contact us
What Is Automatic Test Equipment (ATE)?
products
- Main Products
technical articles
Company News
An impulse test applies a short, high-voltage pulse to insulation and asks whether it survives. IEC 61180-1 defines the waveform, the tolerances and the measurement procedure for low-voltage equipment, and that definition is what makes the test reproducible. The peak voltage is the number people quote; the waveform is what the standard actually specifies. Two impulses with the same peak and different front times stress insulation differently, and a generator that produces a 2 microsecond front is not producing the specified test.
The waveform, defined by three numbers
| Front time | 1.2 microseconds, with a tolerance of plus or minus 30 percent |
| Time to half value | 50 microseconds, with a tolerance of plus or minus 20 percent |
| Peak value | The specified test voltage, with a tolerance typically of plus or minus 3 percent |
The front time is not the time to peak. It is defined on the steep rising part of the curve, between 30 percent and 90 percent of peak, extrapolated to the virtual origin. The time to half value is measured from the virtual origin to the point where the wave has fallen to half the peak. Both definitions exist because a real impulse is not a mathematical curve, and without a defined construction two laboratories would measure different numbers on the same oscillogram.
Why the virtual origin matters
The rising edge of an impulse is not a straight line from zero. It curves, and the initial part is often noisy. The virtual origin is the point where a straight line through the 30 percent and 90 percent points crosses zero. Measuring from that point rather than from the first visible rise is what makes the front time comparable between laboratories.
In practice, a digital recorder with a defined sample rate and a cursor function is the tool; eyeballing a printout is not. A 1.2 microsecond front on a screen without at least 20 samples across the front cannot be measured to the tolerance the standard allows. At 6 kV the same rule applies: the front is 1.2 microseconds whether the peak is 1 kV or 6 kV, so the sample rate requirement does not relax at higher voltage.
What changes the waveform in a real test
- The load. The generator's waveform is specified into a defined load, often a resistive divider or a capacitor. Connecting a product with capacitance changes the front time, usually making it slower.
- The leads. Long leads add inductance, which lengthens the front and can introduce oscillation. Keep the leads short and routed away from grounded metal.
- The generator settings. Front time is set by the generator's series resistance and load capacitance; a setting that was correct for one product is not correct for the next. Adjust and verify for each test object.
The test sequence on a product
Product standards specify how many impulses, at what polarity, and between which points. Three rules that apply broadly:
- Both polarities. Insulation does not fail symmetrically. A test in one polarity only tests half the insulation behaviour.
- Three impulses per condition is the common requirement, with the criterion being no breakdown in any of them.
- Voltage applied between the specified points, which usually means live parts to accessible parts, and live parts to each other where the circuit allows.
Reading the result
Acceptance is the absence of breakdown, where breakdown means a sudden collapse of voltage across the insulation, usually visible as a sharp drop in the recorded waveform. Three findings worth distinguishing:
- Clear breakdown — a step in the waveform followed by a much lower voltage. A fail, with no interpretation needed.
- Partial discharge activity — small high-frequency oscillations superimposed on the wave, without collapse. Requires further investigation, often with a partial discharge test, and is not automatically a fail.
- Waveform distortion without collapse — a changed front time caused by the load. Not a fail, but the test is invalid because the specified waveform was not applied.
The third is the one that wastes the most time, because the product passes and the report is unusable. Verifying the waveform before testing product, on the same setup, is the way to avoid it.
Verification of the generator
An impulse generator is verified with a calibrated divider and a recorder, at each voltage level used, into a load representative of the product. The verification record should include the measured peak, front time, time to half value and the load used. A generator verified only at one voltage and one load has been verified for one condition.
Two further checks matter between calibrations: that the voltage divider ratio has not drifted, and that the recorder's time base is correct. A recorder with a 5 percent time base error produces a front time error of 5 percent, which consumes a fifth of the allowed tolerance before the generator is considered.
The electrical safety test range includes impulse generators with waveform verification outputs; the site index links the high-voltage dividers and recorders used to measure them.





