Practical skills for familiar and unfamiliar experiments

Highlights only0 highlighted

Physics is built on measurement, so the examiners test whether you can plan, carry out and judge an experiment, even though there is no practical exam.

The sixteen core practicals give you experience of particular experiments, but the questions also give you set-ups you have never seen and ask what you would measure, with what, and how you would make the result reliable.

This section gives you a general toolkit for that: variables, instruments and their resolution, techniques that improve accuracy, recording, graphs and safety.

These skills are AS content, examined above all on Paper 3, where about half the marks are practical in kind, and also inside topic questions on Papers 1 and 2.

The full detail of uncertainties and the core practicals is in the practical skills unit and in each topic.

You need to be able to demonstrate your knowledge of practical skills and techniques for both familiar and unfamiliar experiments.

Familiar and unfamiliar experiments

A familiar experiment is one you have done, such as a core practical.

An unfamiliar one is a set-up you have not met: a pulse travelling along a stretched spring, a ball rolling off a bench, a ripple crossing a tray of water.

The physics inside it is always standard.

The trick is to break the experiment into single measurements, and for each one to ask three questions: what instrument, what range and resolution, and what technique makes the reading as accurate as possible.

Variables

The independent variable is the one you choose and change.

The dependent variable is the one you measure as a result.

Control variables are kept the same so they cannot affect the result. Say how each is kept constant ("use the same ball each time"), not just that it is.

Choose the independent variable's values to cover a wide range, with at least five or six values, so a graph can show the pattern.

Where a relationship is not linear (an inverse square, say), spread the values so the points are not crowded at one end.

Choosing instruments

Instrument

Typical resolution

Good for

metre rule

1 mm

lengths from about 10 cm to 1 m

vernier calipers

0.1 mm (digital: 0.01 mm)

lengths of a few mm to about 15 cm, internal and external diameters

micrometer screw gauge

0.01 mm

diameters of wires and small spheres, thicknesses under about 25 mm

stopwatch

0.01 s

times of several seconds or more; human reaction time (a few tenths of a second) dominates its uncertainty

light gate and data logger

0.001 s or better

short times and speeds, with no reaction time

top-pan balance

0.01 g or 0.1 g

masses

digital multimeter

the last digit shown

currents and potential differences

protractor

1°

angles

The photograph below shows four of these instruments side by side: a metre rule, vernier calipers, a micrometer screw gauge and a digital stopwatch.

Notice how much finer the scales of the calipers and the micrometer are than the millimetre marks on the rule.

Colour photograph of four measuring instruments side by side on a bench: part of a wooden metre rule, a pair of steel vernier calipers with outside and inside jaws and a sliding vernier scale, a micrometer screw gauge with its C-shaped frame, anvil, spindle, thimble and ratchet, and a digital stopwatch.

A good choice makes the percentage uncertainty small.

The same absolute uncertainty is a big fraction of a small reading and a small fraction of a large one.

That is why you measure a thin wire with a micrometer, not a ruler, and why you time twenty oscillations, not one.

Techniques that improve accuracy

Parallax: read a scale with your eye level with the mark, and use a set square to line a pointer up with a vertical rule.

Zero error: check that an instrument reads zero when it should (micrometer closed, balance empty, meter disconnected) and correct each reading if not.

Many cycles or many items: time 20 oscillations and divide by 20; measure the thickness of a stack of 100 sheets and divide by 100.

Fiducial marker: time oscillations as the object passes a fixed marker at the equilibrium position, where it moves fastest, so the instant is sharply defined.

Repeat and calculate a mean: this reduces the effect of random error. Write "and calculate a mean"; "repeat the readings" alone does not earn the mark.

Wait for steady conditions: let a swing settle, let a thermometer reach a steady reading, switch off a current between readings so a wire does not heat up.

The photograph below shows a pendulum set up with two of these ideas in practice.

The upright pin under the bob is a fiducial marker at the equilibrium position, so each oscillation can be counted as the bob passes it, and the base of the tall stand is fixed to the bench with a G-clamp so that it cannot topple.

Colour photograph of a pendulum set up for timing: a bob on a string hangs from a clamp near the top of a tall stand whose base is fixed to the bench with a G-clamp, and an upright pin in modelling clay on the bench directly below the bob marks its rest position.

Recording and processing

Record raw readings in a table with the quantity and its unit in each column heading, separated by a solidus: "time for 20 oscillations / s".

Every reading of one quantity is recorded to the same number of decimal places, set by the instrument's resolution.

Record every repeat, not just the mean.

Processed values (a mean, a square, a ratio) are given to the same number of significant figures as the raw data they come from, or one more; log values go to a consistent number of decimal places.

Graphs

Plot the independent variable along the bottom and the dependent variable up the side unless the question says otherwise, or unless rearranging the equation into $y = mx + c$ tells you to plot something else.

Label each axis with the quantity and unit, "distance / m".

Choose a scale that spreads the points over more than half of each axis, with easy steps (1, 2, 5 or 10 per large square, not 3 or 7).

Draw one thin best-fit line with the points balanced on either side; do not force it through the origin unless the data support it.

The gradient and intercept then give the quantities you want, with units.

The graph below shows a student's results for a pulse sent along a long stretched spring: the distance the pulse has travelled from one end, against time, measured with a video at 100 frames per second.

A graph of distance in metres against time in seconds for a pulse on a stretched spring. Five plotted points lie close to a straight best-fit line through the origin, rising to 2.50 m at about 1.0 s.

Safety

A safety answer names the hazard and the precaution, and says how the precaution is carried out.

"Be careful" or "wear goggles" with no hazard named scores nothing.

Examples: clamp a tall stand to the bench with a G-clamp so it cannot topple; wear eye protection when a stretched wire or spring might snap, because the broken end can whip back; keep a heated wire's current low so it cannot burn you; handle a radioactive source with tongs at arm's length.

The core practicals

Each of the sixteen core practicals is taught in full in the topic section that holds it.

Method questions on them are set on Paper 3.

Use this list to place each one.

Core practical

What it determines or investigates

Topic

acceleration of a freely-falling object

Mechanics

electrical resistivity of a material

Electric Circuits

e.m.f. and internal resistance of a cell

Electric Circuits

viscosity of a liquid by a falling-ball method

Materials

Young modulus of a material

Materials

speed of sound in air with an oscilloscope

Waves

effects of length, tension and mass per unit length on a vibrating string

Waves

wavelength of light with a diffraction grating

Waves

force and change of momentum

Further Mechanics

collisions between small spheres, analysed with ICT

Further Mechanics

p.d. across a charging and discharging capacitor

Electric and Magnetic Fields

calibrating a thermistor in a potential divider as a thermostat

Thermodynamics

specific latent heat of a phase change

Thermodynamics

pressure and volume of a gas at fixed temperature

Thermodynamics

absorption of gamma radiation by lead

Nuclear Radiation

an unknown mass from the resonant frequencies of known masses

Oscillations

Recall: not on the list

The resolution of each common instrument, and half the resolution as the uncertainty in one reading.

Speed = distance ÷ time, and the gradient of a distance-time graph as the speed.

How to head a table (quantity / unit) and label a graph's axes.

Key terms

Independent variable: the variable the experimenter chooses and changes.

Dependent variable: the variable measured as the independent variable changes.

Control variable: a variable kept constant so that it cannot affect the dependent variable.

Resolution: The smallest measuring interval and the source of uncertainty in a single reading.

Fiducial marker: a fixed mark used as the reference point for timing each oscillation.

Worked example 1: a gradient from a graph

Determine the speed of the pulse from the graph of the student's results above.

(3)

1. The speed is the gradient of the distance-time graph, because speed = distance ÷ time.

2. Large triangle on the best-fit line, using more than half of it: from (0.10 s, 0.25 m) to (1.00 s, 2.50 m) (1)

3. Gradient $= \dfrac{2.50 - 0.25}{1.00 - 0.10} = \dfrac{2.25 \text{ m}}{0.90 \text{ s}}$ (1)

4. Speed $= 2.5$ m s⁻¹ (1)

Answer: 2.5 m s⁻¹ (accept 2.4 to 2.6 m s⁻¹).

Read the triangle's corners from the line, not from two data points, which would carry their own scatter into the answer.

The line passing through the origin also tells you the timing has no systematic offset.

Worked example 2: choosing an instrument by its percentage uncertainty

A student needs the thickness of a plastic ruler, about 2.50 mm.

Calculate the percentage uncertainty in a single reading made with (a) a metre rule of resolution 1 mm and (b) a micrometer of resolution 0.01 mm, and state which instrument she should use.

1. Uncertainty in one reading is half the resolution: metre rule ±0.5 mm (1)

2. Percentage uncertainty $= \dfrac{0.5}{2.5} \times 100\% = 20\%$ (1)

3. Micrometer ±0.005 mm: $\dfrac{0.005}{2.50} \times 100\% = 0.2\%$ (1)

Answer: 20% with the rule, 0.2% with the micrometer, so use the micrometer.

A 20% uncertainty would swamp any result calculated from this thickness.

Worked example 3: stacking to measure something thin

A student measures the total thickness of a stack of 200 sheets of printer paper with vernier calipers of resolution 0.1 mm.

The stack is 18.6 mm thick.

Calculate the thickness of one sheet and the percentage uncertainty in it.

1. Thickness of one sheet $= \dfrac{18.6 \text{ mm}}{200} = 0.093$ mm (1)

2. Uncertainty in the reading = half the resolution = ±0.05 mm (1)

3. Percentage uncertainty $= \dfrac{0.05}{18.6} \times 100\% = 0.27\%$ , the same for one sheet (dividing by an exact 200 adds nothing) (1)

Answer: 0.093 mm, with a percentage uncertainty of about 0.3%.

Measuring one sheet directly with the same calipers would give ±0.05 mm on 0.1 mm, about 50%.

The stack assumes every sheet has the same thickness and the sheets are pressed flat; both are reasonable for printer paper.

Worked example 4: model answer, Describe a plan (4 marks)

A student wants to find out how the time a toy car takes to roll down a ramp depends on the height of the top of the ramp.

Describe how she should carry out the investigation.

Full-mark answer (any four):

Vary the height of the top of the ramp, measured with a metre rule from the bench to the top of the ramp, over at least five values (1)

Release the car from rest from the same marked point each time, without pushing (1)

Measure the time from release to the bottom with a stopwatch, or with light gates at the top and the bottom connected to a data logger (1)

Keep the ramp length, the car and its load, and the ramp surface the same (1)

Repeat each time measurement and calculate a mean (1)

Plot a graph of mean time against height (1)

Describe needs the steps; reasons are not required.

Weaker answer: "Change the height, time the car and draw a graph."

This scores 1 at most, because no instrument, no control and no repeat-and-mean is given.

Worked example 5: model answer, Criticise a results table (2 marks)

A student records the extension of a rubber cord for several loads.

The table shows her results.

load / N

mean extension / mm

1.0

2.0

25.5

3.0

4.0

50.0

Criticise the recording of these results.

Full-mark answer:

The extensions are not all recorded to the same number of decimal places (1)

Only the means are shown; the individual repeat readings used to calculate them are not recorded (1)

Comments about the number of loads or the range earn nothing here: the question is about how the data are recorded.

Worked example 6: model answer, a safety precaution (2 marks)

A student times a heavy pendulum bob hanging from a clamp on a tall stand.

State one safety precaution and how it should be carried out.

Full-mark answer:

The stand could topple when the heavy bob swings, so it must be made stable (1)

by clamping the base of the stand to the bench with a G-clamp (or putting a heavy weight on the base) (1)

Weaker answer: "Be careful the stand doesn't fall."

This scores 0: no precaution is described, and nothing says how.

Connect it

Used again in: every topic.

Each core practical section uses these skills with its own apparatus, the practical skills unit gives the full uncertainty routine and the Paper 3 question types, and the maths skills unit teaches plotting and the log graphs used to test a relationship.

Exam tip

When a question asks how to make a measurement accurate, name the instrument and the specific technique, and say why it helps in this experiment ("time 20 oscillations, so reaction time is a smaller fraction of the total").

Offering a different instrument instead of improving the technique asked about, or writing "repeat readings" with no mean, has scored nothing.

Common mistakes

Repeating with no mean. "Repeat the readings" earns nothing on its own; the mark is for repeating and calculating a mean, which reduces the effect of random error.

Blaming "human error". It is not a recognised source of error. Name it: reaction time in starting a stopwatch, parallax in reading a scale.

Saying a control is kept "the same" without saying how. "Use the same ball and release it from the same marked point" earns credit; "keep everything fair" does not.

Forcing the line through the origin. Draw the line the data support. A real intercept can reveal a zero error or a physical effect, and hiding it gives a wrong gradient.

Check your understanding

1. A student expects the time period $T$ of a swinging rope to be related to its length $L$ by $T^2 = kL$ , where $k$ is a constant. She plots $T$ against $L$ . The graphs below show four possible results.

Four sketch graphs of T against L, labelled A to D. A is a straight line through the origin. B is a curve through the origin that rises steeply at first and then less steeply. C is a curve through the origin that starts flat and gets steeper. D is a straight line with a positive intercept on the T axis.

Which graph shows her results if her expectation is correct?

(1) Total [1]

2. A student needs to measure the diameter of a wire, which is about 0.3 mm.

Which of the following is the most suitable instrument?

A metre rule

B vernier calipers with a resolution of 0.1 mm

C micrometer screw gauge

D protractor

(1) Total [1]

3. A student times a pendulum for different lengths and records her results in the table below.

length / cm

mean time for 10 swings / s

9.0

40.0

12.7

15.5

80.0

17.9

Criticise the recording of these results.

(2) Total [2]

4. A student sends ripples across a tray of water and records the distance travelled by a ripple at different times.

distance / cm

10.0

20.0

30.0

40.0

time / s

0.42

0.83

1.27

1.66

Determine the speed of the ripples in m s⁻¹.

(3)

Speed = ..............................

Total [3]

5. A balance has a resolution of 0.1 g. A student wants the mass of one paper clip, which is about 1 g.

Describe how he could find the mass of one paper clip as accurately as possible with this balance.

(4) Total [4]

6. A student plots a graph of the extension of a spring against the load on it. The points lie on a straight line, but the line crosses the extension axis at 4 mm, not at zero. She concludes that the extension is directly proportional to the load.

Comment on her conclusion.

(2) Total [2]

7. Explain why a student timing the swing of a pendulum should time 20 complete swings and not one. (2) Total [2]

Answers

1. The only correct answer is B. If $T^2 = kL$ , then $T = \sqrt{k}\sqrt{L}$ , so $T$ rises with the square root of $L$ : steep at first, then less steep, through the origin.

A is not correct because a straight line through the origin would mean $T$ is proportional to $L$ ; that is the graph of $T^2$ against $L$ .
C is not correct because it shows $T$ proportional to $L^2$ .
D is not correct because a straight line with an intercept shows a linear relationship that is not proportional, and $T$ must be zero when $L$ is zero.

2. The only correct answer is C. A micrometer reads to 0.01 mm, about 3% of 0.3 mm (half resolution ±0.005 mm is under 2%).

A is not correct because a 1 mm resolution is bigger than the wire itself.
B is not correct because ±0.05 mm on 0.3 mm is about 17%.
D is not correct because a protractor measures angles, not lengths.

3. The lengths are not all recorded to the same number of decimal places (1); the individual timings used to calculate the means are not recorded (1). Ignore comments on the number of readings.

4. Converts cm to m: 0.100 m to 0.400 m (1); use of gradient or speed = distance ÷ time over the largest interval, for example $\dfrac{0.400 - 0.100}{1.66 - 0.42}$ (1); speed = 0.24 m s⁻¹ (1). Accept 0.23 to 0.25 m s⁻¹. An answer of 24 with no conversion scores the second mark only.

5. Any four from: place a large number of paper clips, for example 50, on the balance together (1); check the balance reads zero with nothing on it, or tare it (1); record the total mass and divide by the number of clips (1); repeat with different sets of clips and calculate a mean (1); this makes the percentage uncertainty much smaller, since ±0.05 g is shared among all the clips (1). MAX 4.

6. A straight line shows a linear relationship, but the line does not pass through the origin, so the extension is not directly proportional to the load as plotted (1); the 4 mm intercept may be a systematic error (such as a zero error in measuring the original length), in which case the corrected extension could be proportional, so the conclusion is not justified by the graph as it stands (1).

7. The uncertainty in the timing (from reaction time) is about the same whether one swing or 20 are timed (1); so timing 20 swings makes the total time larger and the percentage uncertainty in the time for one swing smaller (1). MP2 dependent on MP1.

Practical skills for familiar and unfamiliar experiments

Your course level