Water Bottle Rocket Experiment: Thrust, Mass and Height

An air-only bottle rocket hops. The same bottle with a third of a litre of water in it leaves the pad hard enough that you lose it against the sky for a moment. The air did not change. What changed is that the rocket now has something heavy to throw backwards.

That is the lesson of a water bottle rocket experiment, and it is measurable: vary the fill, measure the height, find the peak. First, read the safety section. This is the one project in the series that can genuinely injure an eye, and the rules below are requirements, not suggestions.

Key Takeaways

• You will launch a pressurised plastic drinks bottle at five different water fill fractions and estimate the height each time.
• The concept is thrust as mass thrown per second times how fast you throw it, which is why water beats air.
• Best for ages 9 to 15, with an adult doing every pressurisation and every release.
• The non-negotiables: outdoors, straight up, goggles on everyone, 8 m standing-back line, 40 psi ceiling, plastic drinks bottle only.
• The result worth chasing: there is a best fill fraction, usually near one third, and you can find it.

What You’ll Need

Item	How much	Where to get it	If you don’t have it
Plastic carbonated-drinks bottle, 1.5 or 2 L	4, undamaged	Supermarket	Nothing else. See the Don’t row; not substitutable
Purpose-made launcher with remote release	1	School science suppliers	Never improvise a cork launcher
Bicycle pump with a pressure gauge	1	Bike shop	A pump with no gauge is unusable; the gauge is a safety device
Long pump hose, 2 m or more	1	Bike shop	So nobody stands over a pressurised bottle
Safety goggles	1 per person present	Hardware shop	None. No substitute
Plastic folder or report cover	1	Stationery drawer	For three fins
Strong tape	1 roll	Toolbox	Duct or electrical tape
Modelling clay	small lump	Stationery shop	A wad of wet paper. Never a hard or heavy object
Two stopwatches	2	Phones	Two give you an error estimate
Measuring jug	1	Kitchen	For accurate fill fractions
Notebook	1	Anywhere	Three launches at each of five fills

Time: prep 30 min, setup 20 min, launching 60 min, cleanup 15 min Cost: the launcher and the goggles; everything else is cheap or already yours Mess level: do this outside, and expect to get wet

Do the Experiment

Step 1 — Inspect the bottle and set up the range

Check each bottle: no cuts, no scratches, no whitened creases, no cracks at the base. Reject any bottle that is not perfect. Then set the range: an open flat area with no overhead wires, nothing above you, no roads and no windows within 20 m. Chalk a launch point and a standing line 8 m back.

Step 2 — Fit three fins and check the balance

Cut three identical fins from the plastic folder and tape them at 120 degrees around the bottom third of the bottle, aligned with the body. Add a small lump of clay inside the nose. Then balance the empty finned rocket on a finger: the balance point must sit forward of the middle of the fin area. Skipping this check produces a rocket that tumbles.

Step 3 — Fill, mount and clear the pad

Measure water with the jug, starting with one third of the bottle’s volume. Mount the bottle on the launcher, connect the long hose, and check the seal is seated. Then everybody, without exception, walks back behind the 8 m line. Goggles on. Nobody returns to the pad while the bottle holds pressure.

Step 4 — Pressurise to the ceiling and launch

An adult pumps, from behind the line, watching the gauge, and stops at 40 psi. Not 45. If the bottle creaks, bulges or whitens at any point, stop pumping and release the pressure remotely. Count down out loud so everyone is looking up, then pull the release cord.

Step 5 — Time the flight twice

Two people start stopwatches at launch and stop when the rocket lands. Record both times. Halve the average to get the time to the top, then compute the height as half of 9.8 times that time squared. Three launches at this fill.

Step 6 — Vary the fill and plot

Repeat at fills of 0, one sixth, one third, one half and two thirds of the bottle volume, three launches each, at the same 40 psi every time. Plot estimated height against fill fraction.

What you should see: the air-only bottle reaches a disappointing fraction of a water-loaded one, which is the headline. A one-third fill at 40 psi typically reaches 25 to 45 m, with total flight times of four to six seconds. Two thirds full it barely clears head height. Your curve rises from zero fill, peaks between one sixth and one half, and falls again. Launch-to-launch spread is around 20 percent, so three launches per point is the minimum.

What’s Actually Going On in a Water Bottle Rocket

Thrust is the rate at which you throw mass backwards multiplied by how fast you throw it, plus a smaller term from the pressure difference at the nozzle (NASA Glenn on thrust). Both factors matter, and that is why the water is not optional. Compressed air leaves fast but weighs almost nothing, so the mass flow rate is tiny. Water is hundreds of times denser, so loading it gives the rocket a heavy propellant and the same stored pressure does far more useful work.

NASA’s own description matches what you will see. The body is an empty plastic two-litre soda bottle, typically filled to about 40 percent of its volume with water, with fins for stability. Air is pumped in, and thrust comes from the water expelled through the nozzle. The weight changes constantly during the powered ascent, and as the water goes the air volume increases, which drops the pressure, the mass flow rate and therefore the thrust. When the water is gone there may be a small extra push as the pressure equalises, after which the rocket coasts and then flies like a ballistic shell modified by drag (NASA Glenn).

That falling pressure is why there is an optimum fill rather than “more water is better.” Too little water and you run out of propellant immediately, so most of your stored energy escapes as a puff of air. Too much and two things go wrong: you are lifting dead weight, and the air space depressurises in hundredths of a second. The best fill is the compromise, a real optimisation problem with a measurable answer.

The underlying quantity is the mass ratio, full mass divided by empty mass, which sets how much velocity a given propellant can deliver (NASA Glenn on mass ratios). Your fill-fraction curve explores the same trade-off that decides how many stages a real launch vehicle needs.

Stability is a separate problem and the usual reason a rocket fails. An unfinned bottle tumbles because the aerodynamic forces act at a point too close to the centre of gravity, so any small disturbance grows. Fins at the back move that centre of pressure rearwards, and the requirement is that it sits behind the centre of gravity (NASA Glenn on rocket stability). Nose weight and rear fins are your two tools, and Step 2 is the test.

One honest word on the heights. Timing the flight and applying the free-fall formula ignores drag, which is substantial for a blunt bottle, so your absolute heights will be overestimates by perhaps 10 to 30 percent. What the method gets right is the comparison between fills, because the error is similar for all of them. Say that out loud: the ranking is trustworthy and the number is not.

How to Explain It to Your Kid

Ages 5–8: Spectator only, and that is fine

A five-year-old watches from behind the line in goggles, not launching. Ask them before each launch whether more water or less will win. Afterwards say the rocket goes up because it throws water down, and the more it throws the harder it pushes itself up, until it gets too heavy to lift.

Ages 9–12: Two things make thrust

Introduce the two factors: how much stuff leaves per second and how fast. Air leaves fast but weighs nothing; water is heavy. Then pose the optimisation as a puzzle: if more water gives more push, why does a full bottle barely move? Let them reach the answer that the bottle also has to lift the water, and the air has nowhere to expand.

Ages 13+: Momentum, mass ratio and honest error bars

Write thrust as mass flow rate times exhaust velocity and reason about how both change as the bottle empties. Compute the mass ratio at each fill and relate it to the measured height. Finish with error analysis: compare your two stopwatch times and discuss why drag biases every height the same way. One experiment with stated uncertainty beats five without.

The question to ask: “If we filled it with syrup instead of water, what would happen?” (Thrust would rise, since syrup is denser, but the viscous syrup leaves the nozzle more slowly and the rocket is heavier, so the outcome is genuinely not obvious. It is a good question precisely because the answer requires weighing two effects rather than recalling one.)

Why Teaching This Early Pays Off

Rockets are the one place where almost every child already has a wrong model, usually that the exhaust pushes against the ground or the air. Building one whose propellant you can see and weigh replaces that with the correct idea: you go forwards by throwing mass backwards. That is also why real launch vehicles are mostly propellant by weight.

The second payoff is the optimisation. Most school science produces monotonic results, where more of something gives more of something else. Here the answer is a peak, and finding a peak needs a series of measurements rather than one trial. The National Research Council’s framework names planning investigations and analysing data among the core practices of school science, and a five-point curve with three launches per point is a real instance of both (NRC, 2012).

Keep the claims honest: one afternoon of launches does not set a career path. What it gives a teenager is a genuine dataset, an optimum they found rather than were told, and the experience of reporting a number with a caveat attached. See also our guides to water rocket physics for ages 8 to 14 and the gentler balloon rocket on a string.

Before You Start: What Parents Should Know

Best age	9 to 15 for participating. Younger children watch from behind the line, in goggles
Supervision	An adult performs every pressurisation and every release, from behind the standing line, with a gauge in view. No exceptions and no turns
Real hazards	Eye injury from a bursting bottle or a launching rocket, which is the serious one. Also a rocket striking a person, a bottle failing while someone leans over it, and a rocket hitting an overhead wire
Don’t	Use a plastic carbonated-drinks bottle only. Never glass. Never a repurposed container of any kind: no milk jugs, no juice bottles, no cleaning-product bottles, no chemical containers. Never exceed 40 psi. Never aim anywhere but straight up, and never at a person, an animal, a building, a road or a window

The full rule set, and all of it matters. Outdoors only, in an open flat area with nothing overhead, no power lines, and no roads, windows or people within 20 m. One direction: straight up, from a launcher set vertical on level ground, decided before the first launch. Goggles on every person present, including spectators, because a bursting bottle sends fragments sideways. An 8 m standing-back line, marked with chalk, which everyone is behind before pressurisation begins and does not cross until the rocket has landed and the launcher has vented. A 40 psi ceiling, read from a gauge, because a bottle is a pressure vessel and the whole risk is pressure. A remote release, so nobody is within reach of a pressurised bottle. A long hose, so the pump operator is behind the line too.

Two more. Retire a bottle after roughly 20 pressurisations, and immediately at the first sign of a whitened crease, a bulge or a scratch; PET fatigues, and a tired bottle fails without warning. And keep the nose soft. A lump of clay or a wad of wet paper taped lightly in the nose is nose weight; a wooden cone, a metal bolt or a cricket ball is a projectile, and a rocket that comes down nose-first from 40 m should be something you would be happy to be hit by.

When It Doesn’t Work

Air hisses out and the pressure never builds. The seal between launch tube and bottle neck is not seated, or the pump connection leaks. Reseat the O-ring, wet it slightly, check the hose fitting. Never fix a leak by pumping harder.

The rocket tumbles instead of flying straight. The centre of pressure is not behind the centre of gravity. Add more clay in the nose or make the fins larger, then redo the balance check in Step 2.

It corkscrews and veers sideways. Fin misalignment. All three fins must be parallel to the bottle axis and spaced at 120 degrees. One fin twisted by a few degrees spins the whole rocket, and a veering rocket is a safety problem, so fix it before the next launch.

You are tempted to build a cork launcher. Don’t. A stopper pushed into the bottle neck fires when the friction happens to give way, which is an unpredictable moment, often while a hand is still nearby. Everything that makes this project safe depends on the launch being triggered deliberately, from a distance. Use a purpose-made launcher with a remote release, or do not launch.

Heights scatter wildly between launches. Some of it is genuine, because fill, seal and wind all vary. Check the wind first: anything beyond a light breeze ruins both the data and the safety case. Then check you are reading 40 psi the same way each time and measuring the water rather than estimating it.

Take It Further With Your Water Bottle Rocket Experiment

Find the optimum fill properly. Seven fills instead of five, three launches each, all at 40 psi, with the median plotted and uncertainty from the spread of your two stopwatch readings. The peak position is the result, and a peak with error bars looks like real science.

Map height against pressure. Hold the fill at your optimum and launch at 20, 30 and 40 psi, three each. Never above 40. The relationship is not linear, because stored energy rises faster than pressure does.

Measure the drag penalty. At fixed fill and pressure, launch with fins only, then with fins plus a smooth paper nose cone. Three launches each. The nose cone should buy measurable height, which makes the point that drag is not a detail.

What to Watch For Over the Next 3 Months

• Week 4: Your kid explains, unprompted, why a rocket needs propellant mass and not just pressure.
• Month 2: They notice staging in a launch video and connect it to carrying dead weight.
• Month 3: They ask about a different propellant or a two-bottle rocket. Encourage the calculation and the sketch, and keep the 40 psi ceiling and the goggles exactly where they are.

Frequently Asked Questions

Why does it need water at all?

Because thrust depends on how much mass you throw backwards each second as well as how fast. Air leaves quickly but weighs almost nothing, so thrust is small. Water gives the rocket a dense propellant to expel, which is why a water-loaded bottle far outperforms an air-only one.

What is the best amount of water?

Usually around a third of the bottle volume, and NASA describes a typical fill of about 40 percent. The exact optimum depends on your bottle, pressure and total mass, which is why finding it by measurement is the point.

What pressure is safe?

Treat 40 psi as a hard ceiling, read from a gauge, with an adult pumping from behind the standing line. Stop immediately if the bottle creaks, bulges or shows a whitened crease. A pressurised bottle is a pressure vessel, and pressure is the entire risk in this project.

Can we use a different bottle?

Only a plastic carbonated-drinks bottle, undamaged, 1.5 or 2 litres. Never glass, and never a repurposed container such as a milk jug, a juice bottle or anything that held a cleaning product. Those plastics are not designed for internal pressure and can fail into sharp pieces.

Do we really need goggles if it is only water?

Yes, for everyone present including spectators. The risk is not the water, it is a bottle failing under pressure and throwing fragments sideways, and a rocket arriving back at the ground faster than you expect. Impact-rated glasses sold for building work are inexpensive and adequate; OSHA’s eye and face protection guidance explains the standard behind them.

---

About the author

Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.

---

Sources

1. NASA Glenn Research Center. “Flight of a Water Rocket.” Beginner’s Guide to Aeronautics. https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/flight-of-a-water-rocket/
2. NASA Glenn Research Center. “Fun with Bottle Rockets.” Beginner’s Guide to Aeronautics. https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/fun-with-bottle-rockets/
3. NASA Glenn Research Center. “Thrust.” Beginner’s Guide to Aeronautics. https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/thrust-force/
4. NASA Glenn Research Center. “Mass Ratios.” Beginner’s Guide to Aeronautics. https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/mass-ratios/
5. NASA Glenn Research Center. “Conditions for Rocket Stability.” Beginner’s Guide to Aeronautics. https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/conditions-for-rocket-stability/
6. Occupational Safety and Health Administration. “Eye and Face Protection.” U.S. Department of Labor. https://www.osha.gov/eye-face-protection
7. National Research Council. (2012). A Framework for K-12 Science Education. National Academies Press. https://doi.org/10.17226/13165