DIY Barometer Experiment: Predict Tomorrow's Weather

The first time the straw on your jar drops two millimetres and your kid says “it’s going to rain tonight” and then it does, something clicks that no weather app can deliver. This DIY barometer experiment takes fifteen minutes to build out of a jar, a balloon and a drinking straw. What makes it worth doing properly is the problem almost every tutorial hides: your barometer is also a thermometer, and if you do not control for that, your readings are noise with a story attached. Fix that, and you have a real instrument.

Key Takeaways

• You seal a glass jar with a stretched balloon, tape a straw to the membrane as a pointer, and log its position twice a day for two weeks.
• The physics is the sealed air inside pushing against the atmosphere outside, so a falling pointer means rising outside pressure.
• Best for ages 8 to 14. Younger kids can build it; reading a 2 mm change needs patience.
• The thing that goes wrong: temperature. A 3 °C room swing moves the membrane as much as a 10 hPa pressure change.
• No hazards worth naming, which makes this one of the few experiments a kid can own completely.

What You’ll Need for the DIY Barometer Experiment

Item	How much	Where to get it	If you don’t have it
Glass jar with a smooth rim	1, 500 ml to 1 litre	Jam or pickle jar; a smooth rim seals far better than a threaded one	A tin can with the lid cleanly removed
Balloon	1, standard round	Party supplies	A square cut from a sturdy rubber glove
Rubber bands	2 or 3, tight	Stationery drawer	Strong tape wound twice
Drinking straw	1, the longest you have	Kitchen	A bamboo skewer, lighter and stiffer
Index card or cardboard	1 piece, about 10 by 15 cm	Craft drawer	The side of a cereal box
Ruler with millimetres and a fine pen	1 each	Home	Any ruler; millimetres matter here
Thermometer	1	Pharmacy or kitchen drawer	A phone’s room sensor, if it has one
Tape	A few centimetres	Home	Glue stick, less reliable

Time: prep 15 min, first reading 2 min, then 2 min twice a day for two weeks Cost: everything is probably in your kitchen Mess level: low

Do the Experiment

Step 1 — Cut and stretch the balloon

Snip the neck off the balloon and discard it. Stretch the remaining dome flat over the jar mouth like a drum skin, pulling it tight and even, then fix it with two rubber bands wound as tight as you can manage. The seal has to be airtight. If air can creep in or out, the jar equalises with the room and your instrument reads nothing forever.

Step 2 — Tape the straw to the membrane

Lay the straw flat across the drum so one end sits at the centre and the other sticks out well past the rim. Tape the centre end down with a single small piece of tape. The overhang is your lever: a 0.2 mm movement at the centre becomes a visible 2 mm swing at the tip if the straw is long.

Step 3 — Build the scale and mark zero

Stand the index card upright behind the straw tip, taped to a block or a book so it will not move. Draw a horizontal line at the tip’s current height, write today’s date and the time, and note the room temperature. Then find the actual sea-level pressure for your town from a weather app or a national weather service page and write that number next to your mark.

Step 4 — Put it somewhere boring and leave it

The right spot is an interior shelf away from windows, radiators, draughts and sunlight, where the temperature barely moves. A kitchen windowsill is the worst place in the house for this and the most common choice.

Step 5 — Log three numbers twice a day

Every morning and evening, record the straw tip’s height in millimetres, the room temperature, and the official pressure. Three columns. Two weeks. That third column is what turns a craft project into a calibration.

Step 6 — Plot it and make a forecast

Graph straw height against official pressure. You should get a rough straight line going the other way: higher outside pressure pushes the membrane in and the tip up. Once the line exists, start forecasting. A steady fall over three to six hours means a low-pressure system approaching, so cloud and rain are likely within a day. A steady rise usually means clearing.

What you should see: Movements of 1 to 4 mm at the straw tip over a few days, not dramatic swings. On a day when a front passes, you may see 5 mm or more. If nothing ever moves, your seal has failed. If it wanders randomly with no relation to the official pressure, check your temperature column before you blame the barometer.

What’s Actually Going On

Your jar holds a fixed number of air molecules. The balloon is the only flexible wall, so it sits wherever the push from inside equals the push from outside. When a high-pressure system moves over your town, the atmosphere presses harder on the membrane, it dents inward, and the far end of the straw rises. When pressure drops, the trapped air wins and the membrane bulges out, dropping the tip. That is the same principle as a real aneroid barometer, which uses an evacuated metal capsule instead of a rubber sheet.

Atmospheric pressure is simply the weight of the air column above you. At sea level the standard value is 1013.25 hPa, equivalently 29.92 inches of mercury, and the H’s and L’s on a weather map are departures from that (NOAA JetStream). Low pressure means air is converging and rising, and rising air cools, condenses and makes clouds. High pressure means air is sinking and warming, which dissolves cloud. That chain, not the number itself, is why a barometer forecasts anything.

Now the honest part. Your instrument obeys the ideal gas law, which links pressure, volume and temperature together: PV = nRT. The membrane cannot tell the difference between the outside pressure dropping and the inside air warming up. Room temperature is around 293 K, so a 3 °C rise is about a 1% increase in absolute temperature, and 1% of 1013 hPa is roughly 10 hPa of fake signal. Real weather swings are often only 5 to 15 hPa. Put the jar on a sunny sill and the daily temperature cycle will swamp the weather entirely. Commercial aneroid barometers include a bimetallic compensator for exactly this reason. Yours does not, so you compensate with a notebook.

One more real effect worth knowing before you over-interpret your graph: pressure has a built-in twice-daily rhythm driven by solar heating of the atmosphere. NOAA notes it bottoms out near 4 a.m. and 4 p.m. and peaks near 10 a.m. and 10 p.m., strongest near the equator. So reading at breakfast one day and at dinner the next introduces a wobble that has nothing to do with the weather. Read at the same two times every day.

How to Explain It to Your Kid

Ages 5–8: the air is sitting on us

Have them hold a hand flat, palm up, and tell them there is a column of air stretching all the way to space standing on it, and it weighs about as much as a bowling ball. They do not feel it because the same air pushes from underneath. Then show the straw moving and say the air is sometimes heavier and sometimes lighter, and the jar notices before we do.

Ages 9–12: a number with a direction

Give them the word hectopascal and the benchmark 1013. The useful idea is that the number alone means little; the direction of change is the forecast. Have them write a prediction every evening based only on which way the straw moved that day, then score themselves over two weeks. A 70% hit rate from a jar is genuinely good.

Ages 13+: the gas law and the confound

Give them PV = nRT and let them work out the temperature sensitivity themselves. Then have them design the fix: either a temperature-controlled location, a correction factor derived from their own data, or a second identical jar kept at the same temperature as a differential reference. Comparing their readings against an official station is a first lesson in instrument calibration.

The question to ask: “If we carried this barometer up a mountain, which way would the straw move, and would the weather have anything to do with it?”

Why Teaching This Early Pays Off

This experiment trains a specific skill that shows up everywhere later: separating a signal from a confounding variable. The National Research Council’s Framework for K-12 Science Education names “planning and carrying out investigations” and “analyzing and interpreting data” among its eight core practices, and controlling a known confound sits at the centre of both.

There is also a teaching structure worth borrowing. Asking for a prediction before each reading is the Predict–Observe–Explain pattern described by Richard White and Richard Gunstone in Probing Understanding, and it works because a wrong prediction makes the real result memorable in a way that passive observation does not. Chi and Wylie’s ICAP framework points the same direction: having a child explain and predict outranks having them watch.

Keep the claim modest. A fortnight of barometer readings will not raise anyone’s grade in physics. What it reliably produces is a kid who understands that measurement instruments have systematic errors, which is more than most adults walk around with.

Before You Start: What Parents Should Know

Best age	8 and up. A 6-year-old can build it, but reading millimetres twice a day for two weeks needs an older kid or a committed parent.
Supervision	Minimal. An adult should handle cutting a tin can if you go that route, because the cut edge is sharp.
Real hazards	None beyond normal kitchen care. A balloon neck is a choking hazard, so bin it immediately rather than leaving it on the table.
Don’t	Do not use mercury for any part of this, and do not improvise a liquid barometer with anything but water. Mercury vapour is a genuine neurotoxin.

When It Doesn’t Work

• The straw never moves at all. The seal leaked, almost certainly at a threaded jar rim where the balloon cannot sit flat. Switch to a smooth-rimmed jar and add a third rubber band.
• The straw drifts steadily in one direction for days. Rubber creeps under tension. Give the balloon two days to settle before you mark your zero line, and re-mark zero after a week.
• The readings bounce around with no pattern. Check your temperature column. If the room swings more than 2 °C between readings, move the jar somewhere duller.
• It moves, but opposite to the official pressure. You have the geometry backwards: check whether the straw is taped at the centre of the membrane, not at the rim. Taped at the rim, it pivots the other way.

Take It Further

Add a second variable and graph both. Log relative humidity from a cheap hygrometer alongside pressure. Falling pressure plus rising humidity is a far stronger rain signal than either alone, and seeing that in your own data beats being told.

Build the control jar. Make a second identical barometer and keep it inside a closed cupboard at stable temperature. Compare the two: the difference between them isolates how much your main instrument is responding to heat. That is a real experimental design, not a craft. Pair it with a backyard weather station and the can-crushing demonstration of how hard air pushes.

What to Watch For Over the Next 3 Months

• Week 4: They check the barometer before deciding whether to take a jacket. Using an instrument rather than admiring it is the first real sign.
• Month 2: They notice their ears pop in a lift or on a mountain road and connect it to the jar on the shelf.
• Month 3: They ask why the forecast on the phone is better than their jar. The answer, which is thousands of simultaneous readings plus a physics model, leads naturally toward how modern forecasting works.

Frequently Asked Questions

How accurate can a jar barometer really be?

With a long straw and a stable room, a few hectopascals, which is enough to spot a passing front. It will never give you an absolute reading without calibrating against an official station, because you do not know the pressure that was inside when you sealed it.

Why does low pressure mean rain?

Low pressure is air converging and rising. Rising air expands and cools, water vapour in it condenses, and clouds form. High pressure is sinking air that warms and dries, which clears cloud away. The pressure is the cause, not the forecast.

We live at 2,000 metres. Does this still work?

Yes, and arguably better, because you are measuring changes rather than absolutes. Your sealed-in pressure simply starts lower. Compare against your local station’s station-pressure reading rather than the sea-level-adjusted figure if you want the numbers to line up.

Can we use a plastic jar?

It works, but less well. Plastic flexes with temperature and pressure itself, which adds a second uncontrolled variable to an experiment whose whole lesson is about controlling one. Glass is rigid, so the balloon does all the moving.

How long before the balloon wears out?

Two to six months. Rubber hardens and loses elasticity, and ultraviolet light speeds that up, which is another reason to keep it out of sunlight. When readings go flat, re-skin it and start a fresh zero.

---

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. NOAA National Weather Service. “Air Pressure.” JetStream: An Online School for Weather. https://www.noaa.gov/jetstream/atmosphere/air-pressure
2. NOAA Global Monitoring Laboratory. “Solar Calculator.” https://gml.noaa.gov/grad/solcalc/
3. National Research Council. (2012). A Framework for K-12 Science Education: Practices, Crosscutting Concepts, and Core Ideas. National Academies Press. https://doi.org/10.17226/13165
4. White, R., & Gunstone, R. (1992). Probing Understanding. London: Falmer Press.
5. Chi, M. T. H., & Wylie, R. (2014). “The ICAP Framework: Linking Cognitive Engagement to Active Learning Outcomes.” Educational Psychologist, 49(4), 219–243. https://doi.org/10.1080/00461520.2014.965823
6. Alfieri, L., Brooks, P. J., Aldrich, N. J., & Tenenbaum, H. R. (2011). “Does discovery-based instruction enhance learning?” Journal of Educational Psychology, 103(1), 1–18. https://doi.org/10.1037/a0021017