Purpose of open-circuit and short-circuit tests on a transformer

Ask which loss dominates when current is small versus when voltage is small.

Model answer.

Both tests let us find the losses and the equivalent-circuit parameters without actually loading a large transformer.

Open-circuit (OC) test

• Side: Usually done on the low-voltage (LV) side with the HV side open. Rated voltage of the LV side is easy to supply, and the safe, low-range meters can be used.
• What is applied: Rated voltage at rated frequency.
• What is measured: Voltage, no-load current and input power.
• Why it gives core loss: The no-load current is only a few percent of rated current, so the copper loss is negligible. Because rated voltage gives rated flux, the wattmeter reads almost purely the core (iron) loss.
• Parameters: The shunt branch: core-loss resistance R₀ and magnetising reactance X₀.

Short-circuit (SC) test

• Side: Usually done on the high-voltage (HV) side with the LV side shorted. HV-side rated current is smaller and easier to supply and measure.
• What is applied: A reduced voltage, typically around 5% to 10% of rated, raised until rated current flows.
• What is measured: Voltage, current and input power.
• Why it gives copper loss: The applied voltage is small, so the flux and the core loss are negligible. With rated current flowing, the wattmeter reads the full-load copper loss.
• Parameters: The series branch: equivalent resistance R_eq and leakage reactance X_eq, referred to the HV side.

Using the results. Core loss is constant; copper loss scales with the square of the load fraction. Together they give the efficiency at any load and the voltage regulation. Maximum efficiency occurs at the load where the copper loss equals the core loss.

Common mistake. Saying the OC test is done on the HV side "because that is the primary." The choice of side is about convenience and safety, not which winding is called the primary.

Likely follow-up. "Why is the no-load power factor so low?" The no-load current is mostly magnetising current, which lags the voltage by nearly 90°.