Free wind tunnel lesson plans for science, physics and design.

Year level

Years 7–9

Subject

Science: forces and motion

Time

50–60 minutes

WindCanvas

Free, 2D, Beginner mode

Setup (2 minutes)

1. Open windcanvas.com/app/. On the first visit choose Skip in the Welcome window. A car appears in a running tunnel.
2. Check that the top bar shows Beginner and 2D.
3. Point out Drag at the left of the status bar along the bottom: the drag coefficient, Cd, averaged as the flow runs.
4. Laptops on battery: under Power in the top bar choose Balanced.

Step by step

1. 0–5 min

   Hook: what do fast things have in common?

   Show the pictures. Ask what the shapes share and collect words on the board: smooth, rounded, pointed, tapered, long.

2. 5–10 min

   Predict

   On the worksheet (Part A) students see three shapes of the same height in the same wind: a flat plate facing the wind, a circle and a teardrop. They rank them from most to least drag and write a reason.

3. 10–15 min

   Demonstrate

   On the projector, open Presets in the panel menu on the right, click the Fundamentals chip and click Flat plate ⊥ (normal). In Beginner mode a click replaces the scene. Point out the smoke, the colours (air speed: blue is slow, red is fast), the Drag reading and the fidelity word beside it. The Explain tab describes the flow in plain English.

4. 15–35 min

   Observe

   1. In Presets → Fundamentals, load Flat plate ⊥ (normal). Watch the smoke and the swirling wake behind the plate.
   2. Wait a minute or so, until the Drag number stops drifting, then record it. It wobbles a little as eddies peel off: record the value it settles around.
   3. Repeat with Circular cylinder, then Teardrop (streamlined body).
   4. For each shape, describe the wake (wide or narrow, calm or swirling) and sketch the smoke (Part B).
   5. Optional: press 4 to colour the flow by vorticity (how fast the air spins) and 2 to go back to speed. Zoom with the mouse wheel or a pinch.
5. 35–50 min

   Explain

   Compare the results with the predictions. Then build the explanation together. Air cannot turn the plate’s sharp edges or follow the back of the cylinder, so the flow separates from the surface and leaves a wide, swirling wake. The wake is at low pressure, the front is at high pressure, and that pressure difference pushes the shape backwards: drag. The teardrop’s gently tapering tail lets the air follow the surface almost to the end, so the wake is narrow and the drag small.

   Streamlining is shaping the back as carefully as the front. Ask for examples: dolphins, penguins and tuna; a cyclist’s teardrop helmet; aeroplane fuselages; the long nose of a fast train. Students answer Part C.

6. 50–55 min

   Exit question

   “A truck company wants its trucks to use less fuel. Use today’s results to suggest one change to the shape of a truck, and explain why it would help.”

Extension

• Turn the teardrop round so its tail faces the wind: press V, click the teardrop and choose Mirror horizontally on the small toolbar beside it. Is it still streamlined? Why does the back matter so much?
• Predict, then test, other Fundamentals shapes: Square cylinder, Half cylinder, D-shaped body and Wedge (triangular prism).
• Delete the preset shape (press V, click it and press Del), then draw your own with the Ellipse tool (O) or the Brush (B) and try to beat the teardrop. Keep brush strokes thick (at least three grid cells), and give the flow a minute to settle after each change.
• Year 9: in the Tunnel tab change Wind speed from 10 to 20 m/s. Cd hardly changes, but the drag force goes up about four times: switch the mode to Enthusiast and look under Forces in the Analysis tab. What does that mean for the fuel a car uses at high speed?

Teacher notes

Flat plate

Cylinder

Teardrop

• The order and the big gaps are the result. All three readings are higher than textbook values: the tunnel walls are close (the Tunnel tab shows a blockage of 10–12 % and says results are biased high) and the free version’s grid is modest, which hits the teardrop’s thin tail hardest. The plate still has about seven times the teardrop’s drag. This is a good moment for the 2D question on the worksheet.
• The preset cards show textbook values (the flat plate’s card says Cd 1.98), so students will notice their measured numbers are higher. Use it: what is different about this tunnel and a real plate in open air?
• “Still settling”. For the cylinder the Analysis tab may keep saying the averages are still settling, because its wake never quite repeats. Read the status-bar Drag after a minute or two; it changes very little after that.
• Why Cd and not newtons? The drag coefficient compares shapes fairly, whatever their size and the wind speed. Bigger and faster means more force; Cd is about shape alone.
• Markers on the shapes. The green dot is the stagnation point, where the air stops against the front; red diamonds mark where the flow separates from the surface or reattaches to it.
• Lift. The plate, the cylinder and the teardrop shed eddies from alternate sides, so the Lift reading swings up and down and its average can sit away from zero for a while. It is not part of this lesson.
• Misconceptions to listen for. “A pointed front is what matters” (the tail matters more); “drag only depends on size” (all three shapes are the same height); “air only pushes on the front” (the low pressure in the wake behind adds to the drag).
• Timing. Loading a preset restarts the flow. The Drag number settles in about a minute on a laptop that is plugged in, longer on Battery saver. If time is short, run the flat plate on the projector while pairs test the other two.

Year level

Years 10–12

Subject

Physics: forces, Newton’s laws, lift

Time

50–60 minutes

WindCanvas

Free, 2D, Enthusiast mode

Setup (2 minutes)

1. Open windcanvas.com/app/ and choose Skip in the Welcome window.
2. Switch the mode in the top bar to Enthusiast. The status bar now shows L/D and the Reynolds number, and the Charts strip opens.
3. Open Presets, click the Aircraft chip, then NACA 2412, a classic light-aircraft wing section. It loads with air at 50 m/s and the wing at an angle of attack of 4°.
4. Press V, click the wing and open Object. The Angle of attack dial is at the top; you can also type an angle into its box. Zoom in on the wing with the mouse wheel or a pinch.

Step by step

1. 0–5 min

   Hook: how does a wing hold up an aeroplane?

   Students write their own explanation in Part A of the worksheet. Collect a few without judging them. Someone will probably say the air over the top has further to go, so it must go faster to meet the air from underneath at the back. Keep it for later.

2. 5–12 min

   Demonstrate

   Show the wing at 4°. Press 3 to colour the flow by pressure: low pressure (suction) over the top, higher pressure underneath. Press 2 for speed: the air over the top moves faster. Point to the smoke behind the wing: the air leaves the trailing edge heading slightly downwards. That downwash is the air the wing has pushed down.

3. 12–35 min

   Collect data

   1. Set the angle of attack to 0°. Wait until Drag and Lift in the status bar stop drifting (30–90 seconds; the Cd / Cl history chart in the Charts strip shows the lines levelling out).
   2. Record Cl (Lift), Cd (Drag) and L/D. In the Analysis tab, under Inviscid (panel method, high-Re) just below Results, record the panel-method Cl: the theory value for frictionless, fully attached flow.
   3. Note whether the smoke follows the upper surface to the trailing edge or breaks away into a wide, swirling wake.
   4. Repeat at 2° steps up to 18°. If time is short, give each pair three angles and pool the class results on the board.
4. 35–45 min

   Graph

   Students plot simulated Cl and panel-method Cl against angle on the same grid (Part C), mark where the simulated lift stops rising, and note where the drag starts climbing steeply.

5. 45–55 min

   Explain

   Two descriptions of the same physics. Pressure: the wing’s shape and angle make the air speed up over the top and slow down underneath, so the pressure on top is lower and the net force points up. Newton’s third law: the wing pushes air downwards (the downwash), so the air pushes the wing upwards. Then test the hook idea: nothing makes the two streams meet at the back, and in reality the air over the top arrives at the trailing edge well before the air underneath. At high angles the flow can no longer follow the upper surface; it separates, the wake widens, drag shoots up and lift stops rising: the stall. Students answer Part D.

6. 55–60 min

   Exit question

   “Why does an aeroplane raise its nose to take off, and why can’t it keep raising it?”

Optional: pressure along the surface (Cp)

Extension

• In the Analysis tab click Instant polar (under the panel method): the theory’s whole lift curve from −6° to 16° in milliseconds. It never stalls. Why not?
• Load NACA 0012, a symmetric section, and compare the panel-method lift at 0° with the cambered NACA 2412’s. What does camber do?
• Flip the wing upside down with Mirror vertically on the small toolbar beside it: lift becomes downforce, as on a racing car’s rear wing.
• Try Wing with slotted flap or Wing with leading-edge slat (Aircraft presets): how do airliners get extra lift for landing?

Teacher notes

4°: attached flow, thin wake

16°: separated flow, wide wake

• Expect the simulated lift to be low. In the free version the wing is only about five grid cells thick (the Tunnel tab shows Cells across body), so the thin layer of air next to the surface cannot be resolved. The simulated air behaves as if it were far more viscous than real air: it leaves the surface early and the lift comes out at a fifth to a half of the panel-method value. The app says so itself: the fidelity grade for these runs is D, “rough”. This is the heart of the evaluation question, not a fault in the lesson.
• What to trust. The panel method is the opposite limit: frictionless flow that never separates, so it never stalls. In published wind-tunnel data for the NACA 2412 at full scale, lift rises about 0.1 per degree, a little below the panel line, until the section stalls at a Cl of about 1.7 (Abbott and von Doenhoff, Theory of Wing Sections). Use the simulation for the shape of the curve, the drag, the flow pictures and the stall, and the panel method for full-scale lift.
• The stall in the simulation is gradual. Lift climbs slowly to about 10°, jumps somewhere between 10° and 14° as the separated region over the upper surface reorganises (pairs may disagree at 12°: a real anomaly worth discussing), then levels off at a Cl of about 1.0–1.1 from 14° to 18°. Meanwhile drag climbs steeply from about 12° and the smoke breaks into a wide, swirling wake. Lift levelling off while drag shoots up is the stall. Real wings at full scale stall more abruptly.
• Lift explanations. The pressure and Newton’s-third-law descriptions are two views of the same momentum balance; Bernoulli’s principle links the faster flow to the lower pressure, but does not say why the air speeds up. The “equal transit time” story is false: nothing requires the two streams to meet again.
• Timing. Each angle restarts the averaging, and the numbers settle in 30–90 seconds. Pooling angles across pairs gives a full curve in a few minutes, and the scatter between pairs is itself worth discussing.

Year level

Years 7–12

Subject

STEM, Design and Technologies

Time

50–60 minutes, plus write-up

WindCanvas

Free, 2D, Enthusiast mode

Setup (2 minutes)

1. Open windcanvas.com/app/ and choose Skip in the Welcome window.
2. Switch the mode in the top bar to Enthusiast, so the Analysis tab offers Compare snapshots.
3. Check the photo or sketch is on the device and the car’s length is written on the worksheet.

Step by step

1. 0–5 min

   Brief

   The goal: less drag, so the car is faster down the track. The rule: change one thing at a time, and keep the real length, the ground gap and the wind speed the same for every test. Teams plan their three changes in Part A.

2. 5–15 min

   Import the original design (version A)

   1. Drag the file onto the app, or open Import and click Drop files here or click to browse.
   2. In the import dialog, compare the Original with the Extracted outline; the Grid preview shows what the solver will see. Fix the outline if needed with the Keep and Erase brushes or Crop. Take any wheels out of the outline: drag the Ground line up to the underside of the body (everything below it is removed), or paint them out with Erase. The teacher notes explain why.
   3. Check that the line at the top of the settings says it looks like a car: the dialog then ticks Mount on ground and faces the car into the wind (if the nose still points right, click Flip to face the wind). Ignore the offer to use a preset instead (usually the STEM Racing car): this lesson is about the team’s own design.
   4. Check the Real length and change it if needed, for example 0.21 m for a 210 mm car (check the unit beside the box). The dialog guesses the length from the closest preset, so check it on every import. Set Gap to the car’s ground clearance in metres (0.005 for 5 mm).
   5. Name it “A original”, tick Apply recommended tunnel & flow (outside Beginner mode it starts unticked) and click Add to tunnel. The tunnel is sized for the car, with air at 20 m/s (about race speed) over a moving floor. Check Wind speed in the Tunnel tab reads 20 m/s, and from now on leave the tunnel settings alone.
3. 15–20 min

   Measure version A

   Let the flow run for about two minutes, until Drag in the status bar stops drifting. It may still wobble a little as eddies peel off the back of the car, so read it three times, ten seconds apart, and record the average and the fidelity grade (Part B). Then, in the Analysis tab under Compare snapshots, click Take snapshot.

4. 20–40 min

   Test three changes (B, C, D)

   Make one change, measure it the same way, take a snapshot, then undo back to version A (Ctrl+Z; snapshots are kept) before the next change. Two ways to change the design:

   • In the tunnel. The Erase tool (E) carves material away: round a sharp corner, slope the rear, slim the nose. The Brush (B) adds material: a smoother nose cone, a fairing, a longer tail. Use [ and ] to change the brush size.
   • On paper. Redraw the change, photograph it and import it again with the same Real length and Gap. Leave Replace current reference body ticked and Apply recommended tunnel & flow unticked, so the tunnel and the wind stay exactly the same.

   Ideas: a rounder or a sharper nose; a smooth step where the nose meets the body; a tapered tail or a cut-off (Kamm) tail; a lower rear wing; a lower overall height. Finally, undo back to version A and measure it once more: if it comes back within a percent or two of the first time, bigger differences between versions are real.

5. 40–50 min

   Compare and decide

   Under Compare snapshots, tick up to four versions: the table shows Cd, Cl, L/D, the Reynolds number, the speed and the fidelity of each, with the percentage change from the first. Which change cut drag the most? Switch the field (View tab, or keys 2 speed and 3 pressure) to see why: a smaller, calmer wake means less drag.

6. 50–60 min

   Rationale and evidence

   Teams draft their design rationale (Part C). For the portfolio, click Make report in the top bar: an A4 report with the set-up, results, charts and flow pictures, plus a comparison of the snapshots you ticked (in Enthusiast mode the Comparison section in the Report tab starts ticked). Print it or save it as PDF. In the free version the report is marked “WindCanvas Free”.

Extension

• Combine the best two changes into version E. Do the savings add up?
• Change Wind speed in the Tunnel tab from 20 to 30 m/s. Cd stays about the same, but the drag force under Forces (Analysis tab) more than doubles. Why?
• If a change alters the car’s height, compare Cd·A (under Forces) as well as Cd: the drag area decides the drag force.
• Import the photo again with the wheels left in. Why does the lift jump? (See the teacher notes.)
• Teams with a CAD model can test the real 3D car with a licence (STL and OBJ import, the 3D tunnel above Low). Registered STEM Racing teams can claim a free licence, and the STEM Racing page has a guide for teams and judges (windcanvas.com/stem-racing/).

Teacher notes

• Why the wheels come out. The 2D tunnel sees a side view as a slice through an infinitely wide car. Solid wheels then become two walls that seal the gap under the body and trap air over the moving road, which distorts lift and drag (the user guide quotes a test sedan with solid wheels reading a lift coefficient of about +3). The app’s own car presets draw their wheels as see-through, visual-only parts for the same reason. Neither choice models what real wheels do to the flow, so keep students’ attention on the body shape.
• Ignore lift in this lesson. A few millimetres of ground clearance spans only two or three grid cells in the free version, so the flow under the car is not resolved and the Lift reading can swing wildly from one version to the next (in our test from −0.3 to +3.3); the Analysis tab says so: Underbody gap under-resolved. Drag comparisons between versions with the same Gap remain useful.
• Typical numbers. Side views give high drag coefficients, above 1, because the 2D tunnel sees an infinitely wide car with a blunt back, so compare versions rather than quoting the number as the car’s Cd. In our test (the 210 mm body in the picture, 5 mm gap, 20 m/s over a moving floor, free version) the original read Cd 1.38, graded C: a 2D run of a car is always marked down, and C still supports comparing versions. Lowering the tall rear and tapering the tail cut drag by about 37 % (Cd 0.86). A blunter nose read lower too (Cd 0.93 on average), but its reading wandered by about ±12 %. Results can surprise: that is why designers test.
• Small differences. Once the flow has settled, the status-bar Drag (an average over the last two passes of air through the tunnel) often holds steady to a fraction of a percent, even while eddies peel off the car, but for some shapes it keeps wandering: by about ±12 % for the blunt-nosed version in our test. The three readings show which case you have: treat a difference smaller than the spread of the readings as a tie.
• Thin parts. Features thinner than about three grid cells, such as a thin rear wing seen side-on, are unreliable at this resolution; the Grid preview in the import dialog shows how thick they come out.
• Warnings are fair to every version. The Analysis tab may list High blockage, Body is under-resolved or Wake reaches the outlet for a small car in the free version’s grid. They are true, and they affect every version alike, so comparisons stay fair.
• Fair-test checklist. The same Real length, Gap and wind speed for every version; the tunnel left alone; the same crop when re-importing; two minutes to settle and three readings every time.
• Keeping the evidence. The free version autosaves in the browser, but saving project files needs a licence: make the report and save it as PDF before the end of the lesson.