Transformer Efficiency: Full-Load Numerical and the Maximum-Efficiency Condition
Transformer efficiency numericals appear in nearly every Electrical Machines exam, and the "condition for maximum efficiency" is a favourite theory question. Both come from the same two kinds of loss.
The two losses
- Iron (core) loss, Pi: hysteresis and eddy currents in the core. It depends on voltage and frequency, so it is essentially constant from no load to full load.
- Copper loss, Pcu: I²R heating in the windings. It depends on the square of the load current, so it changes with load.
Efficiency η = Output power / (Output power + Pi + Pcu)
Worked numerical: efficiency at full load
A 50 kVA transformer has iron loss 500 W and full-load copper loss 800 W. Find the efficiency at full load, 0.8 power factor lagging.
Output power = 50 000 VA × 0.8 = 40 000 W
Total losses = 500 + 800 = 1 300 W
η = 40 000 / (40 000 + 1 300) = 40 000 / 41 300 = 0.9685
Efficiency at full load = 96.85 %
At a fraction of full load
If the load is a fraction x of full load (x = 0.5 for half load):
- Output = x × (kVA × pf)
- Copper loss = x² × (full-load copper loss)
- Iron loss unchanged
At half load (x = 0.5), same power factor:
Output = 0.5 × 40 000 = 20 000 W
Pcu = 0.25 × 800 = 200 W
Pi = 500 W
η = 20 000 / (20 000 + 500 + 200) = 20 000 / 20 700 = 96.62 %
Half load is slightly less efficient than full load here, which tells us something.
The condition for maximum efficiency
Efficiency rises with load at first (the constant iron loss is spread over more output) and then falls (copper loss grows as the square). So there is a peak. Where?
Write the efficiency in terms of load current I, with output = V·I·cosφ:
η = V·I·cosφ / (V·I·cosφ + Pi + I²R)
Maximizing η is the same as minimizing the loss per unit output, which is (Pi + I²R) / (V·I·cosφ), or Pi/I + I·R after dividing through. Differentiate with respect to I and set it to zero:
−Pi/I² + R = 0 → Pi = I²R
Efficiency is maximum when the variable loss equals the constant loss: copper loss = iron loss.
The load at which this happens
If full-load copper loss is Pcu and the load fraction is x, the copper loss is x²·Pcu. Setting it equal to Pi:
x = √(Pi / Pcu)
For our transformer:
x = √(500 / 800) = √0.625 = 0.7906
So maximum efficiency occurs at about 79 % of full load.
The maximum efficiency itself
At that load:
Output = 0.7906 × 40 000 = 31 623 W
Pcu = Pi = 500 W, so total losses = 1 000 W
η_max = 31 623 / (31 623 + 1 000) = 96.93 %
Compare: 96.85 % at full load, 96.62 % at half load, 96.93 % at the peak. The peak sits between them, exactly as the theory predicts.
Why designers care
Distribution transformers run at light load much of the day, so they are designed with a low iron loss and a peak around 50–70 % load. Power transformers that run near full load all day are designed so the peak sits closer to full load. That is also the idea behind all-day efficiency, which averages over a 24-hour load cycle rather than one load point.
Exam checklist
- State both losses and which one varies with load.
- Write η with the variable load fraction x.
- Give the condition Pi = x²·Pcu and the formula x = √(Pi/Pcu).
- Compute output at that x and get η_max. Remember the power factor stays the same.
Sumantrak™ is an AI mentor built for Indian engineering, diploma, BCA and BSc CS/IT students — Electrical Machines, Circuits, Power Systems, and the rest of your syllabus, explained like this. Free to start →