Before you can plan any experiment, you need to ask: "What am I actually measuring, and what tool measures that?" Every physical quantity — length, mass, time, angle, current — has a standard instrument that goes with it. Think of apparatus selection like picking the right kitchen tool: you wouldn't use a tablespoon to measure a teaspoon of salt, and you wouldn't use a metre ruler to measure the diameter of a human hair.

The apparatus you need falls into two categories:

• What you are measuring — e.g. the block of metal, the wire, the beaker of liquid.
• How you are measuring it — e.g. the ruler, the thermometer, the ammeter.

Common Apparatus & What They Measure

Worked Example: Specific Heat Capacity Setup

To measure the specific heat capacity of an aluminium block, you'd need:

• An aluminium block (ideally 1 kg) or a beaker of fluid with known mass
• A thermometer (to measure temperature rise)
• An immersion heater (to supply heat)
• A power source, plus a voltmeter, ammeter, and stopwatch (or a joulemeter)
• Wires and connectors

Notice how the list includes even the "boring" parts — wires and connectors. Examiners want you to think of the entire circuit, not just the star apparatus.

Every instrument has two defining properties, and mixing them up is one of the most common exam mistakes:

Think of range as "how far the ruler stretches" and resolution as "how fine the marks on it are." A metre ruler has a huge range (0–1 m) but a coarse resolution (1 mm) — it simply can't detect a change smaller than a millimetre, no matter how carefully you look. A micrometer has a tiny range (0–25 mm) but an incredibly fine resolution (0.01 mm) — perfect for small, precise objects but useless for measuring your desk.

Worked Example

Two digital thermometers display: Thermometer 1: 80.13 °C. Thermometer 2: 42.0 °C. Which has the better resolution?

Answer: Resolution = smallest increment the device can read. Thermometer 1 reads to 0.01 °C; Thermometer 2 reads to 0.1 °C. Since 0.01 °C is a finer (smaller) increment, Thermometer 1 has the better resolution.

Calibration is comparing a known measurement (something you're certain about) against what the instrument actually reads, so you can correct for any inaccuracy. It's like checking your kitchen scale by weighing a bag of sugar that says "1 kg" on the packet — if the scale reads 1.05 kg, you know it's off by 5%, and you can adjust future readings accordingly.

A simple everyday example: checking a voltmeter or ammeter reads exactly zero before you start taking measurements. If it doesn't, you have a zero error, and every single reading you take afterwards will be shifted by that same amount.

Calibrating a Thermometer

To calibrate a thermometer, you place the correct marks at the correct positions using two known reference points — typically the freezing point of water (0 °C) using melting ice, and the boiling point of water (100 °C) using boiling water. Once these two fixed points are marked correctly, everything in between can be divided evenly, and the thermometer becomes trustworthy across its whole range.

Calibration Curves

Some instruments don't give a reading that's proportional to the thing they're measuring — a thermocouple's e.m.f. doesn't rise in a straight line with temperature, and neither does a thermistor's resistance. In these cases, you can't just read the value directly; you plot a calibration curve (a graph of the sensor's output against the known true value), and then use that curve to convert any future reading back into the actual physical quantity.

Worked Example

A voltmeter always reads higher than the true value, and the size of this error grows as voltage increases. What does its calibration curve (true value vs. meter reading) look like?

Answer: Since the meter always overreads, at true value = 0 the meter would still show something above zero — so the line does not pass through the origin. And because the error grows with voltage, the gap between the "true value" line and what the meter shows must widen as voltage increases (not stay constant). This rules out a simple straight line through the origin, and instead gives a curve that starts above zero and bends away from the 1:1 line as values increase.

This section is essentially the practical partner to Section 1 — it's about matching the correct instrument and technique to whatever variable you need. The golden rule: the resolution of your instrument must be fine enough to detect meaningful changes in what you're measuring. If it isn't, your data will look "stepped" or won't show any variation at all, even if the real physical quantity is genuinely changing.

This is arguably the most exam-heavy section in the whole chapter, so let's be precise about the three types of variable in any experiment:

Here's the intuition: imagine testing whether temperature affects how fast a chemical reaction happens. If you also accidentally change the concentration of your reactants between trials, you can no longer say "temperature caused this change" — because concentration might have caused it instead. That's why controlled variables matter so much: without them, you cannot draw a valid conclusion, even if your results look neat and consistent.

Worked Example: Ideal Gas Laws

An ideal gas has four properties that can be measured/changed: temperature, pressure, volume, and number of moles. The key rule is that only one of these should ever be deliberately changed at a time:

If you failed to keep pressure constant while testing Charles' Law, and the volume changed, you'd have no way of knowing whether that change was due to temperature or the sneaky change in pressure — the results become unreliable and the conclusion is invalid.

No single measurement is perfect — human reaction time, tiny fluctuations in equipment, and small environmental changes all introduce a bit of randomness. Taking repeat readings and averaging them smooths out this randomness, making your data more reliable (i.e. you'd get similar results if you repeated the whole experiment again).

• The spread of repeat readings should be as small as possible — this shows the measurements are precise, not just luck.
• Repeats also help you spot an anomalous result — a value that clearly doesn't fit the pattern of the rest, which you can then investigate or discard.
• Always leave space in your results table for repeats, and always calculate the average value from them to use in further calculations (like plotting a graph).

When Repeats Aren't Straightforward

It's tempting to think "just repeat everything 3–5 times" is always the right answer, but that's not always practical:

• Heating effects: if a component heats up during a reading (like a wire carrying current), you can't just immediately repeat it — you'd need to disconnect it, wait for it to cool, and then retake the reading. This takes far longer than the original experiment.
• Time-dependent variables: if you're measuring something that depends on a specific time of day (e.g. background light levels between 6am–7am), you can't simply repeat it minutes later — you'd have to wait until the same time the next day, which may not fit your timeframe.
• Fatigue/human factors: if a person is the "instrument" (e.g. running up stairs to measure power output), tiredness changes their performance between repeats, so you'd need a proper rest period first.

Safety questions in exams are usually worth just 1 mark, but that mark is easy to lose if you only state the hazard without explaining how it's dealt with. Always follow the pattern: hazard → consequence → precaution.

Masses, Wires & Springs

• Wear safety goggles when a wire might snap under tension.
• Clamp or weight down support stands (commonly with a G-clamp) so they don't tip over.
• Don't overload springs beyond their elastic limit — they could snap while oscillating.
• Place a soft surface (like a cushion) beneath anything that could fall, e.g. a ball bearing when measuring g.

Electrical Circuits

• Keep liquids away from the apparatus — a spill near live equipment could cause a fire.
• Turn off the power supply between readings so components (especially thin wires) don't overheat — this prevents burns AND stops temperature-related resistance changes from skewing your results.
• Never exceed the voltage rating of any component in the circuit.

General Lab Practice

• Keep bags and chairs tucked under desks to avoid trips.
• Stand up during the experiment so you can react quickly if something goes wrong.
• Don't eat or drink while conducting experiments.
• Wear appropriate clothing/lab coat; tie back long hair.
• Mop up spills immediately to prevent slips.
• Never leave apparatus unsupervised — especially Bunsen burners.
• Let hot glass cool, or use gloves (e.g. after specific heat capacity experiments).
• Report spills or equipment problems to a supervisor as soon as possible.

Raw results are rarely exam-ready. Often you need to process them — using standard form, rounding consistently, calculating averages, or rearranging a physical law into the shape of a straight line so you can extract useful information (like a constant) from a graph's gradient or intercept.

Using Standard Form & Prefixes

Physical quantities are often huge or tiny (e.g. the speed of light = 3.00 × 10⁸ m s⁻¹), so standard form avoids writing out long strings of zeros. You should also know the common prefixes:

Significant Figures

Calculations must be reported to an appropriate number of significant figures, and crucially, every value in a data column should be quoted to the same number of significant figures — mixing 22.0, 39.5, and 60 in the same column is inconsistent and will be marked down, even if the numbers themselves are "correct."

Calculating Mean Values

Special rule: when calculating a mean, it's acceptable to increase the number of significant figures by 1 compared to the raw data — this is one of the few places in physics where you're allowed to "gain" precision, because averaging genuinely does improve your estimate.

Turning Equations into Straight Lines

This is one of the most powerful techniques in A-Level Physics: if you suspect two quantities follow a particular relationship, you can rearrange the physics equation to match the format of a straight line, then plot a graph to test it.

If your data produces a straight line through the origin when plotted this way, it confirms the relationship you were testing.

Worked Example: Verifying the Inverse Square Law

The inverse square law for gamma radiation states intensity I ∝ 1/x². Since intensity is proportional to corrected count rate C, we get:

This means plotting 1/√C (y-axis) against x (x-axis) should give a straight line through the origin, where the gradient equals the constant k. Steps to do this in practice:

1. Take repeat count-rate readings at several distances, and calculate an average for each.
2. Measure background radiation separately (with no source present) and subtract it from every reading to get the "corrected" count rate, C.
3. Calculate 1/√C for each corrected value.
4. Plot 1/√C against x and draw a line of best fit.
5. If the line is straight and passes through the origin, the inverse square law is confirmed.