A Review of Dielectric Strength with Distorted Lightning Impulses

Abstract

Non-standard and distorted lightning impulse voltages are often actual dielectric stresses on high voltage insulation, either in laboratory tests or in service. The standard lightning impulse 1,2/50 s conveniently allows comparing the results obtained in different laboratories. However, it is not always possible to generate this impulse voltage in a laboratory because of the presence of distributed electrical parameters, such as inductance and capacitance, associated with big dimensions of circuits and apparatus. In large electric networks the lightning strength voltages produced by direct strokes and travelling surges present marked variations of time-to-front, time-to-tail, oscillations and superimposed overshoots, giving dielectric results quite different from conventional impulses. Much work has been done in the past in order to forecast the breakdown probability for the distorted impulses in insulations under test or, at least, the equivalence from a point of view of the disruptive discharge probability between the distorted voltage wave-shapes (or non-conventional impulses) and standard lightning impulses. The classic approach of the constant area criteria, which considers the time integral of the difference between the applied voltage and a constant voltage, has practical limits depending on the electrode configuration, voltage waveforms and voltage polarity. In the last decade a new approach was introduced for evaluating the parameters of impulse voltages with superimposed overshoot and oscillations on the crest. This approach, the test voltage factor or k-factor, has an experimental basis which facilitates calculation of the equivalent impulse voltage for distorted test voltages. The output of this method is the value of the lightning impulse parameters of an equivalent standard impulse, which should be dielectrically equivalent, i.e. have the same disruptive discharge probability. Then, with some restrictions, this method would estimate the disruptive probability with distorted impulses. Since the first proposal of a single test voltage factor, included in the latest h-v test standards (IEC 60060-1, 2010 and IEEE Std. 4, 2013), recent experiments have shown different values of k-factors in air gaps. The aim of this contribution is to highlight some relevant aspects of insulation performance with distorted lightning impulses.