This guide is for students preparing for the 2027 Singapore–Cambridge Secondary Education Certificate G3 Physics examination, syllabus K323, including the usual Secondary 4 cohort in 2027. It explains the published assessment structure and develops original examples of the reasoning needed in theory and practical work. Students taking a different examination year should check that year’s syllabus. Being in Secondary 4 during 2026 does not by itself make the 2027 SEC arrangements applicable to that student’s examination.
Confirm the registered subject before choosing papers. Standalone G3 Physics is different from the Physics component of combined Science. The G3 combined options include Science (Physics, Chemistry) K326 and Science (Physics, Biology) K327; G2 combined Science uses different codes and requirements. The paper structure below applies to K323. School preliminary examinations are school assessments and may have their own arrangements. Use the official syllabus and the school’s instructions for the actual national examination, and treat the worked examples here as teaching illustrations.
- Check the subject code, examination year and paper instructions before using a practice set.
- Prepare for multiple-choice, structured and practical work; recall alone is insufficient for unfamiliar applications.
- Define the system, identify relevant forces or transfers and state the assumptions behind a calculation.
- Keep quantities, units, graphs and diagrams connected to their physical meaning.
- Analyse experimental limitations specifically instead of writing general phrases such as human error.
- Use feedback to practise the underlying misconception with a fresh example, then explain the solution without looking at a model answer.
Read the official K323 assessment structure
The 2027 G3 Physics syllabus specifies three papers. Paper 2 has a compulsory 70-mark Section A and a 10-mark Section B in which candidates answer one of two questions. Paper 3 has two 55-minute sections, each worth 20 marks. These arrangements belong to standalone Physics, not every subject containing a Physics component. The SEAB G3 subject index identifies the separate registrations.
Use the weightings to plan balanced preparation, not to predict which chapters will appear. Raw marks and percentage contributions are different: the 40 marks in Paper 1 contribute 30 percent, while the 40 practical marks contribute 20 percent. A practice record should therefore distinguish paper performance from an overall weighted estimate. Follow the instructions printed on each actual paper, including question choice and apparatus access. A school exercise that uses a different length or mark total does not change the published national specification.
| K323 paper, 2027 | Duration | Raw marks | Overall weighting |
|---|---|---|---|
| Paper 1: multiple choice | 1 hour | 40 | 30% |
| Paper 2: structured and free response | 1 hour 45 minutes | 80 | 50% |
| Paper 3: practical | 1 hour 50 minutes | 40 | 20% |
Treat a physical quantity as a number with meaning
A measurement needs a magnitude and a unit. Writing that a length is 25 without specifying millimetres, centimetres or metres leaves the physical statement incomplete. Convert units before substituting into a formula that requires a consistent system. For example, a length of 25 centimetres is 0.25 metres. An area of 25 square centimetres is 0.0025 square metres, not 0.25 square metres. The area conversion squares the length factor. A unit check can reveal a major error even when the arithmetic operations were performed correctly.
Use scientific notation to keep scale visible. A current of 5 milliamperes is 0.005 amperes, while 5 kiloamperes is 5,000 amperes. Prefixes cannot be discarded without changing the value. For a calculated result, retain sufficient precision during intermediate steps and round according to the task’s requirements. Do not create extra precision by reporting many calculator digits after a coarse measurement. Ask whether the order of magnitude makes sense for the object or process. A plausible scale is an independent check, not a replacement for correct working.
Distinguish distance, displacement, speed and velocity
Distance measures the length of the path travelled. Displacement describes the change in position with direction. A person walking 30 metres east and then 10 metres west travels 40 metres but ends 20 metres east of the starting point. Average speed uses total distance divided by elapsed time; average velocity uses displacement divided by elapsed time. The two quantities can therefore differ for the same journey. Define the positive direction before assigning signs, and keep that choice consistent throughout the calculation.
Graphs encode these distinctions. The gradient of a displacement–time graph gives velocity, while the area under a velocity–time graph gives displacement with sign. A region below the time axis contributes negative displacement under the chosen convention. Distance travelled is obtained by considering the magnitudes of the relevant areas, not by allowing opposite directions to cancel. Before calculating a gradient or area, read the axes. Treating every sloping graph as an acceleration graph is a common error caused by recognising the shape while ignoring what the graph actually represents.
Use resultant force to explain acceleration
For a constant mass in the usual school model, resultant force equals mass multiplied by acceleration. A resultant force of six newtons on a two-kilogram object produces an acceleration of three metres per second squared. The force in this relationship is the resultant, not necessarily one named force in the diagram. If several forces act, combine them with their directions before calculating. A moving object does not require a nonzero resultant force simply to maintain constant velocity; a resultant force changes the velocity.
Draw a free-body diagram for the chosen object. Include forces acting on that object, not the reaction forces it exerts on other objects. Newton’s third-law partners act on different bodies, so they do not cancel on one body’s force diagram. For a book resting on a table, its weight and the table’s support can balance, but they are not a third-law pair. The partner to the support force is the force the book exerts on the table. This distinction prevents a verbal rule about equal forces from producing an incorrect explanation.
Explain free fall and terminal velocity with forces
In the ideal model of free fall near the Earth’s surface, gravity is the only force considered and the acceleration is approximately constant. The K323 syllabus uses an approximate value of 10 metres per second squared; follow any value supplied by the actual question. A heavier object does not automatically have a larger free-fall acceleration in this model. Its greater gravitational force is accompanied by greater mass. State that air resistance is neglected before applying the idealised reasoning to an object falling through air.
With air resistance included, the force balance can change as speed increases. For a falling object released from rest, the resistive force generally increases with speed until it can balance the weight under suitable conditions. The resultant force then becomes zero and the object continues at terminal velocity. Zero resultant force means zero acceleration here, not zero velocity. Ask the learner to draw force diagrams at different stages and connect them to a velocity–time graph. A horizontal section of that graph indicates constant velocity; it does not show that gravity has stopped acting on the object.
Calculate moments using perpendicular distance
The moment of a force about a pivot is the force multiplied by the perpendicular distance from the pivot to the force’s line of action. A 12-newton force acting with a perpendicular distance of 0.25 metres produces a moment of three newton metres. The length of a sloping handle is not automatically the required distance. Draw the line of action and identify the shortest perpendicular distance. This makes the geometry of the calculation visible and prevents a familiar formula from being applied to the wrong measurement.
For rotational equilibrium, total clockwise and anticlockwise moments balance about the chosen point. A six-newton force at a perpendicular distance of half a metre can balance a three-newton-metre moment in the opposite sense. Translational equilibrium also requires the resultant force to be zero. Do not assume that balanced moments alone establish every condition of equilibrium. In stability questions, examine the position of the centre of gravity and the base of support. Explain how a change affects the tendency to topple rather than simply state that lower objects are always more stable in every possible setup.
Connect pressure to force, area and depth
Pressure is normal force per unit area. If the same force acts over a smaller area, the pressure is greater. A force of 200 newtons distributed over 0.02 square metres produces 10,000 pascals. Convert the area carefully before substitution. A narrow contact can produce a large pressure without increasing the total force. The distinction helps explain why a sharp edge can cut more readily under otherwise comparable conditions. It does not mean that pressure and force are interchangeable quantities.
In a liquid of uniform density, the pressure difference due to a depth difference is described by ρgh in the relevant model. Specify whether the question asks for pressure due to the liquid alone or total pressure including an external contribution. The shape of a container is not the direct variable in this depth relationship. For a connected-fluid problem, identify the level and fluid involved before comparing pressures. An explanation should state the conditions supporting the comparison, such as a liquid at rest, rather than relying solely on how wide the vessels appear in the drawing.
Account for work, energy and power
Work done by a constant force along a displacement is force multiplied by displacement in that direction. Power is the rate of energy transfer or work done. Lifting a two-kilogram object through one metre increases its gravitational potential energy by 20 joules when the problem specifies g = 10 newtons per kilogram. Completing that transfer in two seconds gives an average useful power of ten watts. The force, distance, time and energy are connected, but each answers a different question about the process.
For kinetic energy, one half multiplied by mass and speed squared gives nine joules for a two-kilogram object moving at three metres per second. Doubling speed quadruples this energy when mass is unchanged. During braking, the kinetic energy can be transferred to thermal stores of the brakes, tyres and surroundings. Energy is conserved overall, but the energy of a chosen system can change through transfers across its boundary. A closed system can exchange energy; an isolated system exchanges neither energy nor matter. State the system clearly before claiming its energy remains constant.
Explain thermal behaviour through a particle model
Heating a substance can change the motion and arrangement of its particles. During ordinary thermal expansion, the particles are not described as swelling into larger particles. Their average separation changes within the relevant model. In a gas, pressure arises from particle collisions with the container walls. If a fixed amount of gas is heated at constant volume, its particles move more rapidly on average and the pressure rises. State what is held constant, because allowing the volume to change creates a different situation.
Conduction, convection and radiation describe different pathways of thermal transfer. Convection involves bulk movement of a fluid, whereas radiation can occur through a vacuum. A phrase such as heat rises is incomplete: in a common example, warmer, less dense air rises under gravity and carries energy. Choose the mechanism that matches the setup. A vacuum flask reduces several pathways in different ways, so one feature does not explain its whole performance. Ask which transfer a particular design feature reduces and how that feature acts on the pathway.
Use heating and phase-change equations in the right situations
For a temperature change without a change of state, the model Q = mcΔT relates transferred thermal energy to mass, specific heat capacity and temperature change. Heating 0.20 kilograms of water through 15 degrees Celsius requires 12,600 joules if c is given as 4,200 joules per kilogram per degree Celsius and losses are neglected. The temperature difference can be expressed in degrees Celsius or kelvins with the same numerical interval. The calculation is an idealised energy account; actual electrical input can be greater when energy also warms the container or surroundings.
During a phase change at constant temperature under the stated conditions, use the relevant latent heat relationship rather than assume that no energy is transferred. The energy changes the arrangement and interactions of particles instead of increasing temperature in that interval. Read a heating graph segment by segment. A flat region has a different interpretation from a rising region. Avoid applying Q = mcΔT across the entire process if it includes both temperature changes and a phase change. Account for each stage and then combine the required energies.
Connect wave speed, frequency and wavelength
For a periodic wave, speed equals frequency multiplied by wavelength. If a wave travels at 12 metres per second with a frequency of three hertz, its wavelength is four metres. Identify which quantity changes when the situation changes. A source sets the frequency of a wave crossing a stationary boundary in the usual model; a change in speed is then accompanied by a change in wavelength. Memorising an inverse relationship without stating the fixed quantity can lead to contradictory explanations.
Distinguish the transfer of energy from net transport of the medium. In a simple transverse wave on a rope, points on the rope oscillate while the disturbance travels along it. The rope is not carried along with each crest. Amplitude describes the maximum displacement from equilibrium, while wavelength measures a spatial repeat and period measures a temporal repeat. Label these on an appropriate diagram. A displacement–distance graph and a displacement–time graph may look similar, but their horizontal axes represent different quantities and support different direct measurements.
Read the electromagnetic spectrum through shared properties
Electromagnetic waves travel at the same speed in a vacuum, approximately 3.0 × 10⁸ metres per second. Their frequencies and wavelengths differ. At fixed wave speed, a higher frequency corresponds to a shorter wavelength. Do not say that one category necessarily travels faster in a vacuum because it has more energy per photon. The relationship between frequency and photon energy is distinct from the common vacuum speed. Use the level of detail required by the syllabus when explaining applications and hazards.
When comparing uses, connect a relevant property to the application. An answer should explain why the property is helpful in that context rather than simply match a name to a device. Also distinguish an application from a claim that the radiation is harmless under all conditions. Exposure and interactions matter. For examination practice, use the information supplied and the taught principles. For example, explain the role of ionisation in a stated application or hazard and relate penetration to the shielding material described in the question. A precise scientific explanation stays within the situation and evidence given.
Construct ray diagrams from the geometry
For reflection, measure angles from the normal to the surface. The angle of incidence equals the angle of reflection in the usual ray model. In refraction, the direction can change when light crosses a boundary between media, but normal incidence does not produce a change of direction in the standard situation. Draw the normal and label the rays before applying a relationship. A statement that light always bends on entering glass is therefore too broad. The geometry and the media must both be considered.
Total internal reflection requires light to travel towards a medium of lower refractive index and to meet the boundary at an incidence angle greater than the critical angle. Reversing the media or measuring the angle from the surface changes the analysis. For lenses, use the appropriate principal rays and extend rays only where needed to locate a virtual image. An image’s position should follow from the construction. Do not draw the desired image first and then force the rays to meet it. Check the resulting orientation, size and nature against the setup.
Distinguish charge, current and potential difference
Current is the rate of flow of charge. If 12 coulombs pass a point in four seconds, the current is three amperes. Potential difference is energy transferred per unit charge between two points. A potential difference of six volts means six joules transferred per coulomb in the relevant component. These definitions explain why current is measured through a path and potential difference across two points. A circuit’s current is not a quantity gradually consumed by each component while travelling around the loop.
In electrostatics, charging ordinary objects commonly involves transfer of electrons, not creation of charge from nothing. Opposite charges attract and like charges repel, but a charged object can also attract a neutral object through redistribution or polarisation. Therefore attraction alone does not establish that two objects carry opposite net charges. Use a diagram to show the relevant distribution and state the model’s limits. During circuit work, distinguish conventional current direction from electron motion in a metal. Mixing those conventions within one explanation can reverse an otherwise correct directional conclusion.
Analyse a circuit from its connections
For an ohmic resistor under constant physical conditions, potential difference is proportional to current, giving a constant resistance in that model. A 12-volt potential difference across a six-ohm resistor gives two amperes. Not every component maintains a constant resistance as its temperature or operating conditions change. A filament lamp’s current–voltage behaviour should not be explained by assuming constant resistance without checking the context. Read the axes of an electrical graph before deciding whether its gradient represents resistance or its reciprocal.
In a series circuit, components share the same current. In parallel, branches connected across the same two nodes share the same potential difference, while currents divide and recombine. Identify the nodes rather than judge the circuit from how neatly the symbols are drawn. Redraw a complicated-looking arrangement if needed without altering the connections. When a switch changes position, trace the new paths before recalculating. A remembered rule about adding a resistor is incomplete unless it specifies whether the resistor is added in series or as a parallel branch and what source model is assumed.
Calculate electrical power and energy with consistent units
Electrical power can be calculated as P = VI, with related resistor formulas used under their appropriate conditions. A component carrying two amperes with 12 volts across it transfers energy at 24 watts. Operating at that power for 50 seconds transfers 1,200 joules. Energy and power are different quantities; a watt is a joule per second. If a problem uses kilowatt-hours, keep the kilowatt and hour units together or convert systematically to joules. Do not treat a kilowatt-hour as a unit of power.
For practical electricity questions, connect protective features to their functions. A fuse is selected and connected to interrupt an excessive current under the relevant circuit conditions. Earthing provides a protective path in appropriate equipment designs; double insulation is a different protective approach. Follow the syllabus and school explanations for the specific device shown. Revision should use diagrams and approved laboratory equipment. Household mains systems are not suitable for improvised experiments, and understanding a protective feature in an examination question does not authorise practical electrical installation or repair.
Explain magnetic effects using fields and forces
A magnetic field describes the region in which magnetic effects can be detected. Field lines are a representation: their direction and spacing communicate features of the model, not physical threads in space. For a current-carrying wire or solenoid, connect the direction of current with the field pattern using the convention taught. Reversing the current reverses the associated field direction. A soft magnetic core can change a solenoid’s effect, but the explanation should identify the relevant material behaviour rather than assume every metal acts identically.
A current-carrying conductor in a magnetic field can experience a force when the geometry permits. The direction depends on both the current and the field. Reversing one reverses the force; reversing both can leave its direction unchanged. Draw the three directions before using a hand rule so that the rule supports the geometry. In a motor explanation, connect the forces on the coil to a turning effect and explain how the design sustains rotation. Simply naming the motor effect does not show how the arrangement produces the observed movement.
Describe induction through a changing magnetic relationship
Electromagnetic induction involves an induced electromotive force when the magnetic flux linkage changes. Movement of a magnet or coil can produce this change, but physical movement is not the only possibility: changing the current in a nearby circuit can change the magnetic field. If the induced circuit is closed, an induced current may flow. Keep the distinction between induced electromotive force and current visible. An open circuit can have an induced potential difference without a sustained current around a complete conducting path.
For a generator or transformer question, identify what changes and where energy is transferred. An ideal transformer model assumes no power loss; real devices have losses, so input and output power are not automatically equal in every stated situation. Use the supplied turns and voltage relationships with the correct primary and secondary labels. A numerical ratio can be inverted if the diagram is read carelessly. Check whether the answer matches the description of a step-up or step-down arrangement, then consider any current relationship only under the model and assumptions given.
Interpret radioactive decay statistically
Radioactive decay is random for an individual unstable nucleus, while large collections show a predictable statistical pattern. Half-life is the time for the number of undecayed nuclei, or the activity of a given sample, to fall to half its value. It does not mean every nucleus survives for exactly that time. After two half-lives, one quarter of the original undecayed population remains on average. The relationship is repeated halving, not subtraction of the same original amount during each interval.
For count-rate data, account for background when appropriate. Suppose an initial measured rate is 900 counts per minute and background is 20. The initial source contribution is 880. After two half-lives, the expected source contribution is 220, giving an expected measured rate of 240 when the same background is added. This is an invented arithmetic example, not laboratory measurement data. When discussing radiation, distinguish penetration from ionising ability and use the conditions supplied. Practical handling belongs within authorised school arrangements; a written revision exercise needs no radioactive material.
Plan practical work around a testable relationship
An investigation plan should identify the changed quantity, measured quantity and relevant controlled conditions. Explain how the measurements will test the proposed relationship. For a resistance investigation, the plan must specify the component, circuit, measurements and any relevant temperature control or monitoring. Naming a voltmeter and ammeter is not enough if their connections are unclear. Include a suitable range and repeats where helpful. A good method is sufficiently specific that another trained student could understand what data would be collected and how the comparison would be made.
Evaluate limitations by connecting them to a likely effect. A zero error, inconsistent alignment and a fluctuating reading are different problems and need different responses. Repetition can reduce the effect of random variation in an average, but it does not automatically correct a systematic offset. Use a large appropriate triangle for a graph gradient and retain units. Choose the line or curve justified by the data and instructions, rather than forcing every point onto an expected straight line. Keep observations honest even when the pattern differs from a prediction.
Approach unfamiliar questions through familiar principles
An unfamiliar device does not necessarily require knowledge outside the course. Read the supplied information and identify the relevant quantities, interactions and constraints. Draw a simplified diagram, mark what is known and state what is required. Then select a relationship whose conditions match the situation. Do not choose a formula only because it contains the same letters as the question. A temperature symbol, for example, does not make every problem a specific-heat calculation; the process might involve a phase change or an electrical measurement.
Write enough working to expose the reasoning. Substitute values with units, interpret the sign and compare the result with a sensible estimate. For a written explanation, connect cause, physical principle and consequence. If a question asks for a suggestion, use the provided evidence to develop a plausible explanation and acknowledge any necessary assumption. The objective is not to guess the examiner’s preferred phrase. It is to demonstrate that a known principle has been applied coherently to the new information, with the limits of that application kept visible.
Revise by testing the corrected idea in a new situation
Keep a record of the first incorrect step rather than only the final wrong answer. Sort errors into concepts, diagrams, data interpretation, units, algebra and incomplete reasoning. These categories lead to different actions. A learner who confuses force with energy needs conceptual comparison; one who converts square centimetres incorrectly needs dimensional practice. After correcting the example, change a condition and ask for a new prediction. This shows whether the improvement transfers beyond the exact question whose solution has just been seen.
Balance short targeted practice with complete tasks under the appropriate conditions. Review practical feedback alongside theory marks, and retain time for questions that require explanation rather than calculation. Before the examination, use the school’s instructions to confirm the registered paper, equipment and procedures. This guide does not guarantee particular questions or outcomes. Its purpose is to make preparation more precise: know the course, identify the physical relationship, work carefully with evidence and use each correction to improve the next independent attempt.
Related curriculum and examination guides
- Secondary 4 Science: SEC 2027 Routes and Exam Preparation
- Secondary 3 Science: Mechanisms, Models and Evidence
- Secondary 4 Additional Mathematics: SEC 2027 G2/G3 Guide
Official sources and review date
Fact-checked on 15 November 2026. This independent guide explains the published requirements and offers original teaching examples. The linked examination documents and school instructions govern the actual examination. Examples and suggested home routines below are educational illustrations, not official specimen questions or compulsory school schedules.