Quick answer. For a negative temperature coefficient (NTC) thermistor, resistance decreases as temperature rises. In a series potential divider, the voltage across a component depends on its share of the total resistance. Therefore a temperature rise can make the output voltage fall or rise, depending on which component the output is measured across. The circuit must be identified before a student predicts the direction of the reading. This chapter teaches that diagnostic method through original low-voltage paper models, not unsupervised electrical tests.
Start with the previous EMS circuit-fault audit, Volume 0620: that chapter asked whether measurements distinguish rival circuit explanations. Here we make the problem harder. A reading may change even when nothing is broken, because the component itself responds to temperature. We learn to tell a normal environmental response from an incorrect meter position, a reversed sensor arrangement, a mistaken prediction or a genuinely inconsistent observation.
What this volume adds to the earlier circuit sequence
The D.C. circuits foundation in Volume 0613 established voltage and current locations. Volume 0617 compared distinct fault hypotheses. An earlier potential-divider foundation, Volume 0327 is the owner for general thermistor and LDR introduction. This volume instead owns the narrower skill of interpreting thermistor output during diagnostic calibration, with attention to measurement location and uncertainty.
Check the qualification before following a technique
The first Singapore-Cambridge Secondary Education Certificate (SEC) examination is in 2027. Candidates sit their subjects at G1, G2 or G3. SEAB’s G3 Physics syllabus is K323 in 2027, replacing the earlier reference number 6091; G3 Mathematics is K310 and G3 English Language is K300. These are not paper labels to memorise in place of content. The subject syllabus sets the assessment: Physics Paper 1 multiple choice (one hour, 40 marks, 30%), Paper 2 structured and free response (one hour 45 minutes, 80 marks, 50%), and Paper 3 practical (one hour 50 minutes, 40 marks, 20%). The Physics syllabus explicitly includes NTC thermistors and light-dependent resistors as input transducers in potential dividers. Combined Science students should check their own K326, K327 or K328 syllabus rather than assume the pure-Physics practical applies.
The official 2027 SEAB G3 syllabus directory, updated 26 October 2026, and the K323 Physics syllabus are the authorities for current rules. This learning exercise is not an examination paper, an electrical repair instruction or a replacement for trained supervision.
The sensor is a resistor whose resistance responds to its environment
An NTC thermistor has a resistance that normally decreases when its temperature rises. It is not a thermometer with an answer already displayed in degrees. A circuit can turn a changing resistance into a changing potential difference; a separate calibration or known characteristic is needed to interpret that output as an estimated temperature. Under a particular model a smaller resistance may mean a warmer sensor, but it does not prove that the air in every part of a room is warmer.
Begin with cause and effect: environmental temperature changes, the thermistor’s resistance changes, the circuit redistributes potential difference, and a voltmeter detects the output. A student who says simply that heat produces more voltage has skipped two critical relationships.
Draw two connection points before using an equation
Imagine a 6 V ideal source with a 2.0 kΩ fixed resistor above a thermistor, joined in series. The lower end of the thermistor is at the supply’s reference node. If output is measured between the junction and that reference node, it is the voltage across the thermistor. If the output is taken between the positive supply and the junction, it is the voltage across the fixed resistor. Those outputs are complements and move in opposite directions.
A voltmeter does not reveal which two points it measures merely by displaying a number. Mark the red and black connection points in a sketch, then say which component lies between them. On a timed paper, this often saves more marks than immediate algebra.
Find total resistance before distributing voltage
For two series resistors with values Rf and Rt, total resistance is Rf + Rt. With an ideal source Vs, the current is I = Vs/(Rf + Rt). The voltage across the thermistor is I Rt = Vs Rt/(Rf + Rt); the voltage across the fixed resistor is Vs Rf/(Rf + Rt). The fractions add to one because the potential differences across the two series components sum to the supply voltage.
The underlying relationship is series current, not a memorised arrow showing which voltage rises. Derive it once from V = IR and the series rules. The derivation remains valid when a new resistance is substituted.
Do not confuse input resistance with output voltage
When temperature rises, Rt falls for an NTC thermistor. With fixed Vs and Rf, I increases because total resistance falls. Yet voltage across the thermistor can decrease because its share of the series resistance becomes smaller. More current does not automatically mean more potential difference across each component. Both current and component resistance have changed.
This is a useful example of why single-variable verbal reasoning can be dangerous. A correct prediction accounts for both changes, or relies on the divider equation that includes them.
Reverse the output pair, reverse the trend
With the thermistor below the junction, the voltage across it falls as it warms. The simultaneous voltage across the upper fixed resistor rises. Neither prediction contradicts the other: they answer different questions. A student should name the output terminals in every explanation, especially when the same diagram is described as a temperature alarm.
The result is not merely a diagram convention. If a mark scheme asks for output across Rf, reporting the trend of output across Rt is a physics error even if the resistance trend was correctly recalled.
Build a numerical benchmark that can be checked
Let Vs = 6.0 V and Rf = 2.0 kΩ. At 20 °C suppose Rt = 4.0 kΩ. Total resistance is 6.0 kΩ, current is 1.0 mA, thermistor voltage is 4.0 V and fixed-resistor voltage is 2.0 V. At 40 °C suppose Rt = 1.0 kΩ. Current is now 2.0 mA, with 2.0 V across Rt and 4.0 V across Rf.
The fictional resistance values are an exercise calibration, not a manufacturer’s universal thermistor specification. Real devices differ, which is why the question must provide an appropriate graph, table or characteristic.
Use a calibration table rather than inventing temperatures
A fictional table might provide 20 °C: 4.0 kΩ; 25 °C: 3.0 kΩ; 30 °C: 2.0 kΩ; 35 °C: 1.5 kΩ; and 40 °C: 1.0 kΩ. In our 6 V circuit, thermistor-output voltage is then approximately 4.00, 3.60, 3.00, 2.57 and 2.00 V. Read the table with its assumptions. It describes a specific example, not an exact linear temperature-to-voltage law.
Do not draw a straight line and treat every intermediate point as measured simply because the values are tidy. The resistance–temperature response is generally nonlinear, and the divider makes voltage depend on a ratio.
Separate an observed trend from the reason for the trend
Suppose a student’s output falls from 3.60 V to 3.00 V during a supervised, controlled change in temperature. The observation is a 0.60 V decrease. The explanation is that the NTC thermistor resistance fell and its fraction of the fixed supply became smaller, under the stated circuit arrangement. The inference about temperature requires the calibration and the assumption that the device behaved normally.
A good Physics response does not turn a possible explanation into a certain cause when several variables might have changed. It states the relationship and then the warranted conclusion.
A light-dependent resistor is a useful comparison, not the same physical mechanism
A light-dependent resistor usually decreases in resistance when illumination increases. In a divider its output trend can look similar to an NTC thermistor’s output under warming, but light intensity and temperature are different inputs. A voltage trace alone cannot say which kind of transducer produced it if the circuit and environment are not identified.
Examiners can combine familiar rules in an unfamiliar context. Write the variable being changed above the diagram before making any comparison of output traces.
Plan a fair observation without changing several causes together
A controlled teaching investigation would hold the supply and fixed resistor constant, choose one environmental input, allow adequate time for the sensor to respond, use an appropriate instrument and record readings with units. The school controls the equipment and safety. If the supply varies while temperature rises, a change in output cannot be attributed to the thermistor alone.
Good experimental logic is about isolating explanations. A careful table contains the input, measured output and relevant controlled conditions; it does not treat the apparatus as a black box that always tells the truth.
Distinguish uncertainty, variation and a fault
No instrument reports infinitely precise values. A displayed 3.00 V may be rounded, the sensor may need time to settle, and temperature may not be uniform. If repeated measurements at the same stated condition are 2.98, 3.01 and 3.00 V, the small spread is not automatically a broken wire. If the whole trend reverses relative to the identified output, recheck connections, labeling, source and sensor assumptions before alleging component failure.
The working question is not ‘Is every reading perfect?’ but ‘Are the observations consistent with a stated model to the resolution and conditions supplied?’
Identify an open-path explanation from a change of current
In an ideal series circuit a break in a conducting path means no complete circuit and hence no steady current. That outcome differs from the normal NTC response, in which the resistance changes but the circuit stays connected. A supply current that shifts from 1.0 to 2.0 mA can be compatible with warming under the example; a reading of zero calls for a different hypothesis or a measurement issue.
Never encourage a learner to create a fault physically. The diagnostic task belongs in a printed diagram, teacher-approved model or supervised examination apparatus.
Check extremes before solving detailed numbers
If Rt becomes very large relative to Rf, most of the supply potential difference appears across Rt. If Rt becomes very small relative to Rf, its share approaches zero. The fixed resistor receives the complementary share. These limiting cases give a quick consistency check and can expose a swapped numerator before the calculator is used.
An extreme-case check is a powerful practical habit in Science and Mathematics. It explains why an answer is plausible without assuming an exact real-world zero or infinite resistance.
Use correct language for alarm thresholds
A threshold is a chosen rule applied to a measurement or estimated quantity. It is not an automatic declaration of physical danger. If a fictional chart states ‘below 3.0 V on the thermistor output,’ it describes a circuit output condition, not a definitive diagnosis of overheating everywhere. The meaning of the threshold depends on calibration, sensor position, and the conditions for which it was chosen.
Reports should describe ‘consistent with the example’s warmer calibrated region’ rather than ‘proves the room is unsafe’. A safety decision needs appropriate authorised procedures.
Explain the mathematics without allowing the calculation to replace the physics
At 6.0 V and 2.0 kΩ fixed resistance, an output of 3.0 V across Rt implies Rt = 2.0 kΩ in the ideal model. A learner might solve 3.0 = 6.0 Rt/(2.0 + Rt) using kΩ, or recognise equal voltage shares across equal series resistances. Both methods are legitimate. The short reason is that the same series current flows through each component.
The equation is useful only after the measured component is identified. A perfectly solved equation for the wrong output location is still the wrong model.
Practise under examination conditions while keeping an audit trail
Write ‘NTC: warmer → lower R’ at the top of a rough diagram; label the two output terminals; decide which voltage is requested; choose the divider ratio; substitute numerical values with consistent units; then test whether the direction matches the expected limit. This five-stage process is concise enough for an exam but complete enough for later checking.
If time is short, retain a short line of reasoning and the correct unit. A bare numerical answer hides whether you understood the effect of the environmental variable.
Worked mini-cases: examine the evidence, not only the result
Practice 1: 20 °C benchmark
Question. At 20 °C the sample thermistor is 4 kΩ in series with 2 kΩ. What is the output across the sensor?
Worked reasoning. The 6 V source is shared in the ratio 4:2; the sensor receives 4 V. Current is 1 mA. Common wrong turn. Writing 2 V means the fixed-resistor output was chosen instead of the thermistor output. Next check. Point to the two terminals before reading a voltage value.
A reliable answer records what quantity is changing, what is being kept constant and which equation is justified. Here that means refusing to replace the actual output terminals or calibration condition with a guess. In a structured Physics response, connect the condition to the mechanism and then to the specific observation. When the data permit more than one explanation, name the missing evidence rather than claiming certainty. In a multiple-choice question, reject options whose predicted trend contradicts the resistor share or the series-voltage check.
Practice 2: 40 °C benchmark
Question. At 40 °C its sample resistance is 1 kΩ. Predict current and sensor voltage.
Worked reasoning. Total 3 kΩ gives current 2 mA; voltage across sensor is 2 V. Common wrong turn. Assuming voltage rises because current doubled forgets that sensor resistance fell fourfold. Next check. Recalculate both current and voltage after each change.
A reliable answer records what quantity is changing, what is being kept constant and which equation is justified. Here that means refusing to replace the actual output terminals or calibration condition with a guess. In a structured Physics response, connect the condition to the mechanism and then to the specific observation. When the data permit more than one explanation, name the missing evidence rather than claiming certainty. In a multiple-choice question, reject options whose predicted trend contradicts the resistor share or the series-voltage check.
Practice 3: The other pair of terminals
Question. At 40 °C, what is the output across the fixed resistor?
Worked reasoning. The fixed component receives I × 2 kΩ = 4 V. The 2 V and 4 V readings sum to 6 V. Common wrong turn. Quoting the 2 V across the thermistor answers a different question. Next check. Write the component name beside the V symbol.
A reliable answer records what quantity is changing, what is being kept constant and which equation is justified. Here that means refusing to replace the actual output terminals or calibration condition with a guess. In a structured Physics response, connect the condition to the mechanism and then to the specific observation. When the data permit more than one explanation, name the missing evidence rather than claiming certainty. In a multiple-choice question, reject options whose predicted trend contradicts the resistor share or the series-voltage check.
Practice 4: Equal shares
Question. The thermistor and fixed resistor both have resistance 2 kΩ. Determine junction output across the thermistor.
Worked reasoning. Each takes half the 6 V source, so output is 3 V. This is independent of the fact that the current is 1.5 mA. Common wrong turn. Using resistance in ohms for one and kilohms for another ruins a comparison. Next check. Equal voltage shares in series are a useful fast check.
A reliable answer records what quantity is changing, what is being kept constant and which equation is justified. Here that means refusing to replace the actual output terminals or calibration condition with a guess. In a structured Physics response, connect the condition to the mechanism and then to the specific observation. When the data permit more than one explanation, name the missing evidence rather than claiming certainty. In a multiple-choice question, reject options whose predicted trend contradicts the resistor share or the series-voltage check.
Practice 5: Warming but unchanged output
Question. A student’s voltage remains 3.00 V while temperature appears to rise. Does that prove the thermistor has failed?
Worked reasoning. No. Recheck which temperature was measured, response delay, whether the change was sufficient, stable supply, and the specific measurement interval. It is a clue, not a full diagnosis. Common wrong turn. A single unchanging reading cannot distinguish a faulty sensor from an unchanging sensor temperature. Next check. List one further discriminating observation.
A reliable answer records what quantity is changing, what is being kept constant and which equation is justified. Here that means refusing to replace the actual output terminals or calibration condition with a guess. In a structured Physics response, connect the condition to the mechanism and then to the specific observation. When the data permit more than one explanation, name the missing evidence rather than claiming certainty. In a multiple-choice question, reject options whose predicted trend contradicts the resistor share or the series-voltage check.
Practice 6: A reversed trend
Question. A teacher expects thermistor output to fall with warming, but it rises. Give one plausible model-based explanation.
Worked reasoning. The voltmeter may be across the fixed resistor, where voltage rises as the NTC thermistor becomes less resistive. Confirm the terminal positions. Common wrong turn. Calling this automatically a defective device overlooks a perfectly consistent alternate measurement. Next check. A correct explanation changes only the stated assumption needed.
A reliable answer records what quantity is changing, what is being kept constant and which equation is justified. Here that means refusing to replace the actual output terminals or calibration condition with a guess. In a structured Physics response, connect the condition to the mechanism and then to the specific observation. When the data permit more than one explanation, name the missing evidence rather than claiming certainty. In a multiple-choice question, reject options whose predicted trend contradicts the resistor share or the series-voltage check.
Practice 7: Confusing current and voltage
Question. Current rises from 1 to 2 mA while sensor voltage falls from 4 to 2 V. Is that contradictory?
Worked reasoning. No. At the lower resistance, I Rt equals 2 mA × 1 kΩ = 2 V. Both observations obey Ohm’s law. Common wrong turn. Treating ‘increases’ as applying to all electrical quantities misses the multiplication by changing resistance. Next check. Identify exactly which variable is held fixed.
A reliable answer records what quantity is changing, what is being kept constant and which equation is justified. Here that means refusing to replace the actual output terminals or calibration condition with a guess. In a structured Physics response, connect the condition to the mechanism and then to the specific observation. When the data permit more than one explanation, name the missing evidence rather than claiming certainty. In a multiple-choice question, reject options whose predicted trend contradicts the resistor share or the series-voltage check.
Practice 8: Complementary voltages
Question. At one condition a reading across the sensor is 3.6 V. What should the other resistor read for an ideal 6 V supply?
Worked reasoning. The fixed resistor receives 6.0 − 3.6 = 2.4 V. Common wrong turn. A claim that both have 3.6 V would require 7.2 V across the pair, incompatible with the model. Next check. Conservation of potential difference provides an independent check.
A reliable answer records what quantity is changing, what is being kept constant and which equation is justified. Here that means refusing to replace the actual output terminals or calibration condition with a guess. In a structured Physics response, connect the condition to the mechanism and then to the specific observation. When the data permit more than one explanation, name the missing evidence rather than claiming certainty. In a multiple-choice question, reject options whose predicted trend contradicts the resistor share or the series-voltage check.
Practice 9: Lookup rather than guess
Question. In the exercise calibration the sensor output is 2.57 V. What table temperature does it approximately match?
Worked reasoning. About 35 °C, the row corresponding to 1.5 kΩ and approximately 2.57 V. State that this is an exercise estimate. Common wrong turn. Claiming 35.00 °C implies more precision than the table gives. Next check. State whether the value is measured, interpolated or merely selected from a row.
A reliable answer records what quantity is changing, what is being kept constant and which equation is justified. Here that means refusing to replace the actual output terminals or calibration condition with a guess. In a structured Physics response, connect the condition to the mechanism and then to the specific observation. When the data permit more than one explanation, name the missing evidence rather than claiming certainty. In a multiple-choice question, reject options whose predicted trend contradicts the resistor share or the series-voltage check.
Practice 10: A between-row reading
Question. What should you do with an output of 2.80 V, which lies between the 30 °C and 35 °C sample outputs?
Worked reasoning. It indicates an intermediate region under a monotonic calibration, but exact temperature is not determined by a five-point table. Explain any interpolation assumption if asked. Common wrong turn. Treating the response as strictly linear without justification assigns false certainty. Next check. A graph can estimate a value while still acknowledging its limits.
A reliable answer records what quantity is changing, what is being kept constant and which equation is justified. Here that means refusing to replace the actual output terminals or calibration condition with a guess. In a structured Physics response, connect the condition to the mechanism and then to the specific observation. When the data permit more than one explanation, name the missing evidence rather than claiming certainty. In a multiple-choice question, reject options whose predicted trend contradicts the resistor share or the series-voltage check.
Practice 11: Light instead of heat
Question. A sensor’s resistance falls as the light becomes brighter. Is it an NTC thermistor?
Worked reasoning. The description is consistent with a light-dependent resistor, but it does not establish identity by itself. The stimulus and circuit information are needed. Common wrong turn. Conflating a light change with a temperature change misidentifies the controlled variable. Next check. Begin from the evidence actually given.
A reliable answer records what quantity is changing, what is being kept constant and which equation is justified. Here that means refusing to replace the actual output terminals or calibration condition with a guess. In a structured Physics response, connect the condition to the mechanism and then to the specific observation. When the data permit more than one explanation, name the missing evidence rather than claiming certainty. In a multiple-choice question, reject options whose predicted trend contradicts the resistor share or the series-voltage check.
Practice 12: A missing unit
Question. A question says the resistances are 2 and 4 without units. Can you calculate a numerical current?
Worked reasoning. Not reliably. Their ratio may determine a voltage share, but absolute current requires the resistances’ units. Common wrong turn. A unit-free source current pretends that kΩ and Ω produce the same amperes. Next check. Use ratios when possible and state what is missing.
A reliable answer records what quantity is changing, what is being kept constant and which equation is justified. Here that means refusing to replace the actual output terminals or calibration condition with a guess. In a structured Physics response, connect the condition to the mechanism and then to the specific observation. When the data permit more than one explanation, name the missing evidence rather than claiming certainty. In a multiple-choice question, reject options whose predicted trend contradicts the resistor share or the series-voltage check.
Practice 13: Unstable source voltage
Question. The supply rises from 6.0 V to 6.3 V during the trial. Can the output shift be assigned solely to temperature?
Worked reasoning. No. The divider output is proportional to supply voltage as well as dependent on resistance. A fair interpretation needs the voltage source monitored or controlled. Common wrong turn. Ignoring a changed supply is an uncontrolled-variable error. Next check. Model the source explicitly and compare like conditions.
A reliable answer records what quantity is changing, what is being kept constant and which equation is justified. Here that means refusing to replace the actual output terminals or calibration condition with a guess. In a structured Physics response, connect the condition to the mechanism and then to the specific observation. When the data permit more than one explanation, name the missing evidence rather than claiming certainty. In a multiple-choice question, reject options whose predicted trend contradicts the resistor share or the series-voltage check.
Practice 14: A mistaken diagnosis
Question. A memo says, ‘Voltage went down so the sensor is definitely broken.’ Improve the inference.
Worked reasoning. Voltage can decline normally across an NTC thermistor as it warms in the specified divider. Determine terminals, temperature, supply and calibration before a fault is inferred. Common wrong turn. The word ‘definitely’ is unsupported by a single reading. Next check. Use ‘consistent with’ when alternatives remain.
A reliable answer records what quantity is changing, what is being kept constant and which equation is justified. Here that means refusing to replace the actual output terminals or calibration condition with a guess. In a structured Physics response, connect the condition to the mechanism and then to the specific observation. When the data permit more than one explanation, name the missing evidence rather than claiming certainty. In a multiple-choice question, reject options whose predicted trend contradicts the resistor share or the series-voltage check.
Practice 15: The open circuit
Question. A model shows zero series current, while the supply is available. Is normal thermistor warming a sufficient explanation?
Worked reasoning. No; even a relatively high but finite sensor resistance permits some current in a complete ideal circuit. Investigate an open path or the measurement model. Common wrong turn. Calling zero current a normal response without examining the circuit confuses finite with infinite resistance. Next check. Treat a new measurement as a test of competing hypotheses.
A reliable answer records what quantity is changing, what is being kept constant and which equation is justified. Here that means refusing to replace the actual output terminals or calibration condition with a guess. In a structured Physics response, connect the condition to the mechanism and then to the specific observation. When the data permit more than one explanation, name the missing evidence rather than claiming certainty. In a multiple-choice question, reject options whose predicted trend contradicts the resistor share or the series-voltage check.
Practice 16: Resistor order
Question. Move the sensor from below to above the junction and keep the output measured between junction and lower reference. Does the output trend stay the same?
Worked reasoning. No. The lower resistor is now fixed. Its voltage increases when the upper thermistor warms and resistance falls. Common wrong turn. A diagram rotation with unchanged node meaning can reverse the requested voltage trend. Next check. Do not use a memorised up-or-down arrow without the actual layout.
A reliable answer records what quantity is changing, what is being kept constant and which equation is justified. Here that means refusing to replace the actual output terminals or calibration condition with a guess. In a structured Physics response, connect the condition to the mechanism and then to the specific observation. When the data permit more than one explanation, name the missing evidence rather than claiming certainty. In a multiple-choice question, reject options whose predicted trend contradicts the resistor share or the series-voltage check.
Practice 17: Uncertain readings
Question. Three outputs at a nominally steady condition are 2.98, 3.01 and 3.00 V. What should the student report?
Worked reasoning. Report that the values cluster around 3.00 V and identify the small observed spread. Do not invent a cause or overstate certainty. Common wrong turn. Averaging without acknowledging the measurements may hide useful evidence about variation. Next check. Describe both centre and spread when evaluating reliability.
A reliable answer records what quantity is changing, what is being kept constant and which equation is justified. Here that means refusing to replace the actual output terminals or calibration condition with a guess. In a structured Physics response, connect the condition to the mechanism and then to the specific observation. When the data permit more than one explanation, name the missing evidence rather than claiming certainty. In a multiple-choice question, reject options whose predicted trend contradicts the resistor share or the series-voltage check.
Practice 18: An alarm statement
Question. A notice says ‘below 3 V proves dangerous heat.’ What is missing?
Worked reasoning. It lacks a specified calibrated device, output terminals, applicability conditions and an approved safety criterion. Replace it with a statement about the example voltage threshold. Common wrong turn. Threshold readings do not by themselves certify a physical space as dangerous or safe. Next check. Keep school safety decisions with authorised adults.
A reliable answer records what quantity is changing, what is being kept constant and which equation is justified. Here that means refusing to replace the actual output terminals or calibration condition with a guess. In a structured Physics response, connect the condition to the mechanism and then to the specific observation. When the data permit more than one explanation, name the missing evidence rather than claiming certainty. In a multiple-choice question, reject options whose predicted trend contradicts the resistor share or the series-voltage check.
Practice 19: Working marks
Question. A student writes only ‘2 V’ for a five-mark request to explain a sensor response. What should the student add?
Worked reasoning. Include the NTC resistance trend, the series-divider relationship, a substituted calculation, units and an interpretation for the specified pair of terminals. Common wrong turn. A correct final number cannot show all the causal reasoning asked for. Next check. Use short, precise lines to make every relevant inference inspectable.
A reliable answer records what quantity is changing, what is being kept constant and which equation is justified. Here that means refusing to replace the actual output terminals or calibration condition with a guess. In a structured Physics response, connect the condition to the mechanism and then to the specific observation. When the data permit more than one explanation, name the missing evidence rather than claiming certainty. In a multiple-choice question, reject options whose predicted trend contradicts the resistor share or the series-voltage check.
Practice 20: Transfer to a new set-up
Question. The source becomes 9 V, with fixed 3 kΩ and sensor 6 kΩ. Find output across the sensor.
Worked reasoning. The sensor owns two-thirds of the series resistance; it receives 6 V. Total current is 1 mA. It is the same ratio idea as the benchmark, not the same absolute input. Common wrong turn. Reusing the old 6 V answer without recalculation ignores the new supply. Next check. Start by rewriting the new givens and output location.
A reliable answer records what quantity is changing, what is being kept constant and which equation is justified. Here that means refusing to replace the actual output terminals or calibration condition with a guess. In a structured Physics response, connect the condition to the mechanism and then to the specific observation. When the data permit more than one explanation, name the missing evidence rather than claiming certainty. In a multiple-choice question, reject options whose predicted trend contradicts the resistor share or the series-voltage check.
How to turn twenty cases into independent examination skill
Work through the cases once without looking at the explanations. On the second attempt, redraw the output terminals for each case and predict the sign of the voltage change before calculating anything. On the third attempt, give yourself a different fixed resistance or supply voltage. The aim is to transfer a principle, not memorise that one example gives 4 V and another gives 2 V. If a wrong answer occurs, trace the error to its first incorrect assumption.
Use a simple notebook divided into four columns: situation, selected model, line of working and independent check. For a model problem, write the circuit assumptions. For a data problem, identify the table and units. For an inference, state why alternatives remain. This is especially important in the K323 Paper 2 data-based question and in practical interpretation, where students must use evidence rather than recite a definition. The method also prepares them to explain a Science answer accurately in English.
A three-session learning plan
Session one: practise only the electrical foundations. Label a simple series circuit, show equal current, find total resistance and verify that component voltages add to the source. Work with three new pairs of positive resistances. The learner should not proceed until they can explain which terminals the voltmeter touches and why the chosen numerator belongs in the voltage-divider fraction.
Session two: introduce the environmental input. Read NTC and LDR descriptions separately, calculate predictions from a small table and compare the opposing trends of fixed-resistor and sensor outputs. Pause when a phrase like ‘voltage rises when warm’ appears without a named component. That shortcut needs repair, not another calculator example.
Session three: practise evaluation. Given two different interpretations of a measurement, decide what observation would distinguish them. Explain the difference between an actual recorded voltage, a modelled resistance, an inferred temperature and a safety decision. Finish with a timed written answer in complete sentences, then compare it to the evidence and check which claims remain unsupported.
Parents and teachers: diagnose the first weak link
Ask a learner to explain one incorrect answer aloud. If they know an NTC thermistor decreases in resistance but choose the wrong voltage trend, revisit divider position, not the thermistor definition. If the diagram and equation are right but the current is a thousand times too large, revisit kΩ-to-Ω conversion. If the arithmetic is correct but the conclusion declares a hazard, revisit the boundary between classroom calibration and authorised safety assessment.
For the earlier PSLE bridge, the Science label-to-mechanism guide teaches the same intellectual movement in simpler terms. For further routes use the Science Hub and Mathematics Hub. The next chapter, Volume 0622, develops the numerical calibration and threshold questions in a dedicated Mathematics lesson.
The closing rule
A thermistor does not tell a story on its own. The circuit, the voltmeter terminals, the calibration and the controlled conditions give a reading its meaning. A learner who can say what changed, derive the expected response, calculate a check and state what remains uncertain is practising the sort of reasoning that transfers beyond a familiar school diagram.
Official references: SEAB SEC overview; 2027 G3 subject-code index; K323 2027 Physics syllabus. The original classroom numbers in this guide are illustrative, not official examination questions or safety thresholds.
