This practice outline analyses collisions using calibrated video or suitable motion-measurement ICT. It supports development of a locally approved method and risk assessment; it is not a complete validated procedure. Check current Pearson documents for course-specific requirements.
Aim and scientific principle
Measure both objects' masses and velocity components before and after a collision. Define the system and compare its total momentum over the same interval, assessing net external impulse. Video-only velocity analysis does not directly measure a force-time record. Kinetic-energy conservation is a separate test, not a consequence of momentum conservation alone.
Begin by translating the title into a testable question. Identify what is deliberately changed, what is measured directly, what is calculated and what must remain controlled. Predict a relationship only after naming the mechanism. A prediction without a reason is weak; a reason without measurable variables cannot guide a practical design.
Apparatus and setup
Use dynamics trolleys on a low-friction track, a securely mounted camera with a suitable recording timebase, a scale in the motion plane, tracking markers and a balance. Alternatively use calibrated motion sensors/data logging with common coordinates and a time reference. Record both objects' motion; a force sensor is additional evidence, not a substitute for these velocity data. Consider measuring range, resolution, uncertainty, calibration and suitability for the sample, not the instrument name alone. Draw a labelled scientific setup that shows the working geometry and exactly where readings are taken. Include a control or blank when it distinguishes the target effect from drift, contamination or background response.
Before collecting results, check zero readings, calibration, leaks, alignment or sterility as relevant. Record the instrument resolution beside the first raw-data table. If a digital device displays more digits than its underlying accuracy supports, do not treat every digit as equally certain.
Method
Define the two-object system, include attachments in moving masses and choose coordinate directions. Level and align the track. Mount the camera with its optical axis approximately normal to the motion plane and put the scale in that plane across the tracked region.
Verify the true recording frame interval, including high-speed/slow-motion conversion, dropped frames or non-uniform timing. Playback speed is not the recording timebase. Retain original frames and timing information; use actual elapsed time between analysed frames.
Track a consistent marker representing each object's translation. Fit suitable short position-time intervals separately before and after contact, excluding the interaction itself from constant-velocity fits. Use enough points to assess the fit without extending so far that friction changes the relevant endpoint velocities. For sensor ICT, establish equivalent spatial/time calibration and simultaneous records.
Check this topic from memory
Attempt the matching topic bank before reopening the notes. Use each missed idea to decide what to review next.
Repeat collision conditions and obtain both objects' signed velocities around the same event. Compare momentum components with uncertainty and assess external impulse. If investigating elasticity, separately compare total translational kinetic energy, stating any unmeasured rotation or other energy transfers.
Plan the analysis before starting. This reveals whether an additional measurement is needed for a derived quantity and prevents a completed experiment from lacking a mass, length, concentration, background or calibration value. A method should be detailed enough for another trained student to reproduce, while avoiding decorative steps that do not affect validity.
Variables and controls
Independent variable: initial speed, mass or interaction condition.
Dependent variable: both objects' velocity components and derived total momentum; kinetic-energy change where investigated.
Important controls: track slope, camera/scale geometry, motion plane, time calibration, markers and interaction attachments. If deliberately changing mass or attachments, record the new moving masses and system definition rather than claiming they stay fixed.
For each control, state how it is maintained or monitored. Simply writing "keep temperature constant" is incomplete if the method has no bath, probe, equilibration time or other means of doing so. Choose a range wide enough to reveal the relationship but within safe and scientifically valid limits. Choose the number, range and spacing of values for the expected response and analysis. Concentrating values near an uncertain threshold or optimum can be more useful than uniform spacing; no fixed count establishes an adequate design.
What is distinctive about this practical
Calibrated position and true elapsed time give velocity from a position-time gradient. One fit across the collision averages over the velocity change; it does not give separate before/after velocities. A rotating off-centre marker need not follow the centre of mass. Check perspective and lens distortion across the actual field of motion; one calibration length does not validate all geometry.
For one-dimensional motion, total momentum is p=m1v1+m2v2, using signed velocities along the declared axis. Translational kinetic energy is K=(m1v12+m2v22)/2. For planar motion compare momentum components separately and use full speed magnitudes for kinetic energy; a single projected component is not the full motion. Internal collision forces do not change combined momentum, but net external impulse can. Elasticity additionally requires kinetic-energy conservation within measurement/model limits; bouncing apart alone does not establish elasticity.
Recording and processing results
Retain marked positions, frame numbers/timestamps, scale and mass measurements, fit windows and residuals. Quote signed before/after momentum components and a defined residual with uncertainty. Define any percentage denominator and avoid a near-zero reference. Include common scale/timebase uncertainty shared by velocity estimates; averaging collisions does not remove it. State perspective, tracking, blur and unmeasured-motion limits as well as fit scatter. Put units in column headings, record raw readings consistently with instrument resolution and the method and distinguish repeats from calculated means. Do not delete an anomaly merely because it is inconvenient. Recheck the condition, look for a procedural explanation and state any exclusion rule.
Choose a fitted line, curve or ruled joins according to the data, scientific relationship and question; do not connect or smooth points mechanically. A gradient should use a large triangle and values read from the fitted line. When a relationship is linearised, explain the transformation and connect gradient or intercept to the requested physical quantity. When observations are categorical, define the category rule before seeing the result.
Safety, ethics and responsible practice
Use end stops, keep faces and fingers clear, and secure sensors and masses. A useful safety statement names the hazard, identifies who or what could be harmed and gives a control that reduces likelihood or severity. "Take care" does not specify an action that reduces the identified risk.
Worked application: from raw evidence to a defensible conclusion
This is a constructed timebase check, not measured collision data. If footage recorded at 120 frames per second is analysed as though recorded at 30 frames per second, elapsed time for a given frame count is four times too long. Inferred velocities and momenta are one quarter of correctly timed values; translational kinetic energies are one sixteenth, for unchanged positions and masses.
The same wrong factor in all before/after velocities can leave an apparent momentum balance looking consistent. Agreement alone does not validate the timebase. In real data, independently verify recording time, retain fit windows and uncertainty, and assess external impulse before interpreting momentum or kinetic-energy change.
This chain is stronger than a memorised recipe because every step answers the practical question. If the graph contradicts the prediction, report the evidence honestly and investigate method, range, controls and model assumptions before claiming a new scientific law.
Evaluation: limitation, effect and improvement
Limitation: friction can provide external impulse; off-axis motion, perspective, tracking and timing errors can make velocities incomplete or biased.
Likely effect: friction changes system momentum over the measurement interval; an omitted transverse component makes one-dimensional analysis incomplete. Common scale/timebase error may bias all momenta together and remain hidden by before/after agreement. Fit scatter alone does not reveal shared errors.
Targeted improvement: level and align the track, check the motion-plane spatial calibration and independently verify the recording timebase. Use separate pre/post fit intervals with enough valid frames and consistent markers. Shorter intervals reduce some frictional change but increase sensitivity to position/time uncertainty; longer intervals can average changing velocity. Inspect residuals and state the trade-off rather than choosing a fixed frame count mechanically.
Separate random variation, systematic bias and model limitation. Repeats help reveal random variation but do not remove a zero error. Calibration establishes the relation between reference values and indications; appropriate adjustment or correction can address an identified offset. It does not make biased sampling representative. Finer resolution and greater repeatability are distinct properties, and finer resolution may improve an estimate only when the reading is a material source of uncertainty.
Common misconceptions and corrections
Repeats automatically make a result accurate. Repeats estimate random variation; they do not correct a biased method or calibration error.
A control variable is just a list item. Explain the apparatus or procedure that keeps it stable.
The line must pass through every point. Use the overall pattern and a best-fit relationship, then investigate justified anomalies.
Percentage error and percentage uncertainty are interchangeable. Error compares a measured value with a suitable reference. Uncertainty characterises dispersion of values attributed to the measurand; it may be expressed as a standard uncertainty or an interval with stated coverage, and includes more than resolution alone. State the reference, denominator and uncertainty convention used.
A successful endpoint proves the theory. The observation supports a conclusion only within the method, range and assumptions tested.
More sophisticated apparatus is always better. The improvement must address the dominant limitation and be practical in the stated context.
Assessment guidance
Pearson assesses practical understanding throughout the science papers and concentrates it in written Units 3 and 6. Planning answers need an operational independent variable, a measured dependent variable, a justified range, controls, repeats, safety and a stated analysis. Data answers need units, consistent precision, sensible significant figures and transparent calculations. Graph answers need labelled axes, a scale using the grid well and line treatment appropriate to the data and question. Evaluation answers should use a limitation-effect-improvement chain specific to the apparatus. Do not claim that a memorised core-practical recipe is the only acceptable method when the question supplies a different valid context.
Retrieval practice
Extend or adapt this practical to answer a related question. Where the design uses a continuous variable, justify its planned values. List apparatus with range and resolution where relevant, control methods, a raw record layout, the expected analysis and one hazard-control pair. Then identify one random uncertainty, one systematic bias and one model limitation. Explain which can be reduced by repeats, which needs calibration followed by appropriate correction or adjustment, or a redesign, and how your conclusion would change if the anomaly were confirmed.
