Think of apparatus selection as matching a job to the right tool. Just like you wouldn't use a sledgehammer to hang a picture frame, you wouldn't use a metre ruler to measure the diameter of a wire (0.2 mm — you'd barely see it against the ruler's 1 mm markings, let alone read it accurately).

Apparatus falls into two jobs: what you're measuring (the thing being investigated — e.g. an aluminium block) and how you're measuring it (the instruments — thermometer, ammeter, voltmeter, etc).

These two words get mixed up constantly, so lock in the difference now:

• Range — the highest and lowest value the instrument measure. Think "how far it stretches."
• Resolution — the smallest change the instrument can detect. Think "how fine its teeth are."

A bathroom scale has a huge range (0–150 kg) but poor resolution (nearest 0.5 kg) — great for weighing yourself, useless for weighing a letter for postage.

Calibration means comparing your instrument's reading against a known, trusted value. If a thermometer says a boiling kettle is at 94°C, but you know water boils at 100°C (at standard pressure), the thermometer is wrong by a consistent 6°C — and calibration is how you'd catch and correct that.

The simplest form of calibration is a zero check: before taking readings, confirm a meter reads exactly zero with no input. If it doesn't, you have a "zero error," which is a type of systematic error.

To calibrate a thermometer properly, you mark 0°C at the freezing point of pure water (melting ice) and 100°C at the boiling point of water, then divide the space between into equal degree markings.

iCalibration Curves

Some sensors (like thermistors or thermocouples) don't give a reading directly in the unit you want — a thermocouple outputs a voltage (e.m.f.), not a temperature. A calibration curve is a graph that lets you convert the sensor's raw output into the real physical quantity, by plotting the sensor's output against a known reference.

This topic is really an extension of Topic 1 and 2 combined into a decision-making skill: given a specific measurement task, pick the instrument whose resolution is appropriate for the size of the thing you're measuring.

Every experiment has three types of variable, and mixing them up is one of the most common ways to lose marks:

• Independent variable — the ONE thing you deliberately change.
• Dependent variable — the thing you measure as a result (the outcome).
• Control variables — everything else that could affect the result, which you must keep constant.

A fair test is one where only the independent variable is allowed to influence the dependent variable. If you don't control the other variables, you can't be sure it was your independent variable causing the change you saw — the result becomes unreliable.

iWorked Example: Charles' Law vs Boyle's Law

Notice: in an ideal gas investigation there are four possible factors (temperature, pressure, volume, moles), but only one is ever changed deliberately at a time — everything else must be locked down.

Taking repeat readings and averaging them reduces the effect of random error, making your data more precise (small spread) and more reliable (trustworthy, not a fluke). Repeats also let you spot an anomaly — a result that clearly doesn't fit the pattern of the others, which you'd typically exclude before averaging.

But repeats aren't automatically the right call for every experiment. Ask:

This is the easiest topic to lose marks on despite being the "simplest" — because examiners want a hazard + a specific fix, not just "be careful."

Once you've collected raw readings, you often need to process them before they're useful — usually by getting your data into the form of a straight line, because straight-line graphs are the easiest way to test a relationship and extract meaningful constants.

iStandard Form & Prefixes

Standard form avoids writing out long strings of zeros — e.g. the speed of light is 3.00 × 10⁸ m s⁻¹ rather than 300,000,000 m s⁻¹.

iiSignificant Figures & Means

Every value in a single data column should be quoted to the same number of significant figures — mixing 22.0, 39.5, 60, 81.44, 100 in one column (as in the chapter's example) looks sloppy and inconsistent. Round them all to match, e.g. 20.0, 40.0, 60.0, 80.0, 100.0.

When you calculate a mean from repeat readings, you're allowed to quote it to one extra significant figure beyond your raw data — because averaging genuinely does improve your precision slightly.

iiiGraph Skills

• Recognising when a relationship fits y = mx + c so gradient/intercept can be analysed
• Finding the area under a graph (including non-linear curves, by estimation)
• Using and interpreting logarithmic plots
• Drawing tangents to curves and calculating their gradient
• Understanding when asymptotes are needed

This is probably the most conceptually important — and most confused — topic in the chapter. Let's build it up carefully.

Reducing uncertainty — three practical strategies:

• Take repeat readings (3–5) and calculate the mean
• For non-uniform objects (like a wire), measure at multiple points to check uniformity
• Use the instrument with the appropriate resolution for the size of the quantity (don't use a ruler for millimetre-scale gaps)

iSystematic Errors

A systematic error comes from a faulty instrument or a flawed method, and — crucially — it's : it shifts every single reading by roughly the same amount, in the same direction. This consistency is exactly why it shows up so clearly on a graph.

Reducing systematic errors: recalibrate the instrument (or swap it for a different one), or adjust/correct the technique being used.

Physics doesn't happen in a vacuum (pun intended) — every experiment or new technology has knock-on consequences beyond the lab. Examiners expect you to recognise four core categories, often abbreviated C.L.E.S.:

Beyond these four, also consider economic factors (is the research worth the cost vs. other priorities like schools/healthcare — but weighed against long-term benefits like climate mitigation) and environmental factors (impact on wildlife/geography — e.g. wind turbines are clean energy but can harm birds and bats).