- Impulse Test of Transformer Definition: An impulse test of a transformer checks its ability to withstand high-voltage impulses, ensuring its insulation can handle sudden spikes in voltage.
- Lightning Impulse Test: This test uses natural lightning-like voltages to assess transformer insulation, identifying weaknesses that could cause failure.
- Switching Impulse Test: This test simulates voltage spikes from switching operations in the network, which can also stress transformer insulation.
- Impulse Generator: An impulse generator, based on the Marx circuit, creates high-voltage impulses by charging capacitors in parallel and discharging them in series.
- Testing Performance: The test procedure involves applying standard lightning impulses and recording voltage and current waveforms to identify any insulation failures.
An impulse test of a transformer checks insulation against short high-voltage spikes. Lightning strokes on tall transmission lines send a surge along the line conductor to a power transformer. Switching in the network also produces switching impulses. Those peaks are often a few times the system voltage; about 3.5 times is a typical order of magnitude, not a single IEC limit.
A power-frequency withstand test does not cover those fast spikes. IEC 60076-3 therefore includes a lightning-impulse type test, and a switching-impulse test on higher-voltage classes. Both are dielectric tests of the transformer insulation.
Lightning Impulse
A natural lightning stroke does not have one fixed wave shape. Records of lightning on power systems are grouped into three basic shapes used in tests:
- Full wave
- Chopped wave and
- Front of wave
A real stroke may not match those three shapes exactly. The defined waves still set a minimum impulse dielectric strength for the transformer.
If the surge travels along the line before it reaches the transformer, the shape may be a full wave. If an insulator flashes over after the peak, the wave is chopped.
If lightning hits the transformer terminals, the impulse voltage rises until a flashover relieves it. At that instant the voltage collapses and can form a front-of-wave shape.
Each wave shape stresses insulation in a different way. Any of them can puncture insulation, so the lighting impulse test of transformer is a type test under IEC 60076-3.
Switching Impulse
Switching overvoltages often have front times of several hundred microseconds and then decay. IEC 60060-1 uses a switching-impulse wave of 250 μs front time and 2500 μs time to half-value, with stated tolerances. IEC 60076-3 applies that SI test to higher Um ratings; it is not required on every distribution transformer.
The lightning-impulse test checks that the transformer insulation can withstand the lightning overvoltage expected in service.
The pictured impulse generator uses a Marx circuit. The impulse capacitors Cs (12 capacitors of 750 nF on this kit) are charged in parallel through the charging resistors Rc (28 kΩ) (highest charging voltage 200 kV on this kit). When charging reaches the set value, spark gap F1 is triggered. After F1 breaks down, the potential at the next stage (points B and C) rises. Series resistors Rs are low compared with discharging resistors Rb (4,5 kΩ) and charging resistor Rc, and low-ohmic discharging resistor Ra is held off by auxiliary spark-gap Fal, so the voltage across spark-gap F2 rises and F2 breaks down.
The spark-gaps then break down in sequence and the capacitors discharge in series. High-ohmic resistors Rb shape switching impulses and low-ohmic resistors Ra shape lightning impulses. When the auxiliary gaps break down, Ra is placed in parallel with Rb after a delay of a few hundred nanoseconds.
That staging lets the Marx generator erect without the later gaps firing too soon.
Wave shape and peak value are measured with an impulse analysing system (DIAS 733 on this kit) on the voltage divider. Peak voltage is set by the number of series stages and the charging voltage. Parallel or series-parallel generator connections raise discharge energy by paralleling some capacitors during discharge.
Series and discharge resistors of the generator set the impulse shape.
Front time is estimated from:
For R1 >> R2 and Cg >> C (15.1)
Tt = .R.C.123
and time to half-value from
T ≈ 0,7.R.C
Labs then trim the circuit from experience. Those compact formulae are teaching estimates, not a substitute for IEC 60060-1 wave-shape definitions.
Performance of Impulse Test
The lightning-impulse test uses standard negative-polarity waves. IEC 60060-1 defines front time T1 and time to half-value T2.
Standard lightning impulse
Front time T1 = 1,2 μs ± 30%
Time to half-value T2 = 50 μs ± 20%
Low-voltage windings of high rating, and windings with high input capacitance, often cannot hold the exact standard wave. Negative polarity reduces stray flashover in the test circuit. Most objects need resistor and capacitor adjustment, using earlier tests on similar units or a pre-calculation.
IEC 60076-3 uses one reduced full wave (RW), often at 75% of full amplitude, then three full waves (FW) at the rated impulse level. Some ratings also require chopped waves; this page describes the full-wave sequence. Voltage and current are stored on a digital transient recorder and compared between the reduced and full levels. On an OLTC transformer one phase is tested at rated tap voltage and the other two phases at the extreme taps.
Connection of Impulse Test
Dielectric tests check the insulation level of the unit. The generator produces the specified 1.2/50 μs lightning voltage wave. One reduced voltage impulse, 50% to 75% of full test voltage, is followed by three full-voltage impulses.
On a three phase transformer each phase is impulse-tested in turn.
The impulse is applied to each line terminal in turn, with the other terminals earthed.
Current and voltage waves are recorded. Distortion between the reduced and full-wave records is the usual failure sign.
