Transformer Short Circuit Impedance and Leakage Reactance Testing

Short circuit impedance, also called leakage impedance or percent impedance, is one of the most fundamental parameters of a power transformer. It determines how much fault current the transformer will deliver to a downstream short circuit, controls voltage regulation under load, and governs how the unit shares load when paralleled with another transformer. The value is set during design by the physical geometry of the windings: the number of turns, the radial and axial spacing between the high- and low-voltage coils, and the geometry of the leakage flux path between them. It is printed on the nameplate as a percentage, typically in the range of 4–8% for distribution transformers and up to 12–15% for large power transformers with high fault-current limiting requirements.

Because leakage impedance depends on winding geometry, any event that moves the windings, a through-fault producing large electromagnetic forces, an internal fault, seismic loading, or rough handling during transport, can change the measured impedance. That change is detectable in the field before the transformer is returned to service, and it is one of the most reliable indicators of mechanical winding damage.

The physics of leakage impedance

When load current flows through a transformer, not all of the magnetic flux produced by the primary winding links the secondary winding perfectly. Some flux, leakage flux, follows paths through the insulation and oil between the windings rather than through the core. This leakage flux produces a voltage drop that opposes the applied voltage, creating an effective series reactance in each winding. The sum of the primary and secondary leakage reactances, referred to one side, is the leakage reactance of the transformer. Combined with the series winding resistance, this gives the total leakage impedance.

The leakage flux path depends directly on the geometry of the space between the windings. If the high-voltage and low-voltage coils are displaced axially, one coil has shifted up or down relative to the other, the leakage flux pattern changes. If a coil has deformed radially, buckled inward or outward, the effective air gap changes. Either type of displacement changes the leakage reactance, and therefore the measured percent impedance. This is why impedance measurement is a mechanical condition indicator, not just an electrical one.

Test procedure

The short circuit impedance test is simple. The low-voltage winding terminals are shorted together, and a reduced AC voltage is applied to the high-voltage winding. The voltage is increased until rated current flows in the high-voltage winding. At that point, three measurements are recorded: the applied voltage (V_sc), the current (I), and the input power (W). From these, the percent impedance, the resistance component, and the reactance component are calculated.

Percent impedance is V_sc divided by rated voltage, expressed as a percentage. The resistance component (percent R) is derived from the power measurement. The reactance component (percent X) is calculated from percent impedance and percent R using the right-triangle relationship. The reactance component is the most diagnostically useful number because resistance can be affected by temperature and connection resistance, while reactance reflects only the geometric leakage flux path.

Field tests are rarely performed at full rated current, the test set required to drive rated current through a large transformer is impractically large. Instead, the test is run at a reduced current level (commonly 10–25% of rated) and the results are corrected to rated conditions. The correction for resistance must account for winding temperature at the time of test, referenced to the nameplate reference temperature (usually 85°C for ONAN cooling). Reactance does not require temperature correction.

The test is run on each winding pair independently for two-winding transformers (HV-LV), and on all three winding combinations for three-winding units (HV-LV, HV-TV, LV-TV). Each pair produces its own impedance value, and each is compared independently to the nameplate or factory test data.

Interpreting deviations from nameplate

The measured percent impedance is compared to the nameplate value. The allowable deviation before the result is flagged as abnormal depends on the applicable standard and transformer type.

IEEE C57.12.00 (standard for liquid-immersed distribution and power transformers) allows a manufacturing tolerance of ±7.5% of nameplate impedance for transformers below 10 MVA, and ±7.5% for larger units where no tolerance is specified, though most large power transformer purchase specifications tighten this to ±5% or ±3%. IEC 60076-1 specifies ±10% for two-winding transformers below 100 MVA and ±7.5% above that. NETA ATS Table 100.19 recommends investigating any deviation of more than 3% from the factory test data for in-service diagnostic testing, not from the nameplate tolerance band, but from the specific factory test result recorded in the test report.

The distinction between comparing to nameplate and comparing to factory test data matters. A nameplate that reads 7.5% impedance may have been measured at 7.6% during factory acceptance testing. If the field result is 7.8%, that is a 2.6% deviation from the factory measurement, within NETA's 3% threshold, but if the nameplate test report is unavailable and the field result is compared to the 7.5% nameplate value, the calculated deviation is 4%. Always use the factory test report value when available.

An increase in measured impedance from the factory baseline indicates increased leakage flux path, the windings have moved apart, typically from axial displacement or radial buckling that increases the effective gap. A decrease indicates the windings have moved closer together, which can occur with inward radial collapse of the inner winding under compressive fault forces. A symmetric change across all three phases suggests a systematic shift; an asymmetric result, where one phase deviates significantly from the other two, points to a localized fault on that phase.

Post-fault assessment and the relationship to SFRA

The most common trigger for short circuit impedance testing in the field is a through-fault event, a downstream fault that drove large fault current through the transformer before the protective relay cleared it. IEEE C57.109 (guide for through-fault protection) defines cumulative fault duty curves that estimate when a transformer has absorbed enough through-fault energy to warrant mechanical inspection. Short circuit impedance testing is the first electrical test in the post-fault assessment because it is fast, requires minimal equipment, and gives a definitive bulk measurement of winding geometry change.

Sweep frequency response analysis (SFRA) is complementary to short circuit impedance in post-fault assessment. SFRA measures the transformer's frequency response across a wide spectrum and is sensitive to subtle winding deformation, particularly at mid- and high-frequency resonances that reflect localized mechanical changes. Short circuit impedance is a single-frequency bulk measurement, it averages the leakage flux geometry across the entire winding, while SFRA resolves spatial variation within the winding. A transformer can show a normal short circuit impedance result but an abnormal SFRA signature if the deformation is localized enough that its effect on the bulk average is below the impedance measurement threshold. For this reason, both tests are recommended after significant through-fault events on large power transformers. See our article on SFRA Transformer Testing for a detailed explanation of frequency response interpretation.

The post-fault test sequence for a large power transformer typically runs in this order: visual inspection of the tank, bushings, and conservator; turns ratio test on each tap position to check for turn-to-turn shorts; winding resistance to verify connections and detect open circuits; short circuit impedance to check for gross winding displacement; SFRA for sensitivity to localized deformation; and dielectric testing (power factor, insulation resistance) to confirm insulation integrity was not compromised during the fault event.

Parallel operation and load sharing

Short circuit impedance testing is also performed at acceptance when transformers will be operated in parallel. Two transformers in parallel share load in inverse proportion to their impedances, a unit with lower impedance carries a larger share of the load. If the impedances are not closely matched, the lower-impedance unit can be overloaded while the higher-impedance unit runs light. As a practical rule, transformers paralleled in service should have impedances within 7.5% of each other. Field-measured impedance values, temperature-corrected and referred to the same MVA and kV base, are compared before paralleling to verify compatibility. A unit with a field impedance that differs significantly from its nameplate, beyond manufacturing tolerance, may parallel poorly with units whose measured impedance is close to nameplate.