At What Temperature Does the Air Start Making Water?

Measuring the Dew Point — When Does Invisible Water Appear?

A cold glass becomes wet on the outside. Where did that water come from—and can we measure the temperature at which it first appears?

Bring a cold drink into a warm room and droplets soon gather on its surface.

The water has not leaked through the glass. It was already present in the surrounding air as invisible water vapour.

With a shiny metal container, some ice and a thermometer, we can investigate the moment that invisible vapour begins becoming visible liquid.

This gives us an estimate of the dew point: a temperature with connections to misty mornings, wet grass, cloud formation, steamed-up windows and the comfort of a summer afternoon.

The title poses a deliberately intriguing question, but the distinction matters:

The air is not making new water. We are revealing water that was already there.

Why “The Humidity Is 70%” Does Not Tell the Whole Story

“The humidity is 70%” sounds informative.

But 70% at what temperature?

Relative humidity compares the actual water-vapour pressure with the saturation water-vapour pressure at the same temperature:

Relative humidity (%) = 100 x actual vapour pressure / saturation vapour pressure

The saturation value increases with temperature. Consequently, two days with the same relative humidity can contain very different amounts of atmospheric moisture.

Dew point asks a different question:

To what temperature would this air need to cool, at constant pressure and without changing its water-vapour content, to reach saturation?

At a given pressure, a higher dew point indicates more water vapour. It can therefore be more useful than relative humidity alone when comparing how moist different air masses are. Relative humidity remains useful, particularly alongside temperature. weather.gov

What Is Actually Happening?

Water molecules continually move between liquid water and its vapour. Cooling changes the balance between evaporation and condensation.

When a surface becomes sufficiently cold, water vapour in the neighbouring air can condense onto it.

At first, we may see only a faint misting of the metal. Later, small droplets become obvious.

That first persistent misting is the observation we want.

The whole room does not have to reach its dew point. A cold surface can develop condensation while the room air remains much warmer.

This also provides a useful opportunity to distinguish invisible water vapour from visible liquid droplets. The white cloud above a kettle consists mainly of tiny droplets, rather than the invisible gaseous water itself.

The Experiment: Find the First Trace of Dew

Suitable level: GCSE upwards, with extensions for A-level students.
Time: About 20–30 minutes for an initial investigation; longer for repeated measurements.

Equipment

• A clean, thin-walled, shiny metal cup or beaker.
• Water initially close to room temperature.
• Small pieces of ice.
• A digital thermometer or temperature probe.
• A stirring rod or spoon.
• A dry cloth, tray and notebook.
• Optional: a contact surface-temperature probe and a second thermometer for room temperature.

A smooth metal vessel makes the first change in appearance easier to see. Use a manufactured cup or beaker rather than a cut tin with a sharp rim.

Step-by-step method

1. Measure the room temperature. Keep the thermometer away from direct sunlight, radiators and the cooling vessel.

2. Prepare the vessel. Half-fill it with room-temperature water. Clean and dry the outside completely.

3. Choose an observation area. Watch the side below the water level, where the metal is being cooled by the water.

4. Position the temperature probe. Ideally, use a suitable contact probe against the outside near the observation area, leaving adjacent metal visible. For the simpler method, immerse the thermometer in the water without touching the wall or bottom.

5. Add a little ice. Stir gently without splashing. Allow the temperature to settle before adding more.

6. Slow down near the transition. Record the temperature at regular intervals and inspect the surface using good side lighting.

7. Record the boundary. Note the last temperature at which the surface is definitely dry and the first temperature at which a fine mist persists.

8. Repeat the investigation. Let the vessel return to room temperature, replace its water as necessary, dry the exterior and complete at least three trials.

Avoid breathing directly onto the observation area: your breath changes the air you are trying to measure. Handle the vessel as little as possible during readings.

Use a tray, wipe up spills and keep water away from electrical equipment.

Are we measuring the water or the surface?

The surface temperature is the relevant quantity.

An immersed thermometer gives an approximation. In a thin metal vessel with well-stirred water and slow cooling, the water and wall temperatures should be reasonably close, but they will not necessarily be identical.

Describe the result as an estimate of dew point, rather than claiming laboratory precision.

The metal-can-and-ice method is also described by the NOAA National Weather Service. Dry Bulb/Web Bulb/Dew Point

A Worked Example: The Moment Matters

These readings are illustrative, rather than measurements from my laboratory.

Surface temperature	Observation
15.0 C	Dry
14.0 C	Dry
13.0 C	Dry
12.5 C	Definitely dry
12.0 C	First persistent misting
11.5 C	Clear droplets developing

The transition lies somewhere between the final dry reading and the first misted reading.

A sensible report would be:

“The dew point was approximately 12 C, with condensation first observed between 12.0 C and 12.5 C.”

That interval describes the observation boundary. It is not a complete uncertainty calculation: probe accuracy, temperature differences and difficulty judging the first mist also matter.

If the room temperature was 20 C:

Dew-point depression = air temperature - dew-point temperature

Dew-point depression = 20 - 12 = 8 C

This tells us how far the room air would need to cool to reach saturation under the stated conditions.

A window surface could already be colder than that, even while the room remains at 20 C.

What If the Experiment Does Not Behave Perfectly?

Some of the most useful learning comes from understanding why an experiment gives an uncertain or unexpected result.

Condensation appears almost immediately

The starting vessel may already be below the dew point.

Bring the vessel and water closer to room temperature, dry the outside and restart.

Different parts of the vessel mist at different times

The wall may not have a uniform temperature. Ice pressed against one side can create a particularly cold patch.

Stir gently and observe a consistent area away from direct ice contact.

The surface is wet before cooling

Splashes can be mistaken for condensation.

Begin with a dry surface and add ice carefully.

No condensation appears

In sufficiently dry air, the dew point may be below the temperature you can reach using plain ice and water. A mixture containing both ordinarily settles near 0 C.

Record the lowest temperature reached and the absence of visible condensation.

That is a useful limitation of the method, rather than a failed investigation. For this introductory experiment, avoid adding salt to push the temperature lower.

The thermometer displays lots of decimal places

A reading of 12.37 C does not establish that the dew point is accurate to 0.01 C.

The observation and measuring system must justify the precision reported.

A precise-looking number is not necessarily an accurate measurement.

A Thermal-Camera Trap Worth Demonstrating

With a FLIR camera available, it is tempting to point it at the shiny vessel and read its temperature.

However, shiny metal has low emissivity and reflects infrared radiation from its surroundings. Its apparent temperature can therefore be misleading.

A suitable high-emissivity tape patch, allowed to reach the wall temperature and measured with appropriate camera settings, can improve the reading. Keep a separate bare-metal area for observing condensation and compare with a contact probe. Flir

This turns a measurement problem into another lesson:

An instrument produces a reading, but we still have to understand what it is measuring.

From a Metal Cup to the Morning Weather

Dew: a surface reaches the threshold

On clear nights, surfaces can lose energy by radiation and become colder than the air above them.

Grass and car roofs may reach the dew point even when the reported air temperature remains higher.

That helps explain why one surface is wet while another nearby is dry. Met Office

Fog: droplets suspended in air

Dew forms on surfaces. Fog consists of tiny droplets suspended near ground level.

A small difference between air temperature and dew point means that relatively little additional cooling is needed to reach saturation.

However, that difference alone does not guarantee fog. Wind, moisture supply and the way the air cools also matter. Radiation fog, for example, develops as the ground and neighbouring air cool, with light winds favouring its formation. weather.gov

Students could compare morning visibility with their measurements, while avoiding the conclusion that every small temperature gap must produce fog.

Clouds: the connection goes upwards

Rising air expands and cools. If it reaches saturation, condensation onto suitable particles can produce cloud droplets.

Our vessel provides a cold surface; the atmosphere can provide cooling through expansion. Both connect cooling with condensation, although the mechanisms and conditions differ. weather.gov

Frost: sometimes the water becomes ice

Dew can freeze after forming. Hoar frost can also develop when water vapour deposits directly as ice on a sufficiently cold surface.

For advanced students, distinguish saturation over liquid water from saturation over ice. Dew point and frost point are related, but are not identical below freezing. weather.gov

Why Does the Bedroom Window Get Wet?

Imagine room air with a dew point of approximately 12 C.

The window’s inner surface cools to 9 C.

That creates conditions for condensation on the glass, even if the room air is still 20 C.

Three useful questions follow:

• Can we reduce the water vapour being added to the room?
• Can appropriate ventilation remove moisture?
• Can we keep the surface warmer?

Cooking, showers and drying washing indoors can add moisture. Insulation and improved glazing can affect surface temperatures.

Simply warming the air lowers its relative humidity if its moisture content stays the same. It does not, by itself, remove that water vapour or substantially lower its dew point at unchanged pressure.

This makes an excellent connection between a classroom measurement and understanding a home.

Condensation is one possible source of damp; leaks and other causes need separate investigation.

Why Do Some Summer Days Feel Sticky?

A high dew point indicates moist air. That generally makes sweat evaporation less effective, reducing the cooling it provides.

Dew point helps explain why a warm day can feel oppressive even when its relative humidity is lower than on a cool morning. weather.gov

It is not a complete measure of comfort or heat stress. Temperature, sunshine, wind, clothing and activity also matter.

Dew point and wet-bulb temperature describe different things; they should not be treated as interchangeable.

Turn One Lesson into a Weather Investigation

Over several weeks, students could record:

Record	What to include
Date and time	Use a reasonably consistent time
Location	Indoors or outdoors; exact observation position
Air temperature	Temperature in C
Estimated dew point	Temperature in C
Temperature gap	Air temperature minus dew point
Conditions	Cloud, wind, fog, dew or window condensation

For outdoor trials, work in shade and avoid rain or direct sunlight on the equipment.

Useful investigation questions include:

• Does a higher dew point always accompany a warmer day?
• Is morning dew more common after clear nights?
• Does indoor dew point change after cooking?
• How closely do repeated estimates agree?
• Does the hand-measured estimate agree with a weather-station value?

In my home laboratory, this would make a useful link between a visible experiment, PASCO temperature logging and the Davis weather station.

We should compare like with like. An indoor cup measurement and an outdoor station reading describe different air. Even outdoors, location, timing and sensor uncertainty can account for differences.

A simple graph of air temperature and dew point over time can reveal much more than an isolated humidity percentage.

What I Want Students to Take Away

After around forty years of teaching, I value experiments that begin with something familiar and lead to a better question.

A wet glass is familiar.

Asking exactly when it becomes wet turns it into an investigation.

I would begin by asking students to predict where the water comes from. Then I would let them decide what counts as the first reliable evidence of condensation.

The most valuable discussion may concern a blurred boundary, an unexpected reading or disagreement between observers.

Those are opportunities to practise scientific judgement:

Identify limitations, improve the method and report evidence honestly.

At Hemel Private Tuition, this is the sort of practical I want to connect with the ideas students meet in their courses. Temperature, particles and changes of state become tools for explaining something real.

Conclusion: Make the Invisible Measurable

Dew point gives us a temperature at which invisible atmospheric moisture can reveal itself.

It connects a metal cup on the laboratory bench with a misty river, dew on grass, a cloud overhead and droplets on a bedroom window.

The equipment is simple. The thinking can go much further.

Next time a cold drink develops droplets, ask: what temperature did its surface have to reach before the water appeared?

Further Reading

• NOAA National Weather Service: Dew Point vs. Humidity — https://www.weather.gov/arx/why_dewpoint_vs_humidity
• NOAA National Weather Service: Temperature — Dry Bulb/Wet Bulb/Dew Point — https://www.weather.gov/source/zhu/ZHU_Training_Page/definitions/dry_wet_bulb_definition/dry_wet_bulb.html
• Met Office: Dew — https://weather.metoffice.gov.uk/learn-about/weather/types-of-weather/temperature/dew
• FLIR: How Does Emissivity Affect Thermal Imaging? — https://www.flir.com/en-asia/discover/professional-tools/how-does-emissivity-affect-thermal-imaging/
• NOAA National Weather Service: Radiation Fog — https://www.weather.gov/safety/fog-radiation
• NOAA National Weather Service: Cloud Development — https://www.weather.gov/source/zhu/ZHU_Training_Page/clouds/cloud_development/clouds.htm
• NOAA National Weather Service: Dew and Frost Development — https://www.weather.gov/source/zhu/ZHU_Training_Page/fog_stuff/Dew_Frost/Dew_Frost.htm
• NOAA National Weather Service: Dew Point Statistics — https://www.weather.gov/tbw/dewpoint

 

Conway's Game of Life — Can Mathematics Become Alive?

Four simple rules. An infinite chessboard. And behaviour nobody explicitly programmed.

Imagine drawing a few squares on a sheet of graph paper, applying four very simple mathematical rules, and watching the pattern apparently come alive.

Some shapes disappear almost immediately. Others remain unchanged indefinitely. Some alternate between different arrangements, while a few seem to travel across the paper as though they have somewhere to go.

Even more remarkably, certain patterns can produce an endless stream of moving structures.

There is no intelligence controlling them. No computer player is making decisions. Nobody has programmed the shapes to move, reproduce or interact.

Everything follows from four elementary rules.

Welcome to Conway's Game of Life, one of the most fascinating examples of how simple mathematics can produce extraordinary complexity.

It is also a wonderful way to introduce GCSE and A-level students to mathematics they are unlikely to encounter in their normal lessons.

1. A Mathematical Game Without Players

The Game of Life was developed by the British mathematician John Horton Conway and became widely known in 1970 through Martin Gardner's Mathematical Games column in Scientific American.

Despite its name, it is not really a conventional game.

There is no opponent, no score and no winning move.

Instead, it is what mathematicians call a cellular automaton.

That sounds complicated, but the underlying idea is remarkably straightforward.

Imagine a sheet of squared paper extending indefinitely in every direction.

Each square, or cell, has only two possible states:

• Alive — represented by a filled or coloured square.

• Dead — represented by an empty square.

Time advances in discrete steps called generations.

At each generation, every cell examines the eight squares surrounding it and changes according to the same rules.

The fascinating part is that the cells do not need to know anything about the rest of the grid.

They only respond to their immediate neighbours.

Yet when thousands of cells follow those rules simultaneously, remarkable structures emerge.

2. The Four Rules That Govern Life

Each cell has eight possible neighbours: above, below, left, right and the four diagonals.

The four rules are:

Rule 1: Underpopulation

A living cell with fewer than two living neighbours dies.

Rule 2: Survival

A living cell with two or three living neighbours survives into the next generation.

Rule 3: Overpopulation

A living cell with more than three living neighbours dies.

Rule 4: Reproduction

A dead cell with exactly three living neighbours becomes alive.

That is the entire mathematical system.

One particularly important detail is that all cells update simultaneously. We calculate the next generation from the existing one, rather than changing squares individually as we examine them.

Otherwise, the order in which we visit the cells would influence the result.

The rules are often abbreviated to B3/S23.

B3 means that a dead cell is born when it has exactly three living neighbours.

S23 means that a living cell survives when it has two or three living neighbours.

Nothing in these rules explicitly says that a structure should move, oscillate, grow indefinitely or behave like a miniature machine.

Nevertheless, all these things are possible.

3. Our First Experiment — Three Squares in a Row

We do not need a computer to start exploring.

Take some squared paper and colour three adjacent squares in a horizontal line.

Represent living cells by X and dead cells by a dot.

Generation 0

.....
.....
.XXX.
.....
.....

Apply the four rules to every square.

The cells at either end of the line have only one living neighbour, so they die.

The centre cell has two living neighbours and survives.

Meanwhile, the empty squares immediately above and below the centre each have exactly three living neighbours.

Both become alive.

Generation 1

.....
..X..
..X..
..X..
.....

Our horizontal line has become vertical!

Now apply the rules again.

Generation 2

.....
.....
.XXX.
.....
.....

We are back where we started.

The pattern will alternate between these two arrangements indefinitely, provided nothing else interferes.

This little structure is called a blinker.

It is an example of an oscillator, a mathematical pattern that repeats after a fixed number of generations.

The blinker has a period of two.

Period = number of generations required for a pattern to repeat.

Notice something else.

In every generation, the blinker has exactly three living cells.

Its position and total population remain unchanged, but its shape changes.

This gives students three different mathematical quantities to investigate: population, position and repetition period.

4. When Nothing Changes — Still Lifes

Not all patterns move or oscillate.

Try starting with four living squares arranged into a 2 × 2 block.

....
.XX.
.XX.
....

Each living cell has exactly three living neighbours.

Therefore, every cell survives.

No surrounding dead cell has exactly three living neighbours.

Consequently, no new cells appear.

The pattern remains unchanged forever.

This is called a still life.

It provides an excellent opportunity to ask students a question:

Can you design a different pattern that remains completely unchanged?

One example is the beehive:

......
..XX..
.X..X.
..XX..
......

Students can investigate whether every living cell has two or three living neighbours and whether any neighbouring dead cell has exactly three.

This turns an apparently recreational activity into an exercise in logical reasoning and proof.

Finding a stable pattern is one thing.

Proving that it is stable is rather more interesting.

5. The Glider — A Pattern That Travels

Now for something genuinely surprising.

Arrange five living cells as follows:

.X.
..X
XXX

This is called a glider.

Apply the rules four times.

The five-cell pattern will return to its original shape, but it will have moved one square diagonally downwards and to the right.

Continue the simulation and it will keep travelling.

The pattern does not literally contain the same five living cells throughout its journey. Cells die and new cells are born.

What moves is the arrangement of living cells.

This is rather like a wave travelling across water.

The water does not have to move along with the wave, but the pattern of disturbance does.

That distinction between the movement of matter and the movement of a pattern is a valuable idea in physics as well as mathematics.

How fast does a glider move?

The glider moves one square diagonally every four generations.

Its displacement per generation is therefore:

Horizontal displacement = 1/4 cell per generation.

Vertical displacement = 1/4 cell per generation.

Students familiar with Pythagoras' theorem can go further.

Diagonal displacement after four generations = sqrt(1² + 1²) = sqrt(2) cells.

Average diagonal speed = sqrt(2)/4, or approximately 0.354 cell widths per generation.

Strictly speaking, the glider's position is changing in discrete steps, not travelling continuously.

Nevertheless, we can use familiar ideas of displacement, direction and average speed to describe its behaviour.

This is a lovely example of GCSE mathematics appearing in an unexpected setting.

6. A Pattern That Produces Other Patterns

The glider is remarkable enough.

However, mathematicians discovered something even more extraordinary.

In 1970, Bill Gosper and colleagues at the Massachusetts Institute of Technology discovered a configuration now known as the Gosper glider gun.

It contains a structure that repeatedly produces gliders.

Every 30 generations, another glider emerges and travels away.

The original structure continues operating.

Given an infinite grid, an undisturbed glider gun can create an unlimited number of gliders over time.

That was mathematically significant because it demonstrated that a finite starting configuration could produce unbounded growth.

Conway had originally conjectured that this might be impossible.

The glider gun disproved that conjecture.

What I particularly like about this example is the way it illustrates the importance of experimentation in mathematics.

A mathematician can develop a hypothesis, explore examples, and sometimes discover a counterexample that overturns the original idea.

That is genuine mathematical investigation.

7. What Happens If We Start With Random Cells?

Instead of carefully designing a blinker or a glider, suppose we fill a large grid randomly.

Perhaps each square has a 30% probability of starting alive.

Initially, the pattern looks like scattered noise.

As generations pass, clusters form, collide and disappear.

Some settle into stable blocks.

Others become oscillators.

Occasionally, moving gliders escape from the surrounding activity.

What began as randomness can produce recognisable structure.

This is an example of emergent behaviour.

Emergence describes situations in which the interactions of relatively simple components produce complex large-scale behaviour.

We can find related ideas in many other fields:

• Flocks of birds, where individuals respond to nearby birds.

• Traffic flow, where simple driver interactions can produce congestion.

• Ant colonies, where local interactions contribute to organised activity.

• Crystal growth, where local physical processes generate large-scale patterns.

• Computer networks, where many independently operating devices form a larger system.

These are analogies, not examples of systems following Conway's exact rules.

Nevertheless, they raise an important question.

How much complexity can arise from simple local interactions?

For students who have been taught to associate mathematics primarily with equations and calculations, this can be a rather different way of thinking.

8. Can Mathematics Really Become Alive?

The title raises an interesting question.

Have we actually created life?

Not in the biological sense.

A glider does not have cells containing DNA. It does not metabolise, maintain a biological internal environment or evolve by natural selection merely because it moves.

The Game of Life is a mathematical model, not evidence that a computer simulation has become biologically alive or conscious.

Nevertheless, it demonstrates something important.

Apparently purposeful behaviour does not always require a central controller.

A glider looks as though it is travelling towards a destination.

A glider gun looks as though it is manufacturing objects.

Some elaborate configurations can even transmit information and perform computations.

Yet every event is determined by exactly the same four rules.

This brings us to a deeper question.

When we see complicated behaviour, how can we determine whether it arises from intelligence, design or simple underlying processes?

That is a question extending far beyond GCSE mathematics.

9. From Coloured Squares to Computers

One of the most extraordinary properties of Conway's Game of Life is that it is capable of universal computation.

With sufficiently elaborate configurations, its patterns can represent information and implement logical operations.

In principle, arrangements of cells can simulate a general-purpose computer, given enough time and space.

The idea is not that every glider is a little computer.

Rather, carefully constructed interactions between patterns can be used to implement computational mechanisms.

This connects Conway's creation with the work of Alan Turing and the foundations of computer science.

It also explains why the Game of Life appeals to students studying A-level Computer Science alongside Mathematics or Further Mathematics.

The connection is much deeper than simply writing a program to draw coloured squares.

The evolving squares themselves can be organised to perform computations.

The same mathematical world that produces a three-cell blinker can also support remarkably sophisticated logical structures.

10. A Classroom Investigation — Can You Predict the Future?

Here is an investigation I would enjoy giving to a student.

Start with the following L-shaped arrangement:

.....
.XX..
.X...
.....
.....

Without using a computer, predict the next generation.

Then predict the generation after that.

For the first generation, the correct result is a 2 × 2 block:

.....
.XX..
.XX..
.....
.....

The pattern then remains stable.

This is a useful exercise because it demonstrates that three living cells do not necessarily behave in the same way simply because the total population is identical.

Their arrangement matters.

Extension question: Can you find two different initial configurations with the same number of living cells that behave completely differently?

Try a horizontal line of three living cells and the L-shaped arrangement above.

One becomes a repeating oscillator.

The other becomes stable.

Same number of initial cells. Same rules. Different outcomes.

That is an important lesson about initial conditions and structure.

11. Building Your Own Game of Life in Excel

Students do not need sophisticated programming software to explore cellular automata.

Excel or another spreadsheet can be used.

A possible approach is to:

1. Create a 20 × 20 grid with living cells represented by 1 and dead cells by 0.

2. Calculate the number of living neighbours surrounding each cell.

3. Apply the Game of Life rules to calculate the next generation on a separate worksheet.

4. Use conditional formatting to colour living cells.

5. Repeat the calculation to observe how the pattern evolves.

For example, if a cell is located at C3 and its neighbours occupy the surrounding 3 × 3 range, the number of living neighbours can be calculated using:

=SUM(B2:D4)-C3

Subtracting C3 removes the central cell from the count.

The next generation must be calculated separately using the original values for all cells.

For an A-level student, the spreadsheet opens the door to investigating populations, proportions, repeated patterns and mathematical modelling.

There is an important practical detail: Excel grids are finite.

A limited grid needs clearly defined boundary conditions.

We might treat every cell outside the grid as dead, or construct a wrapping grid in which the left edge connects to the right and the top connects to the bottom.

These choices can change the behaviour of a pattern.

In other words, even the boundary conditions form part of the mathematical model.

12. An Even Better Challenge — Write the Program Yourself

For students studying Computer Science, writing a Game of Life simulator is a rewarding programming challenge.

Python is an obvious choice, although JavaScript, C# and many other languages would work.

The central algorithm is:

Create a grid
Choose which cells begin alive

Repeat:
    For each cell:
        Count its eight living neighbours

        If the cell is alive:
            Keep it alive if the count is 2 or 3
            Otherwise make it dead

        If the cell is dead:
            Make it alive if the count is 3
            Otherwise keep it dead

    Replace the old grid with the new grid
    Display the result

Notice that we must calculate the complete new grid before replacing the old one.

A common programming mistake is to change cells directly while counting their neighbours.

That introduces an unintended dependence on the order in which the cells are processed.

The challenge therefore teaches more than loops and conditional statements.

It introduces algorithms, arrays, state, simultaneous updating, debugging and software testing.

A particularly good test is to check whether the program correctly reproduces the blinker.

If it does not return to its original position after two generations, something is wrong.

13. A-Level Extension — Mathematics Behind the Simulation

For A-level students wanting more mathematical depth, there are several possibilities.

Population growth and decline

Let P(n) represent the number of living cells in generation n.

We can investigate how P(n) changes with time.

Does it:

• fall to zero?

• eventually become constant?

• fluctuate periodically?

• continue growing?

We can plot P(n) against n for different starting configurations.

A blinker has constant population:

P(n) = 3

A glider also has constant population:

P(n) = 5

However, a complicated initial configuration can produce a much less predictable population graph.

The important point is that a constant population does not necessarily imply a constant pattern.

Probability

Suppose every cell is independently alive with probability p.

For an interior cell on an initially random grid, the probability that exactly three of its eight neighbours are alive is:

P(exactly 3 neighbours) = 56 × p³ × (1 - p)⁵

The number 56 comes from choosing which three of the eight neighbours are alive:

8! / (3! × 5!) = 56

For a new live cell to appear, the central cell must also be dead.

Therefore:

P(birth at a cell) = (1 - p) × 56 × p³ × (1 - p)⁵

This simplifies to:

P(birth at a cell) = 56 × p³ × (1 - p)⁶

This formula describes the probability of a birth in the first update of the randomly initialised grid.

It does not automatically apply to every subsequent generation because neighbouring cells become statistically dependent as the simulation evolves.

That distinction is a valuable extension into probability and mathematical modelling.

Determinism

The Game of Life is deterministic.

If we start from exactly the same configuration and use the same rules and boundaries, the same sequence of generations will occur.

Yet complicated patterns may be extremely difficult to predict without actually running the simulation.

This illustrates the distinction between a system being deterministic and its long-term behaviour being easy to predict.

Students can investigate the sensitivity of particular patterns to small changes in their starting arrangements.

It is tempting to call every complicated outcome chaos, but mathematical chaos has a more specific meaning. Conway's Game of Life is best introduced as a deterministic discrete dynamical system with complex emergent behaviour.

14. Ten Mathematical Challenges for Students

For anyone who enjoys mathematical puzzles, here are ten investigations.

Challenge 1 — Extinction

Find the smallest non-empty starting pattern that completely disappears after one generation.

Challenge 2 — Stability

Construct a stable arrangement containing exactly six living cells.

Challenge 3 — The oscillator

Find a repeating pattern other than the blinker.

Challenge 4 — The glider

Demonstrate that a glider returns to its original shape after four generations, displaced diagonally.

Challenge 5 — Population

Find a pattern whose population increases during its first generation.

Challenge 6 — Initial conditions

Start with a blinker and add one extra living cell. Investigate how the behaviour changes depending on the position of that cell.

Challenge 7 — Collisions

Create two gliders travelling towards one another. What happens when they meet?

Challenge 8 — Randomness

Generate a 30 × 30 grid with 20% of its cells initially alive. Repeat with 40% and 60%. Compare the population after 50 generations.

Challenge 9 — Changing the rules

Modify the survival rule so that cells survive with two, three or four neighbours. How does the resulting world differ?

Challenge 10 — Design a machine

Investigate whether a system of moving patterns could be arranged to transport or represent a simple item of information.

These challenges range from accessible GCSE investigations to questions suitable for enthusiastic A-level Mathematics, Further Mathematics and Computer Science students.

Some can be answered using squared paper.

Others become much more interesting with a computer.

The important part is not simply obtaining an answer but explaining why the patterns behave as they do.

15. Why I Would Teach This at Hemel Private Tuition

Having taught mathematics and science for many years, I have found that some of the most memorable lessons begin with a question that does not initially look like an examination question.

Conway's Game of Life is particularly appealing because it is accessible without being trivial.

A GCSE student can understand the rules in minutes.

Within the same lesson, that student can explore coordinate grids, probability, pattern recognition, logical reasoning and mathematical proof.

An A-level student can go much further into discrete mathematics, combinatorics and dynamical systems.

A Computer Science student can create a working simulation, optimise the algorithm and experiment with more sophisticated cellular automata.

What makes this such an attractive teaching topic is that students can make their own discoveries.

They do not simply follow a worked example.

They ask questions, make predictions, test their ideas and discover that the results are not always what they expected.

There is also room for failure.

A carefully designed pattern may disappear after one generation.

Another may behave in a completely unexpected way.

That is not wasted effort. It is part of the investigation.

Although Conway's Game of Life is not generally a prescribed topic in GCSE or A-level Mathematics, exploring it helps develop skills that are valuable throughout those courses.

These include logical reasoning, clear explanations, systematic investigation, mathematical modelling and the ability to move between different representations of a problem.

At Hemel Private Tuition, I believe there is considerable value in occasionally going beyond the examination syllabus.

Examination success matters, but so does helping students recognise that mathematics is a creative and continually developing subject.

Sometimes the most effective way to build mathematical confidence is to let students explore something that does not have an obvious answer printed at the back of a textbook.

16. Try the Game of Life Yourself

The best introduction is to experiment.

Begin with three cells in a row.

Then try a square block.

Next, construct a glider.

Finally, create a random population and watch what happens.

Try making predictions before running the simulation.

An excellent starting point is the ConwayLife community website, which provides information, examples and links to simulation tools.

Useful resources include:

• ConwayLife: https://conwaylife.com/

• LifeWiki: https://conwaylife.com/wiki/Conway%27s_Game_of_Life

• Game of Life online simulator: https://playgameoflife.com/

For students who enjoy coding, the next challenge is to create a simulator rather than simply using one.

Conclusion — Complexity Does Not Always Need Complicated Rules

There is something wonderfully surprising about Conway's Game of Life.

Four simple rules applied to a collection of squares can produce stability, movement, repetition, interaction and structures capable of computation.

Nobody has to instruct the glider to travel.

Nobody tells the blinker to oscillate.

Nobody directs the glider gun to produce its next glider.

Those behaviours arise from the rules themselves.

That is one of the most important lessons mathematics can teach us.

Complex behaviour does not necessarily require a complex explanation at the level of individual components.

Sometimes extraordinary outcomes emerge from very simple foundations.

And perhaps the most exciting question is not whether Conway's mathematical cells are genuinely alive.

It is this:

If four simple rules can create such an astonishing mathematical world, what other remarkable things might we discover by experimenting with mathematics?

---

Beyond the Syllabus at Hemel Private Tuition

At Hemel Private Tuition in Hemel Hempstead, I teach GCSE Mathematics, A-level Mathematics, Further Mathematics and Computer Science, helping students develop both examination skills and a deeper understanding of their subjects.

My teaching combines worked examples, investigation, technology and practical demonstrations wherever these can help make difficult ideas clearer.

Lessons are available in person and online.

Mathematics is much more than passing examinations. It is learning to think, question and discover.

Philip M Russell — Hemel Private Tuition

https://www.hemelprivatetuition.uk

 

How Fast Does Sound Travel? Five Experiments to Measure the Speed of Sound

Can you measure something you cannot see? With a microphone, a tuning fork, a tube and some clever physics, the answer is yes.

Hemel Private Tuition | Practical Physics | GCSE and A-Level Physics

Sound Is All Around Us — But How Fast Is It Moving?

Imagine standing in a field and watching someone hit a metal object with a hammer.

You see the hammer strike before you hear the sound.

Or perhaps you have watched a thunderstorm. There is a brilliant flash of lightning, followed some seconds later by a rumble of thunder.

Why the delay?

The answer is that light and sound travel at enormously different speeds.

Light moves at approximately 300 million metres per second in a vacuum, whereas sound travels through ordinary air at roughly 340 metres per second.

That is an extraordinary difference.

But how do we know that sound travels at around 340 metres per second? More importantly, could we measure it ourselves?

At Hemel Private Tuition, I particularly enjoy experiments where students can discover a physical quantity rather than simply accepting a value printed in a textbook.

The speed of sound provides an excellent opportunity because there is more than one way to measure it.

We can use resonance, echoes, microphones, oscilloscopes and computer-based data logging. We can even investigate why different experimental methods sometimes give slightly different answers.

That final point is especially important for A-Level Physics.

A good experiment does not simply produce the expected answer. It produces evidence that we can analyse, question and improve.

1. What Actually Is Sound?

Before measuring the speed of sound, we need to understand what is moving.

Sound is a mechanical wave produced by vibrations.

When a loudspeaker vibrates, it causes neighbouring air molecules to move backwards and forwards. Those molecules interact with neighbouring molecules, passing the disturbance through the air.

The molecules themselves do not travel all the way from the loudspeaker to your ear. Instead, energy travels through the medium.

In air, sound waves are longitudinal waves.

This means the particles vibrate parallel to the direction in which the wave travels.

We describe the alternating regions as:

• Compressions: regions where the pressure and density are slightly higher.

• Rarefactions: regions where the pressure and density are slightly lower.

Sound requires a medium, such as a gas, liquid or solid. It cannot travel through a perfect vacuum.

Its speed depends upon the properties of that medium.

Sound generally travels much faster through water and steel than through air, although the precise values depend on temperature and material properties.

The fundamental wave equation

For a periodic sound wave:

Wave speed = frequency × wavelength

In symbols:

v = f × λ

Where:

• v = wave speed in metres per second (m/s)

• f = frequency in hertz (Hz)

• λ = wavelength in metres (m)

For example, suppose we produce a sound with a frequency of 500 Hz and measure its wavelength as 0.686 m.

Then:

v = 500 × 0.686

v = 343 m/s

That is close to the expected speed of sound in air at room temperature.

But the important question is how we can measure that wavelength — or alternatively measure how long sound takes to travel a known distance.

2. Experiment One — Measuring the Speed of Sound Using Resonance

This is one of the classic experiments in sound physics.

It is elegant, relatively inexpensive and gives students a direct connection between sound waves and standing waves.

Equipment

A resonance tube, water reservoir or adjustable piston, tuning forks of known frequency, metre rule, thermometer and optional microphone or sound-level sensor.

The principle

A tube containing air can resonate when the frequency of the sound matches one of its natural frequencies.

Consider a tube that is open at one end and effectively closed at the other by water.

At the water surface, the air has approximately a displacement node.

At the open end, it has approximately a displacement antinode.

For the lowest resonant frequency, the effective length of the air column is approximately one-quarter of a wavelength.

Therefore:

L = λ / 4

Rearranging:

λ = 4L

And:

v = 4fL

This simplified equation ignores a small correction associated with the open end of the tube.

Practical procedure

1. Fill the resonance apparatus with water, leaving a short air column.

2. Strike a tuning fork gently with a suitable rubber striker.

3. Hold the vibrating fork just above the opening without touching the tube.

4. Gradually lower the water level, increasing the length of the air column.

5. Listen for the point at which the sound becomes noticeably louder.

6. Measure the length of the air column.

7. Repeat using different tuning forks.

At resonance, the sound will become noticeably more intense.

This is not because the tuning fork suddenly begins vibrating faster. The air column is responding particularly strongly to the frequency being produced.

Worked example

Suppose a tuning fork has a frequency of 512 Hz.

The first resonance occurs when the measured air-column length is 0.160 m.

Ignoring the end correction:

λ = 4 × 0.160 = 0.640 m

v = 512 × 0.640

v = 327.7 m/s

The measured value is slightly lower than the expected value of approximately 343 m/s at 20°C.

Why?

One important reason is that the displacement antinode is actually located a little beyond the open end of the tube.

The effective resonating length is therefore slightly longer than the measured tube length.

An A-Level improvement — use two resonances

A particularly good method is to find two consecutive resonance positions.

For a tube closed at one end:

First resonance: L1 + e = λ / 4

Second resonance: L2 + e = 3λ / 4

Here, e represents the end correction, assuming it remains approximately constant.

Subtracting the equations gives:

L2 - L1 = λ / 2

Therefore:

λ = 2(L2 - L1)

And:

v = 2f(L2 - L1)

This method has an important advantage.

The end correction cancels out.

Suppose the first resonance occurs at 0.160 m and the second at 0.495 m.

The difference is:

0.495 - 0.160 = 0.335 m

Therefore:

λ = 2 × 0.335 = 0.670 m

v = 512 × 0.670

v = 343.0 m/s

An excellent result.

These are illustrative measurements rather than actual laboratory observations, but they demonstrate how using two resonance positions can improve the method.

What students learn

This experiment brings together frequency, wavelength, standing waves, nodes, antinodes, resonance and experimental corrections.

It is also an opportunity to discuss why taking more measurements does not necessarily remove a systematic error.

3. Experiment Two — Two Microphones and the Time Taken for Sound to Travel

This is one of my favourite ways of introducing modern experimental physics.

Instead of measuring a wavelength, we measure a time interval.

The idea is remarkably simple.

Place two microphones a known distance apart. Produce a short, sharp sound near the first microphone. Measure how long the sound takes to reach the second microphone.

Then calculate the speed.

Equipment

Two microphones, a dual-channel oscilloscope or simultaneous-channel audio recording interface, a metre rule, connecting cables, and a sharp sound source such as a finger snap or hand clap.

Setting up the experiment

Imagine two microphones arranged along a straight line.

The first microphone is positioned near the sound source.

The second microphone is placed exactly 1.00 m farther away.

When a sound pulse is produced, the first microphone detects it before the second.

The oscilloscope displays two signals with a small difference in arrival time.

The relationship is:

v = d / t

Where:

• d = distance between microphones

• t = difference between arrival times

• v = speed of sound

Worked example

Distance between microphones = 1.00 m

Time delay = 0.00292 s

Therefore:

v = 1.00 / 0.00292

v = 342.5 m/s

This is close to the expected value at room temperature.

Why use an oscilloscope?

An oscilloscope makes extremely short time intervals visible.

A delay of just three milliseconds would be difficult to measure directly with a stopwatch.

However, on an oscilloscope it becomes the horizontal separation between two waveform features.

Students can measure the time difference using cursors.

With a dual-channel instrument, both signals must share a reliable common timebase.

This experiment is also used in university teaching demonstrations, including an example documented by the University of Maryland.

lecdem.physics.umd.edu

Making the experiment more accurate

I would encourage students to repeat the experiment at several different microphone separations.

For example:

Separation (m)	Illustrative delay (ms)
0.50	1.46
1.00	2.92
1.50	4.37
2.00	5.83

If students plot separation (vertical axis) against time delay (horizontal axis), the gradient gives the speed of sound.

Gradient = change in distance / change in time

This is often better than relying upon one result.

It also introduces a crucial A-Level concept: a straight-line graph can identify the physical quantity we want while helping reveal systematic offsets.

A subtle problem

What happens if the microphone signals are not processed at exactly the same time?

Some recording systems introduce delays between channels.

If the channels are not synchronised, students could mistake an electronic delay for the travel time of sound.

For a sound travelling just one metre, even a small timing offset may create a significant error.

I would use a single synchronised two-channel system, where possible, and reverse the microphones as a useful check. The sound source should remain collinear with the microphones, and the pulses must not be so loud that they cause clipping.

The apparent simplicity of this experiment makes it ideal for discussing the difference between a straightforward formula and a genuinely reliable measurement.

4. Experiment Three — Measuring the Speed of Sound Using an Echo

An echo is sound that has travelled to a reflecting surface and returned to the listener.

It gives us another way to measure the speed of sound.

If sound travels a distance d to a wall and then returns, its total distance is 2d.

Therefore:

v = 2d / t

Where t is the total round-trip time.

The outdoor experiment

Imagine standing 50 m from a large, flat wall.

You produce a sharp sound and record the interval until its echo returns.

Suppose the time interval is 0.292 s.

The sound travels:

2 × 50 = 100 m

Therefore:

v = 100 / 0.292

v = 342.5 m/s

A pleasingly simple experiment.

However, a stopwatch is not an ideal instrument for measuring such a short interval.

Human reaction times are much too significant.

Instead, we can use a microphone and audio-recording software.

The waveform should show an initial pulse followed by a reflected pulse.

The separation between these events gives the echo delay.

An indoor laboratory version

We can reproduce the same principle using a long tube with a closed end.

Place a microphone close to the open end, pointing along the tube.

Produce a short sound pulse near the microphone.

The pulse travels down the tube, reflects at the closed end and returns.

Using a tube length of 1.50 m, the approximate return time would be:

t = 2 × 1.50 / 343

t = 0.00875 s

That is 8.75 milliseconds.

With suitable data logging, this is easy to display.

PASCO has published experiments using precisely this general technique, including configurations with a sound sensor, a closed tube and Capstone software.

PASCO

+1

The challenge of reflections

In a real laboratory, sound may reflect from walls, benches, equipment and other objects.

The microphone may detect several overlapping pulses.

Students must identify which signal is the required echo.

This is useful experimental problem-solving: the apparatus may be working perfectly, yet interpreting the signal still requires care.

For an outdoor echo experiment, choose a safe location away from roads and avoid very loud impulsive sounds. A computer-generated click through a modest loudspeaker can be a more controlled source.

5. Experiment Four — Using an Oscilloscope to Measure Wavelength

There is another approach that combines two microphones with a continuous sound of known frequency.

Instead of recording a single pulse, we use a steady tone.

A loudspeaker connected to a signal generator produces a sound at a chosen frequency.

Two microphones detect the resulting wave.

As one microphone is moved, its signal changes phase relative to the other.

When the signals return to the same phase, the microphone has moved by one whole wavelength, provided we follow successive alignments.

The procedure

1. Generate a continuous tone, perhaps at 1,000 Hz.

2. Place two microphones along the direction of propagation.

3. Connect them to separate oscilloscope channels.

4. Keep one microphone fixed.

5. Move the other microphone slowly away.

6. Identify successive positions where the waveforms are aligned in phase.

7. Measure the separation between these positions.

At 1,000 Hz and a sound speed of 343 m/s:

λ = v / f

λ = 343 / 1000

λ = 0.343 m

So the distance between successive in-phase positions should be approximately 34.3 cm.

The experiment demonstrates that sound behaves as a wave rather than merely as a sequence of isolated events.

A useful extension

Repeat the experiment at different frequencies.

Students should discover that, in ordinary air under the same conditions, the speed remains approximately constant while wavelength decreases as frequency increases.

A 500 Hz tone has approximately twice the wavelength of a 1,000 Hz tone.

That provides a convincing practical demonstration of:

v = f × λ

There are some practical complications. Reflected waves can disturb the phase measurements, and moving a microphone sideways rather than along the sound path will create misleading results.

A directional sound source, careful alignment and a suitable distance from reflective surfaces all help.

6. Experiment Five — Using PASCO Sensors and Data Logging

In my home laboratory, I use PASCO equipment because it provides opportunities to explore measurements in much greater detail.

A major benefit of computer-based data logging is that it allows the student to inspect what actually happened during an experiment.

Instead of simply writing down one number, we can examine the waveform itself.

PASCO Capstone can display sound waveforms in a scope-style view. PASCO also provides published methods for measuring sound speed from reflected pulses, including a procedure sampling at 20 kHz.

PASCO Help Center

+1

A suggested PASCO investigation

I would use a sound sensor and an adjustable closed-end tube to perform an echo experiment.

The procedure would be:

1. Measure the tube length.

2. Position the microphone close to the open end.

3. Set an appropriately high sampling rate.

4. Record a short impulse and the returning reflection.

5. Use the software cursors to measure the time interval.

6. Calculate the speed of sound.

7. Change the tube length and repeat.

Why sampling rate matters

Suppose the equipment records 20,000 measurements per second.

The interval between samples is:

1 / 20,000 = 0.00005 s

This is 50 microseconds.

During that time, sound travelling at 343 m/s moves approximately:

343 × 0.00005 = 0.01715 m

That is around 1.7 cm.

It means choosing a waveform peak only to the nearest sample can introduce appreciable timing uncertainty, particularly over short distances.

Students can investigate whether increasing the tube length reduces the percentage uncertainty.

They can also examine whether the waveform provides a sharply defined event or a less obvious series of peaks.

One caution: not every PASCO sound sensor or interface supports simultaneous high-speed recording from multiple microphones. The actual hardware specifications and recording mode must be checked before attempting a two-microphone time-of-flight experiment.

For this reason, a single-sensor echo method is often the more straightforward starting point.

7. Does Sound Always Travel at the Same Speed?

Here is a fascinating question to ask a student:

Would sound travel faster through the air in our laboratory on a hot summer afternoon or on a cold winter morning?

The answer is that sound travels faster through warmer air.

For ordinary temperatures, we can use the approximate relationship:

v = 331 + 0.6T

Where:

• v = speed of sound in m/s

• T = air temperature in degrees Celsius

This is a useful approximation for air under normal conditions.

Physics Classroom

+1

How the speed of sound changes with temperature

Approximate theoretical values for air, calculated using v = 331 + 0.6T.

325 m/s332.5 m/s340 m/s347.5 m/s355 m/s051015202530

At 0°C:

v = 331 + (0.6 × 0) = 331 m/s

At 20°C:

v = 331 + (0.6 × 20) = 343 m/s

At 30°C:

v = 331 + (0.6 × 30) = 349 m/s

That difference is measurable.

Why does temperature matter?

Increasing temperature increases the average kinetic energy of the air molecules.

The relationship between temperature and the speed at which pressure disturbances propagate can be described using the properties of a gas.

At A-Level, we can go further:

v = sqrt(γRT / M)

Where:

• γ = ratio of specific heat capacities

• R = molar gas constant

• T = absolute temperature in kelvin

• M = molar mass of the gas

For a given ideal gas composition, the speed increases in proportion to the square root of absolute temperature.

This also explains why we should measure the laboratory temperature whenever we undertake a speed-of-sound experiment.

Comparing an experimental value directly with 343 m/s without checking the temperature is not always appropriate.

A further investigation

Students could repeat one of the experiments on different days and compare the results against the predicted temperature dependence.

For an indoor experiment, this is preferable to deliberately heating air in an open laboratory, where temperature gradients and convection might introduce further complications.

Humidity also affects sound speed, although temperature is generally the dominant factor in ordinary school experiments.

8. Which Experiment Gives the Best Answer?

This is where the experiments become particularly valuable for A-Level students.

Different methods may produce different results even when measuring the same physical quantity.

Consider the following fictional set of results, illustrating what students might obtain at a laboratory temperature of 20°C.

Method	Measured speed	Difference from 343 m/s
Resonance, one position	328 m/s	-4.4%
Resonance, two positions	343 m/s	0.0%
Two microphones	342 m/s	-0.3%
Echo method	346 m/s	+0.9%
Phase and wavelength	339 m/s	-1.2%

These are illustrative results, not measurements from my laboratory.

The important question is not simply which experiment produced a value closest to 343 m/s.

We should ask why the answers differ.

Random errors and systematic errors

Random errors cause repeated measurements to vary.

For example, students may find it difficult to identify the exact resonance position because the sound becomes louder gradually rather than suddenly.

Systematic errors can shift every measurement in one direction.

Examples include:

• Ignoring the end correction in a resonance tube.

• Using microphones with a fixed electronic timing offset.

• Measuring the wrong acoustic path.

• Selecting a reflected waveform instead of the direct sound.

• Using an assumed temperature rather than the measured air temperature.

Repeating measurements can help reduce the influence of random errors, but it will not automatically eliminate systematic errors.

This distinction is frequently important in GCSE and A-Level experimental questions.

Calculating percentage difference

We can compare an experimental result against a predicted value using:

Percentage difference = |measured value - predicted value| / predicted value × 100

If the measured speed is 339 m/s and the predicted value is 343 m/s:

Percentage difference = |339 - 343| / 343 × 100

Percentage difference = 1.17%

However, a small percentage difference does not automatically mean a measurement is reliable.

We must also consider its uncertainty.

For example, a result of 343 ± 15 m/s is less precise than one of 341 ± 2 m/s, even though the first happens to agree exactly with our expected value.

That is a distinction I particularly want students to understand.

9. Measuring Uncertainty — An A-Level Challenge

Suppose students use two microphones separated by:

d = 1.000 ± 0.005 m

They measure a time interval of:

t = 2.92 ± 0.05 ms

Their calculated speed is:

v = 1.000 / 0.00292 = 342.5 m/s

For independent uncertainties in distance and time, a first-order estimate of the fractional uncertainty is:

Δv / v = sqrt[(Δd / d)^2 + (Δt / t)^2]

Substituting:

Δv / v = sqrt[(0.005 / 1.000)^2 + (0.05 / 2.92)^2]

Δv / v = 0.0178

This corresponds to approximately 1.8%.

The absolute uncertainty is:

Δv = 342.5 × 0.0178

Δv = 6.1 m/s

So we might report:

v = 343 ± 6 m/s

This uncertainty includes only the quoted distance and timing contributions. It does not account for any unrecognised systematic errors.

For a conservative school-level approach, adding the percentage uncertainties gives a larger estimate of about 2.2%.

Either way, students begin to see that reporting a measurement properly is more than writing down the result of a calculation.

10. Can We Combine the Experiments into One Investigation?

This is the experiment I would be particularly interested in running at Hemel Private Tuition.

Rather than carrying out each method in isolation, I would organise a comparative investigation.

Students would measure the speed of sound using three approaches:

• Resonance in an air column.

• Time of flight between two microphones.

• The delay between an initial sound pulse and its echo.

All three methods would be used under approximately the same laboratory conditions.

We would measure the air temperature, repeat each experiment and record the uncertainty.

Then we would compare the findings.

Interactive experiment comparison

Select a method to see what students measure and the main source of uncertainty.

Resonance

Microphones

Echo

What to measure

Frequency and the distance between successive resonance positions.

Equation

v = 2f(L2 - L1)

Main uncertainty

Identifying resonance positions, frequency uncertainty and acoustic end effects.

An investigation suitable for GCSE and A-Level

At GCSE level, students could focus on calculating speed, drawing graphs and explaining differences between results.

At A-Level, I would extend the exercise to include:

• Uncertainty propagation.

• Distinguishing precision from accuracy.

• Identifying systematic errors.

• Using best-fit lines and gradients.

• Evaluating assumptions about wave behaviour.

• Deciding whether different results agree within their uncertainties.

It is an excellent example of how one physical concept can support several levels of teaching.

11. From the Laboratory to the Real World

The speed of sound is much more than a textbook number.

Thunderstorms

When we see lightning, the light arrives almost instantaneously compared with the sound.

At approximately 343 m/s, sound travels about one kilometre in three seconds.

This gives us a rough method of estimating the distance to a lightning discharge.

However, this is not a reliable safety test. A storm can still be dangerous even if the thunder appears distant.

Medical ultrasound

Ultrasound scanners transmit high-frequency sound waves into the body and detect reflections from boundaries between tissues.

By measuring the time taken for reflected pulses to return, an instrument can estimate depth.

The principle resembles our laboratory echo experiment, although the speed of sound within tissue is different from that in air.

Sonar

Boats and ships use sound waves to investigate what lies beneath the water.

A sonar system sends out a pulse and measures the return time.

Depth is calculated using the round-trip distance and the appropriate sound speed in water.

The speed varies with temperature, salinity and pressure.

Music and musical instruments

Wind instruments depend upon standing waves and resonance.

The length of an air column helps determine the frequencies at which it resonates.

The same physics that explains our resonance tube helps us understand organ pipes, flutes and clarinets.

As someone interested in both physics and music, I find this connection particularly satisfying.

The mathematics of waves is not confined to the science laboratory. It is fundamental to the sounds we hear in music every day.

12. Why I Believe Students Should Carry Out These Experiments

After many years of teaching, I have become increasingly convinced that there is an important difference between recognising a scientific equation and understanding what that equation means.

A student might remember:

Speed = distance / time

They might also remember:

Wave speed = frequency × wavelength

And they might be perfectly capable of using both equations in an examination.

But ask them to measure the speed of something they cannot see, and the problem becomes much more interesting.

How do we know where the sound is?

How can we measure a time interval of a few milliseconds?

Why does a tuning fork become louder at certain air-column lengths?

Why do two apparently correct methods give different answers?

What does it mean when our data do not agree with the textbook?

These are the questions that turn a practical into a genuine learning experience.

One advantage of teaching in a well-equipped home laboratory is the opportunity to move between traditional experiments and modern instruments.

A tuning fork and resonance tube can sit alongside an oscilloscope, computer-based data logger and PASCO sensors.

Students can then see how different measuring techniques reveal different aspects of the same physical phenomenon.

I would rather have a student obtain a value of 336 m/s, recognise why it might be inaccurate and suggest a sensible improvement than produce 343 m/s without understanding how the measurement was made.

The first student is beginning to think like an experimental physicist.

13. Bringing Practical Physics to GCSE and A-Level Tuition

At Hemel Private Tuition in Hemel Hempstead, I teach GCSE and A-Level Physics with a strong emphasis on practical understanding.

My laboratory allows students to investigate scientific principles using real equipment rather than relying exclusively on diagrams, simulations or videos.

We can combine experiments with digital data logging, graph analysis, worked calculations and examination-style questions.

Online students can also benefit from live demonstrations using multiple cameras, close-up views of the equipment and screen sharing of experimental data.

The intention is not simply to make Physics entertaining.

It is to help students connect scientific theory with experimental evidence and develop the confidence to tackle unfamiliar examination questions.

For A-Level students in particular, the ability to interpret data and evaluate experimental methods is just as important as remembering formulae.

Conclusion — Measuring the Invisible

Sound travels through the air around us every moment of the day.

We hear it, recognise it and use it to communicate.

Yet the disturbance itself is invisible.

With a tuning fork and a tube, we can measure its wavelength.

With two microphones, we can measure how long it takes to travel a known distance.

With an echo, we can investigate the journey of a sound pulse to a reflecting surface and back.

And with an oscilloscope or data logger, we can make the behaviour of sound visible on a screen.

The fascinating part is that these experiments use different methods but should lead us towards the same physical answer.

They also teach something more valuable than a single numerical value.

They demonstrate that science depends upon measurement, evidence, uncertainty and the willingness to question our results.

The speed of sound may be approximately 343 metres per second at room temperature. But discovering how to measure it is where the real learning begins.

For GCSE and A-Level Physics tuition, including practical laboratory work, visit Hemel Private Tuition.