How to Calculate Copper Loss in Transformer Precisely

Calculate transformer copper loss by adding the I²R losses of every winding and phase, using RMS current and DC resistance corrected to the required winding temperature. For a single-phase transformer:

Pcu,dc = I₁² × R₁ + I₂² × R₂

This gives the DC-resistance component of winding loss. A short-circuit test also captures additional losses, so its reading will usually be higher. Matching those two results starts with checking temperature, connection type and measurement boundaries.

1. Choose the Loss Quantity

“Copper loss” needs a defined calculation boundary. Decide which result the project requires before selecting the formula:

• DC-resistance winding loss: the sum of Irms² × Rdc(T) across all winding paths.
• Total AC winding loss: includes additional conductor losses caused by nonuniform current distribution, including skin and proximity effects.
• Transformer load loss: includes winding losses and leakage-flux losses in other conducting parts.

A short-circuit wattmeter reading represents load loss rather than isolated DC copper loss. Calling the complete reading “pure copper loss” can distort temperature corrections and comparisons with resistance-based calculations.

Use Define → Normalize → Verify to organize the calculation. Define the requested loss quantity, normalize the input conditions, then verify the result against a compatible measurement or engineering check.

2. Measure and Normalize the Inputs

Capture Stable Winding Resistance

Reliable resistance data requires separate current-injection and voltage-sensing connections. Use a suitable transformer winding-resistance tester and follow this sequence:

1. Isolate the transformer, prove it dead, and follow the approved grounding and test procedure.
2. Record the tap position, vector group, conductor material, rated current and winding temperature.
3. Connect Kelvin sensing leads directly to the intended terminals, inside the current connections.
4. Wait for stable readings, then record resistance, units, test current and terminal pair.
5. Complete the instrument’s discharge sequence before disconnecting leads, and demagnetize as required by the test procedure.

A drifting reading is not ready for calculation. Transformer inductance can delay stabilization, particularly with closed-delta windings. Never disconnect the test circuit while DC current flows: stored magnetic energy can produce hazardous voltage.

Excessive test current can heat the winding and change the resistance being measured. Select the current from the instrument instructions and approved procedure. Increasing current simply to obtain a faster reading can introduce an error.

Correct Resistance to the Target Temperature

Resistance must correspond to the temperature stated in the calculation. Using the copper convention selected for this example:

Rtarget = Rmeasured × (234.5 + Ttarget) ÷ (234.5 + Tmeasured)

Where:

• Rmeasured = measured DC resistance, Ω.
• Tmeasured = winding temperature during measurement, °C.
• Ttarget = required calculation temperature, °C.
• Rtarget = resistance corrected to the target temperature, Ω.

The material constant and reference temperature must match the governing test method. Do not automatically apply the copper constant to aluminum windings or assume that 75°C applies to every transformer.

Ambient temperature is a valid substitute only when the required thermal conditions justify it. A recently disconnected transformer may still be cooling. Preserve the raw resistance and measured temperature alongside the corrected result.

Match Line Quantities to Phase Quantities

The phase-based three-phase formula requires actual phase current and phase resistance.

Pcu,dc = 3 × [(Iph,H)² × Rph,H + (Iph,L)² × Rph,L]

Here, H and L identify the HV and LV windings; Iph and Rph are phase current and phase resistance.

A terminal-resistance shortcut works for balanced, symmetrical windings under sinusoidal loading.

Pcu,dc = 1.5 × [IHV² × RHV,LL + ILV² × RLV,LL]

Here, IHV and ILV are operating line currents. RHV,LL and RLV,LL are DC resistances measured between line terminals, corrected to the required temperature.

Both star and closed-delta connections reduce to this expression when their assumptions hold.

• Star: RLL = 2 × Rphase; Iphase = Iline.
• Closed delta: RLL = ⅔ × Rphase; Iphase = Iline ÷ √3.

Check all three terminal pairs before using the shortcut. External parallel paths, unequal phase resistances, unbalanced loading and harmonic circulation require a more detailed winding model.

3. Calculate a Three-Phase Example

This example uses assumed inputs to demonstrate a reproducible calculation. Consider a 500 kVA, 10 kV/400 V, copper-wound Dyn11 transformer with balanced sinusoidal currents and equal phase resistances. All resistance readings correspond to a verified winding temperature of 20°C; the selected target temperature is 75°C.

Input	HV winding	LV winding
Rated line voltage	10,000 V	400 V
Rated line current	28.8675 A	721.6878 A
Terminal-to-terminal DC resistance at 20°C	1.800 Ω	0.001000 Ω

Rated line current follows from three-phase apparent power and line voltage.

Iline = S ÷ (√3 × Vline)

Use S in VA and Vline in volts to obtain current in amperes.

Calculate Copper Loss at 20°C

The two windings contribute 3,031.25 W at rated current and 20°C.

Pcu,20 = 1.5 × [28.8675² × 1.800 + 721.6878² × 0.001000]

Pcu,20 = 2,250 + 781.25

Pcu,20 = 3,031.25 W

Correct the Result to 75°C

Temperature correction increases the calculated DC copper loss to approximately 3.686 kW.

KT = (234.5 + 75) ÷ (234.5 + 20)

KT = 1.216110

Pcu,75 = 3,031.25 × 1.216110

Pcu,75 = 3,686.33 W

Half rated current produces approximately 922 W at the same 75°C winding temperature.

Pcu,half = 0.5² × 3,686.33

Pcu,half = 921.58 W

The displayed decimal places support arithmetic checks rather than measurement accuracy. Round the reported result to match the uncertainty of the inputs.

4. Reconcile the Short-Circuit Test

Obtain a Comparable Load-Loss Reading

A controlled short-circuit test provides an independent load-loss measurement. Qualified personnel short one winding and energize the other from a suitable current-limited source at rated frequency, starting from zero and increasing the reduced voltage until the specified current flows. Never apply rated terminal voltage to this setup.

The comparison requires matching current, temperature, tap position and measurement boundaries. Account for instrument and shorting-connection losses using the approved test procedure.

Correct the Components Separately

The difference between corrected load loss and calculated DC copper loss represents additional load losses within the selected boundary. Suppose the example’s corrected short-circuit result is 3,600 W at rated current and 20°C:

Padditional,20 = Pload,20 − Pcu,20

Padditional,20 = 3,600 − 3,031.25

Padditional,20 = 568.75 W

Ohmic and stray components require different temperature corrections under the method used here. For this all-copper example:

Pload,75 = Pcu,20 × KT + Padditional,20 ÷ KT

Pload,75 = 3,686.33 + 467.68

Pload,75 = 4,154.01 W

Applying the copper-resistance factor to the entire 3,600 W reading would overstate the corrected result.

Incorrect calculation: 3,600 × 1.216110 = 4,378.00 W

That shortcut produces a result approximately 5.4% higher than the component-based calculation.

A negative residual requires investigation. Check resistance units, line/phase conventions, temperature, current scaling and instrument corrections before interpreting it as a transformer characteristic.

5. Check Measurement Uncertainty

Low power factor makes loss measurements sensitive to phase error. For sinusoidal voltage and current:

P = V × I × cos(φ)

Approximate relative power error from phase error:

|ΔP| ÷ P ≈ |tan(φ) × Δφ|

Here, φ is the voltage–current phase angle, and Δφ is the net measurement-system phase error in radians.

A 0.05° phase error at power factor 0.05 contributes approximately 1.74% power error. Check the combined behavior of the power analyzer, current sensor and voltage channels at the actual test conditions.

Current uncertainty has twice its relative influence in an I²R calculation. For one loss contribution, small worst-case input limits give:

Relative loss error ≲ 2 × relative current error + relative resistance error

A ±0.5% current limit and ±0.5% resistance limit already imply approximately ±1.5% before temperature effects. Use this as an initial sensitivity check; a formal uncertainty budget must account for all material contributions and their relationships.

6. Calculate Operating Energy Loss

Account for Harmonic Current

Total RMS winding current captures DC-resistance heating, but it does not describe every frequency-dependent conductor loss.

Pdc component = Rdc × (I₁² + I₂² + I₃² + …)

Here, I₁ is the fundamental RMS current, while I₂, I₃ and subsequent terms are harmonic RMS currents in the winding path.

Delta circulation can hide winding current from an upstream line-current measurement. Triplen harmonics may circulate internally, so a clean upstream waveform does not establish their absence inside the transformer.

Total AC winding-loss calculations require harmonic-resolved AC resistance or a validated model. Calculate each winding path and harmonic contribution, then add them:

Loss per contribution = harmonic RMS current² × AC resistance at that harmonic and temperature

This calculation excludes losses in the tank, clamps and other structural parts.

Square Each Load Interval

Energy calculations must average squared current rather than square average current.

Ecu = ∑ [Pcu,rated(Tj) × kj² × Δtj]

Where:

• ∑ means add all time intervals.
• Pcu,rated(Tj) = rated-current copper loss at the temperature of interval j.
• kj = interval current ÷ rated current.
• Δtj = interval duration.

Use kW and hours to obtain kWh.

Equal periods at 50% and 100% current give a mean squared load of 0.625.

Mean squared load = (0.5² + 1.0²) ÷ 2 = 0.625

Squared average load = [(0.5 + 1.0) ÷ 2]² = 0.5625

Squaring average load understates loss even before temperature changes enter the calculation.

7. Record a Reproducible Result

Another engineer should be able to reconstruct the result from the report. Record transformer identification, tap, connection diagram, conductor material, raw resistance and temperature, current basis, correction method, instrument calibration, uncertainty and the exact loss quantity reported.

Current efficiency requirements reinforce the need to separate operating estimates from compliance results. As checked on October 20, 2026, the U.S. distribution-transformer program lists a June 2026 information request and an April 23, 2029 compliance date for amended standards. For covered equipment, establish compliance through the prescribed test procedure.

FAQs

How Do You Calculate Copper Loss in a Single-Phase Transformer?

Add the primary and secondary I²R contributions.

Pcu,dc = I₁² × R₁ + I₂² × R₂

Use RMS winding currents and DC resistances corrected to their respective winding temperatures. This calculates the DC-resistance component of winding loss.

How Do You Find Copper Loss in a Transformer Without Winding Resistance?

A corrected short-circuit test provides load loss, which approximates copper loss when additional losses are negligible. It cannot independently isolate pure DC copper loss. Obtain resistance data when that distinction affects the result.

Is Short-Circuit Loss Equal to Copper Loss?

Only approximately in a simplified transformer model. Actual short-circuit measurements include additional load losses. State whether the reported figure means DC copper loss or total load loss.

Why Is Copper Loss at Half Load One-Quarter of Full-Load Loss?

Halving current reduces I² to one-quarter. The relationship assumes unchanged resistance and winding temperature. Actual half-load operation may have a different temperature.

Can I Use Line Current With Measured Terminal Resistance?

Yes, for balanced, symmetrical windings under the sinusoidal conditions defined above. Calculate each three-phase winding contribution as:

Pwinding,dc = 1.5 × Iline² × RLL

Add the HV and LV contributions. Do not insert terminal resistance directly into a phase-resistance formula.

Can Transformer Copper Loss Be Calculated From kVA Alone?

No; kVA and voltage establish rated current, but they do not establish resistance. You still need winding resistance and temperature, or an appropriately defined loss-test result.