Real power, reactive power and apparent power describe different parts of an AC load account. A useful explanation preserves the reference point, phase basis, waveform assumption, averaging window and sign convention instead of treating every number with an electrical unit as energy consumption.
The central fictional load has P=60 kW and Q=+80 kvar, where positive Q is declared inductive. Under the simplified sinusoidal model, |S|=100 kVA and the power-factor magnitude is0.60. Adding a stipulated -50 kvar contribution makes net Q=30 kvar and |S| about67.08 kVA while P remains60 kW.
The values are invented for learning. They do not size conductors, specify a capacitor bank, interpret a live tariff, guarantee savings or replace competent electrical engineering and current safety rules.
This specialist guide belongs to the Utilities root. It also connects to The Gold Standard of Education | Learning That Transfers into Life and the How X Works hub. The worked cases are original teaching examples; their numbers are stipulated so that the reasoning can be checked.
eduKateSG · Utilities
Find your next learning step
Write P, Q and |S| in separate columns, declare the reactive sign convention and only then calculate the power triangle.
ROUTE 5 · CHAPTERS 21–24
Report and continue
Repair shortcuts, state limits and follow verified routes.
Chapter contents
Define the quantities · Chapters 1–5
1. Name real, reactive and apparent power
2. Respect the quantity-specific units
Work the mechanism · Chapters 6–10
6. Complete the central calculation
7. Add a compensating reactive contribution
8. Separate instantaneous power from interval energy
Audit the evidence · Chapters 11–15
11. Preserve the reference point
12. Preserve phase aggregation
13. Check meter channels and scaling
Transfer the reasoning · Chapters 16–20
17. Transfer to inverter-based equipment
18. Transfer to utility billing
Report and continue · Chapters 21–24
21. Repair the reactive power is wasted energy shortcut
22. Repair the better power factor saves the same energy shortcut
Start by naming the physical or logical quantity rather than the displayed number. Real power P describes net energy transfer per unit time; reactive power Q describes alternating energy exchange; apparent power magnitude |S| combines them under the stated AC model.
Work the original teaching case in a visible order. A balanced teaching load is assigned P=60 kW and Q=+80 kvar, with the positive sign defined as inductive.
Make the reasoning independently checkable. Write P+jQ before taking a magnitude and preserve the sign convention.
Test transfer by changing one condition while holding the others still. Reverse Q to -80 kvar while keeping P fixed.
Keep the claim boundary visible. Equal magnitudes of Q with opposite signs describe different exchange directions even though |S| is unchanged.
Before accepting the result, predict what should change and what should remain fixed. Apply this check: Write P+jQ before taking a magnitude and preserve the sign convention. Then run the changed case: Reverse Q to -80 kvar while keeping P fixed. Record any mismatch without rewriting the original assumptions.
The evidence receipt for “Name real, reactive and apparent power” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
Keep observation, transformation and conclusion in separate evidence columns. Kilowatt, kilovar and kilovolt-ampere distinguish quantities that share related dimensions but answer different questions.
Work the original teaching case in a visible order. The ledger labels 60 kW, 80 kvar and 100 kVA instead of calling every entry kilowatts.
Make the reasoning independently checkable. Check each value against the meter channel or calculation that owns it.
Test transfer by changing one condition while holding the others still. Relabel one channel incorrectly and identify the false inference it invites.
Keep the claim boundary visible. A unit label cannot rescue an undefined measurement boundary.
Build a compact audit receipt with input, unit, model, calculation, observation and limit. Use this independent check: Check each value against the meter channel or calculation that owns it. For transfer, Relabel one channel incorrectly and identify the false inference it invites. Preserve the first result beside the changed one.
The evidence receipt for “Respect the quantity-specific units” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
Use the short answer only after the boundary and unit have been fixed. For sinusoidal steady state in the simplified model, |S|=sqrt(P²+Q²).
Work the original teaching case in a visible order. sqrt(60²+80²)=100, so the stipulated apparent power is 100 kVA.
Make the reasoning independently checkable. Square, sum and take the root before rounding.
Test transfer by changing one condition while holding the others still. Use P=60 and Q=30 to obtain about67.08 kVA.
Keep the claim boundary visible. The triangle is a model relationship, not evidence that a real waveform is sinusoidal.
Ask a second reader to reproduce the decision from the stated evidence. They should be able to apply: Square, sum and take the root before rounding. Next, Use P=60 and Q=30 to obtain about67.08 kVA. If the conclusion changes, name the changed condition explicitly.
The evidence receipt for “Build the power triangle” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
Trace the mechanism in the order another reader would need to reproduce it. Under the stated single-frequency model, the magnitude of power factor is P/|S|.
Work the original teaching case in a visible order. The central case gives 60/100=0.60.
Make the reasoning independently checkable. Multiply |S| by0.60 and recover60 kW.
Test transfer by changing one condition while holding the others still. Use the compensated case 60/67.08≈0.894.
Keep the claim boundary visible. A magnitude alone does not state leading or lagging direction.
Turn the claim into a falsifiable test. First, Multiply |S| by0.60 and recover60 kW. Then change one consequential condition: Use the compensated case 60/67.08≈0.894. State which observation would count against the explanation.
The evidence receipt for “Calculate displacement power factor” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
Treat a plausible result as a prompt for a check, not as proof by itself. An inductive or capacitive convention assigns the sign of Q; software and industries may present signs differently.
Work the original teaching case in a visible order. This article declares +Q inductive before any calculation.
Make the reasoning independently checkable. Read the instrument manual and phasor convention.
Test transfer by changing one condition while holding the others still. Apply a convention in which inductive Q is negative and translate every sign consistently.
Keep the claim boundary visible. Calling one sign good or bad without the declared convention is meaningless.
Before accepting the result, predict what should change and what should remain fixed. Apply this check: Read the instrument manual and phasor convention. Then run the changed case: Apply a convention in which inductive Q is negative and translate every sign consistently. Record any mismatch without rewriting the original assumptions.
The evidence receipt for “Keep the reactive sign visible” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
Start by naming the physical or logical quantity rather than the displayed number. The three quantities should reconcile before any operational conclusion is drawn.
Work the original teaching case in a visible order. P=60 kW and Q=80 kvar produce |S|=100 kVA and power-factor magnitude0.60.
Make the reasoning independently checkable. Verify with both the triangle and P/|S|.
Test transfer by changing one condition while holding the others still. Reduce Q to30 kvar without changing P.
Keep the claim boundary visible. The example is stipulated and does not size equipment or prescribe correction.
Build a compact audit receipt with input, unit, model, calculation, observation and limit. Use this independent check: Verify with both the triangle and P/|S|. For transfer, Reduce Q to30 kvar without changing P. Preserve the first result beside the changed one.
The evidence receipt for “Complete the central calculation” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
Keep observation, transformation and conclusion in separate evidence columns. Reactive sources can change net Q while the real-power demand in the simplified snapshot remains fixed.
Work the original teaching case in a visible order. A -50 kvar contribution changes net Q from80 to30 kvar; apparent power falls from100 to about67.08 kVA.
Make the reasoning independently checkable. Add signed Q values before taking the magnitude.
Test transfer by changing one condition while holding the others still. Over-correct with -100 kvar and obtain net Q=-20 kvar.
Keep the claim boundary visible. Real installations require voltage, harmonics, switching and resonance checks.
Ask a second reader to reproduce the decision from the stated evidence. They should be able to apply: Add signed Q values before taking the magnitude. Next, Over-correct with -100 kvar and obtain net Q=-20 kvar. If the conclusion changes, name the changed condition explicitly.
The evidence receipt for “Add a compensating reactive contribution” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
CHAPTER 8 OF 24 · Work the mechanism
8. Separate instantaneous power from interval energy
Use the short answer only after the boundary and unit have been fixed. Power is a rate; energy accumulates over a stated time window.
Work the original teaching case in a visible order. At a steady60 kW for0.5 h, the real-energy register advances30 kWh.
Make the reasoning independently checkable. Multiply only after confirming the load is steady or integrating a time series.
Test transfer by changing one condition while holding the others still. Let P vary between40 and80 kW during the interval.
Keep the claim boundary visible. A snapshot cannot establish an interval total.
Turn the claim into a falsifiable test. First, Multiply only after confirming the load is steady or integrating a time series. Then change one consequential condition: Let P vary between40 and80 kW during the interval. State which observation would count against the explanation.
The evidence receipt for “Separate instantaneous power from interval energy” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
Trace the mechanism in the order another reader would need to reproduce it. A meter may integrate reactive power over time, producing a reactive-energy quantity that remains distinct from billed real energy.
Work the original teaching case in a visible order. Steady80 kvar for0.5 h corresponds to40 kvarh under the teaching assumptions.
Make the reasoning independently checkable. Keep kWh and kvarh on separate register columns.
Test transfer by changing one condition while holding the others still. Compensate halfway through the interval and integrate each segment.
Keep the claim boundary visible. Tariffs and meter definitions are jurisdiction-specific.
Before accepting the result, predict what should change and what should remain fixed. Apply this check: Keep kWh and kvarh on separate register columns. Then run the changed case: Compensate halfway through the interval and integrate each segment. Record any mismatch without rewriting the original assumptions.
The evidence receipt for “Handle reactive energy separately” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
CHAPTER 10 OF 24 · Work the mechanism
10. Distinguish displacement and true power factor
Treat a plausible result as a prompt for a check, not as proof by itself. Harmonic currents can make total apparent power and true power factor differ from the fundamental displacement factor.
Work the original teaching case in a visible order. The central triangle assumes sinusoidal voltage and current at one frequency.
Make the reasoning independently checkable. Inspect waveform distortion or the meter definition before importing the triangle.
Test transfer by changing one condition while holding the others still. Add a third-harmonic current while keeping the fundamental phase angle fixed.
Keep the claim boundary visible. One cosine value need not describe a distorted load.
Build a compact audit receipt with input, unit, model, calculation, observation and limit. Use this independent check: Inspect waveform distortion or the meter definition before importing the triangle. For transfer, Add a third-harmonic current while keeping the fundamental phase angle fixed. Preserve the first result beside the changed one.
The evidence receipt for “Distinguish displacement and true power factor” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
Start by naming the physical or logical quantity rather than the displayed number. P and Q measured at a machine terminal, panel, transformer or utility meter include different losses and parallel loads.
Work the original teaching case in a visible order. The example is declared at one service panel.
Make the reasoning independently checkable. Draw the electrical boundary beside the numbers.
Test transfer by changing one condition while holding the others still. Move the measurement upstream of a transformer.
Keep the claim boundary visible. Values from different boundaries cannot be reconciled by arithmetic alone.
Ask a second reader to reproduce the decision from the stated evidence. They should be able to apply: Draw the electrical boundary beside the numbers. Next, Move the measurement upstream of a transformer. If the conclusion changes, name the changed condition explicitly.
The evidence receipt for “Preserve the reference point” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
Keep observation, transformation and conclusion in separate evidence columns. Three-phase totals depend on whether channels are summed from phase values or inferred under balance assumptions.
Work the original teaching case in a visible order. The fictional load is explicitly balanced.
Make the reasoning independently checkable. Compare per-phase readings with the reported total.
Test transfer by changing one condition while holding the others still. Unbalance one phase while preserving total P.
Keep the claim boundary visible. A balanced formula can conceal neutral current and unequal phase loading.
Turn the claim into a falsifiable test. First, Compare per-phase readings with the reported total. Then change one consequential condition: Unbalance one phase while preserving total P. State which observation would count against the explanation.
The evidence receipt for “Preserve phase aggregation” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
Use the short answer only after the boundary and unit have been fixed. Current transformers, voltage transformers, polarity and channel mapping affect recorded P and Q.
Work the original teaching case in a visible order. A reversed current-transformer polarity flips one channel’s signs.
Make the reasoning independently checkable. Reconcile scaling ratios and phase association.
Test transfer by changing one condition while holding the others still. Swap two voltage references.
Keep the claim boundary visible. A neat power triangle can be built from miswired inputs.
Before accepting the result, predict what should change and what should remain fixed. Apply this check: Reconcile scaling ratios and phase association. Then run the changed case: Swap two voltage references. Record any mismatch without rewriting the original assumptions.
The evidence receipt for “Check meter channels and scaling” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
Trace the mechanism in the order another reader would need to reproduce it. Rounding and instrument accuracy propagate through |S| and power factor.
Work the original teaching case in a visible order. Treat60 and80 as rounded inputs and report sensible precision.
Make the reasoning independently checkable. Recompute upper and lower bounds before printing decimals.
Test transfer by changing one condition while holding the others still. Allow each input a one-unit uncertainty.
Keep the claim boundary visible. Extra decimal places do not create metrological certainty.
Build a compact audit receipt with input, unit, model, calculation, observation and limit. Use this independent check: Recompute upper and lower bounds before printing decimals. For transfer, Allow each input a one-unit uncertainty. Preserve the first result beside the changed one.
The evidence receipt for “Carry measurement uncertainty” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
Treat a plausible result as a prompt for a check, not as proof by itself. Load composition and Q can change across operating states.
Work the original teaching case in a visible order. The facility has start-up, production and idle intervals.
Make the reasoning independently checkable. Plot P, Q, |S| and power factor on the same timestamps.
Test transfer by changing one condition while holding the others still. Shift a motor start into the demand interval.
Keep the claim boundary visible. One favourable snapshot does not establish all-day behaviour.
Ask a second reader to reproduce the decision from the stated evidence. They should be able to apply: Plot P, Q, |S| and power factor on the same timestamps. Next, Shift a motor start into the demand interval. If the conclusion changes, name the changed condition explicitly.
The evidence receipt for “Use time series instead of one peak” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
Start by naming the physical or logical quantity rather than the displayed number. An induction motor can draw magnetising reactive power while delivering real mechanical output through losses and conversion.
Work the original teaching case in a visible order. Compare an unloaded and loaded motor at the same supply voltage.
Make the reasoning independently checkable. Keep shaft work, electrical P and Q in separate columns.
Test transfer by changing one condition while holding the others still. Add mechanical load while observing both channels.
Keep the claim boundary visible. The example does not diagnose a particular motor.
Turn the claim into a falsifiable test. First, Keep shaft work, electrical P and Q in separate columns. Then change one consequential condition: Add mechanical load while observing both channels. State which observation would count against the explanation.
The evidence receipt for “Transfer to motor loads” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
Keep observation, transformation and conclusion in separate evidence columns. Converters may control reactive exchange independently within current and control limits.
Work the original teaching case in a visible order. A stipulated inverter holds P=40 kW while Q changes from0 to20 kvar.
Make the reasoning independently checkable. Check the apparent-power and current ceiling.
Test transfer by changing one condition while holding the others still. Request more Q until the ceiling binds.
Keep the claim boundary visible. A control setpoint is not proof of delivered grid support.
Before accepting the result, predict what should change and what should remain fixed. Apply this check: Check the apparent-power and current ceiling. Then run the changed case: Request more Q until the ceiling binds. Record any mismatch without rewriting the original assumptions.
The evidence receipt for “Transfer to inverter-based equipment” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
Use the short answer only after the boundary and unit have been fixed. Bills can include energy, demand and power-factor provisions with different windows and definitions.
Work the original teaching case in a visible order. The teaching bill keeps kWh, peak kW and kvar-related terms separate.
Make the reasoning independently checkable. Read the current tariff and meter-data interval.
Test transfer by changing one condition while holding the others still. Change the billing window without changing physical load.
Keep the claim boundary visible. This article does not interpret a live tariff or guarantee savings.
Build a compact audit receipt with input, unit, model, calculation, observation and limit. Use this independent check: Read the current tariff and meter-data interval. For transfer, Change the billing window without changing physical load. Preserve the first result beside the changed one.
The evidence receipt for “Transfer to utility billing” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
Trace the mechanism in the order another reader would need to reproduce it. Conductors and transformers respond to current and apparent loading as well as real-energy delivery.
Work the original teaching case in a visible order. The compensated case carries the same60 kW with lower stipulated |S|.
Make the reasoning independently checkable. Check nameplate, voltage and current instead of using kW alone.
Test transfer by changing one condition while holding the others still. Lower voltage while preserving real power.
Keep the claim boundary visible. Thermal and protection studies need more than the triangle.
Ask a second reader to reproduce the decision from the stated evidence. They should be able to apply: Check nameplate, voltage and current instead of using kW alone. Next, Lower voltage while preserving real power. If the conclusion changes, name the changed condition explicitly.
The evidence receipt for “Transfer to equipment loading” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
Treat a plausible result as a prompt for a check, not as proof by itself. A complete record names waveform assumption, phase basis, reference point, sign convention, averaging window and meter channels.
Work the original teaching case in a visible order. Another reader reconstructs100 kVA and0.60 from the saved row.
Make the reasoning independently checkable. Remove one field and identify the ambiguity.
Test transfer by changing one condition while holding the others still. Version the receipt after compensation is switched.
Keep the claim boundary visible. Reproducibility checks the calculation, not the safety of an installation.
Turn the claim into a falsifiable test. First, Remove one field and identify the ambiguity. Then change one consequential condition: Version the receipt after compensation is switched. State which observation would count against the explanation.
The evidence receipt for “Build a reproducible power receipt” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
CHAPTER 21 OF 24 · Report and continue
21. Repair the reactive power is wasted energy shortcut
Start by naming the physical or logical quantity rather than the displayed number. Reactive exchange can increase current and capacity requirements without being identical to net real-energy consumption.
Work the original teaching case in a visible order. The load has80 kvar but the0.5 h real-energy result remains30 kWh.
Make the reasoning independently checkable. Keep physical effects and billing rules separate.
Test transfer by changing one condition while holding the others still. Reverse Q while preserving its magnitude.
Keep the claim boundary visible. The word wasted hides mechanism and context.
Before accepting the result, predict what should change and what should remain fixed. Apply this check: Keep physical effects and billing rules separate. Then run the changed case: Reverse Q while preserving its magnitude. Record any mismatch without rewriting the original assumptions.
The evidence receipt for “Repair the reactive power is wasted energy shortcut” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
CHAPTER 22 OF 24 · Report and continue
22. Repair the better power factor saves the same energy shortcut
Keep observation, transformation and conclusion in separate evidence columns. Reducing Q can reduce current-related losses, but the size of any real-energy change depends on the network and operating profile.
Work the original teaching case in a visible order. The simplified compensation holds P fixed and makes no loss calculation.
Make the reasoning independently checkable. Measure losses or model resistance explicitly.
Test transfer by changing one condition while holding the others still. Set conductor resistance to zero in a thought experiment.
Keep the claim boundary visible. A power-factor improvement is not a universal kWh saving percentage.
Build a compact audit receipt with input, unit, model, calculation, observation and limit. Use this independent check: Measure losses or model resistance explicitly. For transfer, Set conductor resistance to zero in a thought experiment. Preserve the first result beside the changed one.
The evidence receipt for “Repair the better power factor saves the same energy shortcut” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
Use the short answer only after the boundary and unit have been fixed. The ledger can check whether real, reactive and apparent quantities are being compared consistently.
Work the original teaching case in a visible order. It can flag a mixed-unit claim before engineering review.
Make the reasoning independently checkable. Write the decision beside the evidence.
Test transfer by changing one condition while holding the others still. Change the decision from classroom comparison to capacitor-bank design.
Keep the claim boundary visible. Design, switching and safety require competent electrical engineering and current rules.
Ask a second reader to reproduce the decision from the stated evidence. They should be able to apply: Write the decision beside the evidence. Next, Change the decision from classroom comparison to capacitor-bank design. If the conclusion changes, name the changed condition explicitly.
The evidence receipt for “State the supported decision” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
CHAPTER 24 OF 24 · Report and continue
24. Continue through verified utility routes
Trace the mechanism in the order another reader would need to reproduce it. This distinction sits beside energy-versus-demand, connected-versus-coincident load and reliability metrics.
Work the original teaching case in a visible order. Choose the sibling whose denominator matches the next question.
Make the reasoning independently checkable. Keep metric ownership explicit.
Test transfer by changing one condition while holding the others still. Move to load factor only after fixing the energy and peak windows.
Keep the claim boundary visible. This article owns the real-versus-reactive power distinction under a declared AC model.
Turn the claim into a falsifiable test. First, Keep metric ownership explicit. Then change one consequential condition: Move to load factor only after fixing the energy and peak windows. State which observation would count against the explanation.
The evidence receipt for “Continue through verified utility routes” records source, input, transformation, changed-context prediction and limit on separate lines. Keep waveform, phase basis, sign convention, meter boundary and time window attached to every number.
Continue through the verified reading route
Energy and peak-demand windows
Energy and peak-demand windows. Use this when interval energy and maximum demand are being combined into load factor.
Connected load and coincident peak demand
Connected load and coincident peak demand. Use this when nameplate totals are confused with simultaneous measured demand.
SAIFI, SAIDI and CAIDI
SAIFI, SAIDI and CAIDI. Use this when service reliability, customer counts and interruption duration are the real question.
Frequently asked questions
Does reactive power appear as ordinary kWh consumption?
No. Real-energy and reactive-energy registers are distinct; tariffs and meter channels must be read as defined.
Can correction reduce apparent power without reducing real power?
Yes in the stipulated model: reducing net Q reduces |S| while P remains fixed.
Does a higher power factor guarantee a fixed energy saving?
No. Any real-loss change depends on the network, resistance and operating profile.
Authoritative sources and further reading
NIST SP 330 SI units. Official SI reference distinguishing power in watts and the IEC special name var for reactive power.
U.S. Department of Energy: Reducing Power Factor Cost. Official DOE source illustrating real power, reactive power, apparent power and capacity implications.
U.S. Department of Energy electricity-bill training. Official training material keeping real power, reactive power and power-factor provisions distinct.
NIST Guide to the SI. Official unit reference for watt, joule, volt and ampere relationships.
Return to the Utilities subject guide, The Gold Standard of Education, or the How X Works hub.