A tornado in a box is one of the most visually striking science-fair projects you can build, and the basic version requires little more than a cardboard box, a small computer fan, and a source of fog or smoke. The idea is simple: create a column of rising air inside an enclosed space, then introduce rotation so the rising air forms a visible spinning vortex. The whole build can take an afternoon, costs under twenty dollars with salvaged parts, and demonstrates the same fluid-dynamics principles that researchers use in professional laboratory vortex chambers.
What You Need
The materials list is short, and most items can be found around the house or picked up at a hardware store. You are building three things: an enclosure, something to pull air upward, and something to make the airflow visible.
- Enclosure: A cardboard box roughly 18 to 24 inches tall works well. Taller boxes give the vortex more room to develop, but anything from a shoebox standing on end to a large moving box will produce results. Clear acrylic or plexiglass panels are ideal if you want the vortex visible from every angle, but you can also cut a large viewing window in one side of a cardboard box and tape a sheet of clear plastic over it.
- Fan: A small 12-volt DC computer fan (the kind that cools a desktop PC) mounted in a hole cut in the top of the box. This pulls air upward and out, creating the updraft. A 120mm fan is a common choice because it moves enough air without being overpowered.
- Power source: A 9-volt battery or a small 12-volt power adapter to run the fan. A 9-volt battery will spin most 12-volt fans at slightly reduced speed, which actually works fine for a gentler, more stable vortex.
- Fog or smoke source: Incense sticks are the classic option. A small piece of dry ice in warm water produces thicker fog. Fog machines designed for Halloween decorations also work if you have access to one.
- Guide vanes (optional but recommended): Small pieces of cardboard or plastic angled at the base of the box to direct incoming air into a spin. These are the secret ingredient that turns a simple updraft into a rotating vortex.
You will also need a box cutter or utility knife, tape or hot glue, and a ruler. If you are using dry ice for fog, grab a pair of thick gloves and safety glasses. The same precautions used in other hands-on science projects apply: gloves protect against cold burns from dry ice, and eye protection is sensible whenever you are cutting materials or handling chemicals.
Building the Box Step by Step
Start with the enclosure. If you are using a cardboard box, stand it upright so the opening faces up. Cut a circular hole in the top panel sized to fit your fan snugly. The fan should sit flat over the hole, pulling air up and out of the box. Secure it with tape or hot glue, making sure the airflow direction is correct: most PC fans have a small arrow printed on the frame showing which way air moves. You want air flowing out through the top.
Next, cut air inlets near the base of the box. This is where outside air enters to replace what the fan pulls out. You have two options, and they produce noticeably different results. The simpler approach is to cut a few rectangular slots (about two inches wide and four inches tall) evenly spaced around the bottom edge of the box. Air rushes in through these slots, rises toward the fan, and if conditions are right, you get a loose column of swirling air. The better approach is to add guide vanes.
Guide vanes are small flaps of cardboard glued at an angle just inside each inlet slot. Instead of letting air flow straight into the box, the vanes deflect incoming air so it enters at an angle, all spinning in the same direction. Think of it like a pinwheel: every slot pushes air clockwise (or counterclockwise, just be consistent). This tangential entry is what gives the airflow its rotation. Without vanes, you get a wobbly, inconsistent column. With them, you get a tight, well-defined vortex.
To make the vanes, cut small rectangular tabs of cardboard about two inches wide. Glue each one at roughly a 30- to 45-degree angle to the inner wall of the box, right next to each inlet slot. All vanes should angle the same way. Four to six inlet slots with matching vanes, evenly spaced around the base, is usually enough.
Once the box is assembled, place your fog source at the bottom center. If you are using incense, light a stick and set it on a small fireproof dish. If you are using dry ice, place a small chunk in a shallow cup of warm water. Turn on the fan, wait a few seconds for airflow to stabilize, and you should see the fog gather into a spinning column that rises toward the fan. Adjusting the fan speed (if you have a variable power supply) or partially covering some inlet slots lets you fine-tune the vortex width and stability.
Why It Works
The tornado in your box forms through the same basic mechanism that drives vortices in the atmosphere. Two ingredients combine: an updraft and rotation. The fan creates the updraft by pulling air out through the top, which forces replacement air to rush in through the bottom inlets. The guide vanes give that incoming air angular momentum, meaning it enters the box already spinning. As the spinning air converges toward the center and gets drawn upward, it spins faster, the same way a figure skater spins faster by pulling their arms in. That acceleration is driven by conservation of angular momentum, and it is exactly what happens in a real tornado when horizontally rotating air gets tilted into a vertical column by a thunderstorm’s updraft.
Laboratory tornado simulators used by researchers operate on the same principle, just at a larger scale and with more precise control. Ward’s laboratory vortex chamber, developed in the 1970s, remains one of the most referenced designs because the vortices it produces share dynamical and geometric similarity with the range of real tornadoes observed in nature.1Atmospheric Research. A review of supercell and tornado dynamics Your cardboard box is a simplified version of this kind of apparatus. The core physics is the same, even if the scale and precision differ enormously.
Getting a Tighter, More Dramatic Vortex
If your first attempt produces a fat, lazy swirl rather than a tight funnel, a few adjustments can sharpen it considerably. The single biggest factor is the ratio between the updraft speed and the rotational speed of incoming air. Researchers who study laboratory vortices call this the swirl ratio. When rotation is weak relative to the updraft, you get a broad, disorganized column. When rotation is strong, the vortex tightens into a narrow, fast-spinning funnel.
In numerical comparisons of laboratory and atmospheric tornado vortices, increasing the swirl velocity of incoming air produces dramatically more intense vortices. Even with friction from the lower boundary slowing things down, a strong imposed swirl can generate vortex wind speeds that exceed what a frictionless setup produces, because the interaction between the surface and the rotating air creates a concentrated inflow jet near the base that feeds angular momentum into the core.2Quarterly Journal of the Royal Meteorological Society. A comparative study of atmospheric and laboratory‐analogue numerical tornado‐vortex models You will not be measuring wind speeds at 55 meters per second in your cardboard box, but the same physics applies at your scale: more rotation equals a tighter funnel.
To increase rotation in your box, angle the guide vanes more sharply (closer to 45 degrees rather than 30). You can also narrow the inlet slots slightly, which forces the same volume of air through a smaller opening and increases its speed. If you have a variable-speed fan or a dimmer switch wired to the power supply, try turning the fan speed down a little. Counterintuitively, a slightly slower updraft with strong rotation often produces a more visible and stable vortex than a fast updraft that yanks the fog out of the box before it has time to spin up.
Surface texture inside the box matters too. Researchers studying real boundary-layer vortices have found that surface roughness modulates vortex development by affecting how air converges and stretches near the ground.3Quarterly Journal of the Royal Meteorological Society. Dynamics of boundary layer and lower‐level vortices east of the second‐step terrain over the Yangtze River Valley In your box, a smooth floor lets air glide inward freely, while a rough or textured floor creates a thin layer of slower-moving air along the bottom that can actually help concentrate the vortex. Experiment with placing a piece of sandpaper or a rough mat on the floor of the box and see whether the funnel changes shape.
Choosing and Managing Your Fog Source
The vortex exists whether or not you can see it. Fog is just the tracer that makes the invisible airflow visible. Different fog sources have different trade-offs, and the choice affects how long your demonstration lasts and how impressive it looks.
Incense is the easiest option. It produces a thin, wispy smoke that shows the vortex shape clearly and works well in a small box. The downsides are that incense smoke is not very dense, so the vortex can be hard to see in a brightly lit room, and the smell can be distracting during a science-fair presentation. One incense stick typically burns for 20 to 30 minutes, which is more than enough time.
Dry ice in warm water produces a much denser, more dramatic fog. It hugs the floor of the box (since CO₂ is heavier than air) and gets swept up into the vortex beautifully. The visual impact is far greater than incense. However, dry ice requires careful handling. Wear insulated gloves whenever you touch it, and do not seal it inside an airtight container. Use it in a well-ventilated area, since CO₂ displacing oxygen in a small enclosed space is a real concern. For a science project being demonstrated in a gymnasium or classroom, dry ice is usually fine, but check with your teacher first. Safety protocols for hands-on science projects involving dry ice call for gloves and eye protection at minimum.4Physics Education. Cloud chamber kit for active learning in a first-year undergraduate nuclear science seminar class
A small fog machine is the most convenient option if you already own one. It produces consistent, thick fog on demand, and you can refog the box repeatedly without relighting anything. The glycol-based fog from these machines is nontoxic in the small quantities you will use. The main drawback is size: some fog machines blast out more vapor than a small box can handle, overwhelming the vortex rather than tracing it. If you go this route, use short, gentle bursts.
When the Vortex Breaks Down
Sometimes the vortex forms beautifully for a few seconds and then collapses into a disorganized mess. This is not a failure of your build. It is actually a well-studied phenomenon in fluid dynamics called vortex breakdown. In laboratory experiments on swirling flows, researchers have observed that when the upward momentum pushing air through the vortex drops below a critical threshold, a bubble of reversed flow forms on the axis of swirl, and the clean funnel falls apart.5Journal of Fluid Mechanics. Continuation or breakdown in tornado-like vortices
In practical terms for your box, this means the vortex is sensitive to the balance between updraft strength and rotation. If you dump too much fog in at once, the cool, heavy fog can slow the updraft enough to trigger breakdown. If the fan speed fluctuates (common with a dying battery), the vortex destabilizes. Keeping the fan speed steady and feeding fog gradually rather than all at once helps maintain a stable vortex for longer.
Vortex breakdown is actually a great talking point for a science-fair presentation. If a judge asks what happens when the tornado falls apart, you can explain that it is the same instability that limits real tornadoes. When the updraft feeding a tornado weakens, the funnel dissipates. Your box lets you demonstrate that process in real time by simply turning the fan speed down.
Upgrade Ideas for an Advanced Project
The basic cardboard-and-fan version is solid for elementary and middle school projects. If you want to push into high school territory or simply make a more impressive demo, several upgrades are worth considering.
First, swap the cardboard for clear acrylic panels. A fully transparent box lets viewers see the vortex from every angle and eliminates the need for a viewing window. Acrylic sheets can be cut to size at most hardware stores and joined with acrylic cement or even hot glue for a temporary build.
Second, add a variable-speed control to the fan. A simple potentiometer (a dial that adjusts voltage) wired between the power supply and the fan lets you change the updraft speed in real time. This turns the project from a static demonstration into an interactive experiment: you can show how different updraft speeds affect vortex diameter, stability, and the point at which breakdown occurs. Potentiometers rated for small DC fans cost a couple of dollars at electronics stores.
Third, try different guide-vane angles and document the results. Set the vanes at 15 degrees, 30 degrees, and 45 degrees, photograph or video the vortex at each setting, and measure the approximate diameter of the funnel. This gives you actual experimental data to present, which elevates the project from a demonstration to a genuine experiment with variables and results.
Fourth, add a heated base. Real dust devils and some tornado-like vortices in the atmosphere form partly because of temperature differences between the ground and the air above. Laboratory models of dust devil vortices have explored how varying boundary heat input changes the vortex’s core structure, producing different categories of vortex behavior depending on the combination of heating and rotation.6Elsevier / ScienceDirect. A laboratory model of dust devil vortices You can approximate this by placing a small, safe heat source (like a ceramic mug warmer) at the bottom of your box. The warm air naturally rises, supplementing the fan-driven updraft. Combined with fog, this can produce a more vigorous and self-sustaining-looking vortex. Be cautious with heat sources around cardboard, though; an acrylic enclosure is much safer for heated experiments.
How Your Box Tornado Compares to the Real Thing
A real tornado forms when a rotating updraft inside a supercell thunderstorm stretches a column of spinning air from cloud base to the ground. The energy source is the atmosphere itself: temperature differences, moisture, and wind shear over tens of kilometers conspire to concentrate rotation into a tight funnel. Your box substitutes a fan for the thunderstorm’s updraft and guide vanes for the atmospheric wind shear. The scale is different by a factor of thousands, but the fluid-dynamics principles are genuinely analogous. This is not just a metaphor; it is why atmospheric scientists use laboratory vortex chambers that are functionally bigger, fancier versions of your project.
That said, real tornadoes do things your box cannot replicate. They move across the landscape, interacting with terrain and buildings. They can produce multiple sub-vortices spinning within the main funnel. They generate pressure drops that your box’s gentle fan cannot approach. And the energy budget of a severe thunderstorm is staggeringly larger than your 12-volt fan can deliver. Your project is a demonstration of the basic vortex mechanism, not a scale model of a tornado’s full complexity. Being upfront about that distinction actually strengthens a science-fair presentation, because it shows you understand the limits of your model.
Presenting the Project at a Science Fair
A tornado box is a crowd-pleaser because it is visual and interactive, but judges tend to reward understanding over spectacle. A few strategies help you get credit for both. Keep the box running during judging if possible. A live vortex is far more compelling than a poster describing one. Bring extra incense or dry ice so you can restart the fog if it runs out.
Prepare to explain the role of each component. The fan creates the updraft. The guide vanes introduce rotation. The fog makes the invisible airflow visible. If you added a variable-speed control, demonstrate it and explain what changes as the speed goes up or down. If you tested different vane angles, bring your data. Judges at every level respond well to projects where the student actually varied something and observed the result, rather than just building a single configuration and describing it.
Anticipate the toughest question you are likely to get: “How is this different from a real tornado?” The honest answer, that it captures the vortex mechanism but not the energy source, the scale, or the complexity, shows genuine understanding. You can add that professional researchers have used the same basic chamber design and found that the vortices it produces span a range similar to what is observed in nature, which is a satisfying fact to share with a judge who wants to know whether your little box has any real scientific relevance.
Common Mistakes and How to Avoid Them
The most frequent reason a tornado box fails to produce a visible vortex is air leaks. If the box has gaps or unsealed seams, air sneaks in through cracks instead of through your carefully angled inlet slots. The fan still runs, fog still swirls around, but there is no organized vortex because the airflow bypasses the guide vanes. Seal every seam with tape before testing. If the vortex is weak, hold a lit incense stick near each seam and watch where the smoke gets pulled in; those are your leaks.
Another common issue is putting the fog source too close to the inlet slots. When fog enters the box through the inlets rather than rising from the center, it gets pulled sideways before it can join the vortex. Place the fog source as close to the center of the floor as you can. For incense, a small dish in the dead center works. For dry ice, a cup of warm water centered on the floor is ideal.
Finally, do not make the box too short. A vortex needs vertical space to develop. If the box is only six or eight inches tall, the fan is so close to the base that it sucks fog straight up before rotation has time to organize it. Aim for at least 18 inches of height, and taller is generally better up to about three feet, beyond which a single small fan may not move enough air to sustain the updraft.