material for a flow straightener.

jamesabt007 (author)DeusXMachina2014-01-09

That is a good idea to use!
I wish I thought of that before I cut all my straws for my new tunnel... haha
Thanks for the suggestion. :)

Andsetinn (author)2013-07-27

For airflow straightener you might want to Google "Laminar Flow" but straws seem to be the most common things used for this scale.

If you mount your model on a scale you can measure lift (or downforce). Many supermarkets or electric stores sell cheap digital scales with accuracy of 0.1 oz or 1 grams. Mounting another scale behind the model and use a lever to convert horizontal force to vertical force will give you the drag of the model. The scales will not be accurate without calibration but will be good indication and are very useful for comparison between objects.

jamesabt007 (author)Andsetinn2014-01-09

For my new wind tunnel, I used an array of straws to straighten the air. As for the scale, I looked into measuring the amount of drag in the tunnel but found that to be over my $100 budget. If I obtain the resources and time to keep adding and improving to the tunnel, I will look into this because I also encountered a professional at MakerFaire that commented on doing this. Thank you.

eschelar (author)2013-08-28

Great little project.

I am going to try to build a couple myself - hopefully one large enough to take a bicycle.

I'm in Asia though and it can be difficult to get some things here.

For a larger one, say around 1.5mx1.5m for the main tunnel section, what would be the preferred diameter of the flow straighteners? I'm thinking about using some kind of plastic sheathing used for fluorescent tube bulbs or something similar, probably 1 inch in diameter or so and around 10 inches in length. I'd probably build a set of 4 fans attached to a plywood board.

Would I need to have additional fans on the intake side to keep the air moving? I was thinking about using several smaller fans just to increase airflow, but the more I think about it, the more I am questioning whether they would be helpful.

I am thinking about putting a set of fluorescent tubes at the top with a half foot recess and straight edges to make some directional light from the top, with black velvet at the back of the viewing area. That should light up the smoke. Maybe need some lights on the bottom too.


M-Theory (author)2013-06-17

Build Your Own Inexpensive Wind Tunnel‎

Andsetinn (author)M-Theory2013-07-27

This has been moved and is now at. :)

jamesabt007 (author)M-Theory2013-06-26

Haha, this would have made the initial stage a lot simpler, thanks. I hope this helps other out there who look forward to making their own.

aspectr (author)2013-06-16

Nice build! If I may, there are a few things you could do to significantly improve the quality of airflow inside the tunnel:

1) Put the fan at the back. There's no reason to force the air in the front of the tunnel and create a huge amount of turbulence when you can draw in ambient air smoothly at the front and then apply all the fan turbulence to the exhaust air at the back.

2) Use a better flow straightener. Someone already mentioned the box of straws which should work pretty well. I know this is a "as much DIY as possible" build, but if you google "honeycomb flow straightener" you can buy a sheet of the stuff used in actual commercial wind tunnels for like $20. I bought some from a place called saxxon pc, worked pretty well.

3) Move the lights to outside of the flow section somehow. I know this is a hard one with a cardboard build but it's worth mentioning.

Finally, the smoke setup you have is pretty clever. Might I suggest experimenting with some UV dye and a blacklight? UV dye is often used in medium-size wind tunnels. Perhaps there is a way to produce smoke or vapor that is blacklight responsive? Good flow visualization is probably the hardest part of running a wind tunnel...I've personally tried a few different techniques including unsuccessfully trying to use a hydrogen bubble generator and know how annoying it is to get those great photos/videos.

Nice job!

jamesabt007 (author)aspectr2013-07-03

Thanks for the help, I've read the other comments relating to 1 & 2 but I was thinking for something like 3, would it be better to integrate the lights into the cardboard? Just and idea. Thanks again.

Cronos51101 (author)2013-06-16

If I might make two suggestions. A great way to visualize sections of your smoke trail is to use a laser "line". I've seen this used on large scale wind tunnels where they project a flat laser through the smoke so that you can pick out cross sections. You can get a little laser line diode from Amazon for about $7.
If you seem to have trouble with boundary effects causing an irregular velocity profile across the tunnel cross section (faster in the center and slow near the walls.) you can use some window screen to normalize the velocity profile. It causes a pressure drop so depending on your fan, you may see a loss in chamber velocity, but it will improve the effective area of your chamber.

jamesabt007 (author)Cronos511012013-07-03

When you say pick out cross sections, what do you mean exactly?
Also, the window screen would go in the tunnel as a layer correct? Would this be before the straighteners?

guitarpicker7 (author)2013-06-16


A suggestion to vary the airspeed is to insert an adjustable vane before the fan- note that it should be fairly sturdy.

May I also suggest a Halloween-style fog machine as a source for "smoke". If you use one with an adjustable output, you could vary the amount of fog as the air speed changes. I make my own Fog Juice with grocery store distilled water and pharmacy glycerin.

Keep on creating!

With this particular fan, I can adjust the airspeed fairly well. How do I make my own fog juice? Sounds interesting. What would I do with those materials?

MTJimL (author)2013-06-16

Just a couple of thoughts: You'll need ground effect for ground vehicles, so elevate them on a level piece of cardboard. Second, If you move the light tubes to the near wall, over and under the window, you won't have to look at the lights themselves.

Maybe next time you can build one like the NASA two-stage wind tunnel. Great work!

jamesabt007 (author)MTJimL2013-07-03

That is a really good point about the ground effects, I was wondering how to demonstrate down force to my peers, thanks! As for the lights we might implement them next time if we make an improve one like the NASA ones.

nalipaz (author)2013-06-17

Might want to caution against the amount of pressure from dry ice and water inside of a glass container as well. Possible you could crack the glass without adequate ventilation.

jamesabt007 (author)nalipaz2013-07-03

Thanks for the precaution. We hear some slight cracking but took it out before to stop the sound.

saberjockey (author)2013-06-17

You can use plex instead of the lexan. You will need a circular saw with a fine tooth blade. Run a strip of painters masking tape on the area to be cut. A layer of duct tape on the bottom of the saw will minimize scratches to the plex surface. Also, plex can be scoured a quarter of it's thickness with a utility knife and snapped just like glass. This method might require some mild touch ups with sand paper as the snapped edge may be a bit jagged.

jamesabt007 (author)saberjockey2013-07-03

What exactly is plex? Is it cheap? Also what sets it apart from lexan? And I don't have a circular saw nor and idea of how to obtain one. Thanks.

doc_cls (author)2013-06-17

I built several wind tunnels when I was young - long before I joined NASA - and I have some suggestions. First, so-called "open return" tunnels *always* have the fan in the rear, pulling the air through the tunnel. This greatly improves the flow quality as well as the efficiency. The overall layout should be a bellmouth leading into the test section, followed by a long expansion section (the "diffuser") leading to the fan. The total angle of the diffuser walls should be no more than about 10-15 degrees, and you should have flow straighteners at both the beginning and end of the test section; indeed, soda straws work very well.
Good luck!

jamesabt007 (author)doc_cls2013-07-03

I got to say, it really blows my mind the awesome people you have the opportunity to meet on this site... So if I were to do it as you said (this is what I understood it as) it would be like the current design except its swapped around and bell shaped with the inlet wider than the testing section?

redrok (author)2013-06-16

Hi jamesabt007;

The standard "Rule of Thumb" for creating a laminar flow with a flow straightener is make the tubes length at least 10 times the diameter. 20 times is even better.
As others have said plastic soda straws work nicely.


jamesabt007 (author)redrok2013-07-03

Oh, this answered my previous question, thanks!

crjeea (author)2013-06-16

The best idea I saw for the tubes that produce the (almost) laminar flow was instead of using toilet roll tubes to use drinking straws. several packs of 250 all stuck together.
As far as the smoke is concerned that is a nice solution. I would suggest trying a small coil of nichrome wire with a blob of blue-tack wrapped around it. You pass a small current (enough to heat the coil) through the wire and it will produce a nice thick white-ish smoke.
The best place to buy large quantity's of nichrome wire is to buy a reel of electric fence wire. Cut it to length and in thread the nichrome (Some are six threads in parallel extra and some have copper or aluminium even a mixture of mettles, so read before you buy)

Tip of the day... The fastest way to strip enamelled copper wire is to light the end with a lighter.

jamesabt007 (author)crjeea2013-07-03

I really like the idea of using straws but if I used them how long should they be still? Same as before? Also we found dry ice to be a lot easier so we will stick to that, thanks.

jongscx (author)2013-06-15

I could suggest you use an electronic cigarette or fog machine to create your 'smoke' trail instead of burning something and creating actual smoke. The only issue here is that they produce actual vapor which could condense on surfaces and weaken the cardboard structure.

jamesabt007 (author)jongscx2013-07-03

I actually looked into the e-cig and found that if I used it, my parents would kill me thinking I was smoking it. Thanks for the idea though.

raygris (author)2013-06-17

Nice project. I have a few suggestions based on my experience with my wind tunnel I built as a Jr High science project. ( almost 50 years ago) A simple manometer can give you airspeed at the test section. I like your tube nest to smooth the air flow. Another reader suggested straws.They should be pretty short or you will have restricted flow due to boundary layer build-up. I think that's about 7 degrees, but I'm going from memory. If your test section has a small cross section compared to your test model, you might improve flow by flaring the test section at 7 degrees,( also to compensate for boundary layer build-up). An elliptical 's' curve to your nozzle will help keep your flow laminar also. With cardboard that should be pretty easy. It's not that hard with sheet metal. The reader that suggested a raised platform for automobile models is right. Ground effect is important. I used very fine fibers glued to the models skin to view airflow. I'm not sure if the smoke is better or not, but I;m sure it's easier!

jamesabt007 (author)raygris2013-06-26

Thank you for your input. It took me a few minutes and a few searches to fully comprehend what you were talking about since I am not really as experienced as I wish to be in regards to the terminology. I will be sure to share this with my partner and see if he is interested in making a better, more professional wind tunnel, tweaking things here and there to take in reader's suggestions. Thanks again for the help! :)

DustySeven7 (author)2013-06-18

An idea is a small fog machine. Most have a speed control for the amount of fog as well.

jamesabt007 (author)DustySeven72013-06-26

I agree with you, a proper fog machine is needed as well as a speed control, thanks for the great idea!

grazer (author)2013-06-20

How much was the fan to buy, as I need to make one for a Physics project and I have no idea where to get one or how much to pay... Help :(

jamesabt007 (author)grazer2013-06-26

I found that it being summer now, a lot of stores now sell cheap fans (places such as CVS, Walgreens, etc.) They can go from anywhere from 10-30. It is also not necessary to have a boxed shaped fan for the project because you can also use a circular framed fan and create a box shaped cardboard with a circle inside to fit the fan through.

bertus52x11 (author)2013-06-13

Very nice one! I am curious what you are planning to test.

jamesabt007 (author)bertus52x112013-06-13

Thank you very much! I'm asking my teacher to email the photos of our dry-ice tests to post online. :)

Hawkeye_bkj (author)2013-06-12

That is a very nice tunnel. I would like to suggest that if you use smaller diameter tubes as your straightener it will produce better results.

Once/if you go through Fluid dynamics, you will learn about the ratio called Reynolds number which is found with this equation, Re=  ρvD/μ, where ρ is the density of the fluid (air in our case), v is the velocity, D is the diameter of the pipe the fluid is going through, and µ is the dynamic viscosity. In your situation, you can control the velocity or the diameter of the tubes you are flowing through. Ideally for laminar flow, which is a smooth flow, you want your ratio to be below 2300. Of course you don't want to slow the air down, so reducing the diameter is your best option.

A cheap way of doing that would be to use boxes of straws. They retain their shape very well, are of uniform length and have low surface friction... I can't claim that idea because I saw it on Mythbusters, but the science is sound.

jamesabt007 (author)Hawkeye_bkj2013-06-13

I understand what you mean. Since most of the stuff I know about fluids is self taught through random youtube videos, I have a pretty limited knowledge. It does make sense however that using straws would help the flow since with such a small diameter would further restrict any stray currents. Thanks for taking time to explain the concept behind why, I've learned something new! I will try to implement this into my next model if I make one. :)

About This Instructable




Bio: In my free time, I enjoy building/modifying/dismantling anything I can get my hands on. I am currently an engineering student at UCLA and ... More »
More by jamesabt007:DIY Wind Tunnel 2.0, Project "Paperclip"Simple Suspension Bridge ModelEvolution of the Laptop Stand
Add instructable to: