Slow-front overvoltages are oscillatory transients with fronts from tens to thousands of microseconds. They arise when lines are energized or reclosed, when faults start or are cleared, on load rejection, when inductive or capacitive currents are switched and from distant lightning. Insulation co-ordination represents them with the standard 250/2500 µs switching impulse.
Short answer
A slow-front overvoltage is an oscillatory transient whose front lasts from tens to thousands of microseconds, with a tail of similar duration. In high- and extra-high-voltage networks it is usually the one that sets the air clearance phase-to-phase and phase-to-earth. It originates when lines are energized or reclosed, when faults start or are cleared, on load rejection, when inductive or capacitive currents are switched and from lightning far from the line. IEC 60071-2 represents it with the standard switching impulse, 250 µs to the peak and 2500 µs to half of the value on the tail.
What causes them
Clause 2.3.3 of IEC 60071-2 lists five origins:
- Line energization and reclosing. This is the most studied cause. Its magnitude depends on the breaker, on the source and on the length and type of line. Three-phase reclosing can give high overvoltages because of the charge trapped on the line.
- Fault inception and fault clearing. The change from operating voltage to temporary overvoltage on the healthy phases, and the return to operating voltage on the faulted phase, produce phase-to-earth transients.
- Load rejection. It matters in range II, where energization overvoltages are already controlled below 2 p.u.
- Switching of inductive currents and capacitive currents. This includes motors, unloaded transformers, reactors, cables and capacitor banks.
- Distant lightning strokes to a line conductor. Only on lines longer than 100 km. Because lightning currents rarely last more than 200 µs to half value, high amplitudes with critical times to peak do not occur. For that reason they are normally neglected.
How they are represented
The representative voltage has a shape and an amplitude:
- Shape. The standard switching impulse, 250/2500 µs.
- Amplitude. It can be an assumed maximum value or a probability distribution of the amplitudes. Without arrester operation, the distribution is characterized by its 2 % value, its deviation and its truncation value. It may be approximated with a Gaussian distribution between 50 % and the truncation, or with a modified Weibull distribution.
The amplitude is considered independent of the actual time to peak. In some range II systems very long fronts appear, and then the influence of front duration on the dielectric strength of the insulation is taken into account.
The assumed maximum value is the lower of the truncation value of the distribution and the switching protection level of the arrester.
The p.u. values in this guide use the base of the peak of the highest phase-to-earth system voltage, Us·√2/√3.
Fault inception and fault clearing overvoltages
When an earth fault starts or is cleared, transients appear only phase-to-earth. Phase-to-phase overvoltages are neglected. The standard gives these conservative estimates of the assumed maximum, with k the earth fault factor and Us the highest voltage of the system:
| Moment | Assumed maximum peak |
|---|---|
| Fault inception | (2k − 1) · Us · √2/√3 |
| Fault clearing | 2.0 · Us · √2/√3 |
Where they matter:
- Range I. They are considered in networks with isolated or resonant earthed neutral, where k is approximately √3. Here the co-ordination is based on the assumed maximum and the probability of the amplitudes need not be treated.
- Range II. When energization and reclosing are already controlled below 2 p.u., fault inception and clearing overvoltages need careful examination if they are not controlled to the same degree.
The ranges are those of IEC 60071-1: range I from above 1 kV up to and including 245 kV, and range II above 245 kV.
How they are limited
- Pre-insertion resistors in line breakers. This is the most widely used method against line switching overvoltages.
- Controlled closing at the right instant on the voltage wave, and varistors in parallel with the interrupter units. They also serve for inductive and capacitive switching.
- Inductive voltage transformers at the line ends. They reduce the charges trapped after opening, so that a later three-phase reclosing is limited to the level of a simple energization.
- Gapless metal-oxide arresters, and specially designed gapped arresters, for networks with moderate temporary overvoltages.
What the standard says about the arrester:
- It limits the phase-to-earth amplitude to approximately twice the rated voltage of the arrester (peak in kV against r.m.s. value in kV).
- It works against energization, reclosing and switching of inductive and capacitive currents. In general it does not work against fault inception and clearing, because their amplitudes are too low. The exception is faults on series-compensated lines.
- The current that flows during energization or reclosing is below 0.5 to 2 kA. In that range the accuracy of the current matters little, because of the strong non-linearity of the material.
- On a long line with arresters at the ends, the overvoltage at the middle can be considerably higher than at the ends.
- If the phase-to-earth arresters are set so that the overvoltage stays below 70 % of the uncontrolled 2 % value, the phase-to-phase overvoltage can reach close to twice the phase-to-earth protection level.
What this means for the design
The representative slow-front amplitude is the input for setting the switching impulse withstand voltage of each insulation. The standard levels for each piece of equipment are in rated insulation level of switchgear.
To know how much really happens in a specific network, the switching operations must be simulated with a model, as explained in models for EMT studies. The standard states that only a careful simulation, with digital programs or transient analyzers, gives the correct distribution of amplitudes.
When to contact us
If you are going to energize a new line, install a capacitor bank or a reactor, or need to check that your insulation withstands the switching operations in your network, we carry out your insulation co-ordination study and quote the scope, remotely and worldwide. The field part, when needed, is covered by the testing and commissioning service in Colombia, or elsewhere on request.
Related guides
- Insulation co-ordination under IEC 60071: the steps
- Overvoltage when energizing and reclosing a line
- Inductive and capacitive currents: overvoltage on interruption
- Earth fault factor and overvoltage on healthy phases
- Arrester protection level: Upl and Ups
References
- IEC 60071-2 (1996 edition), Insulation co-ordination — Part 2: Application guide, clauses 2.3.3 to 2.3.3.7.
- IEC 60071-1:2006, Insulation co-ordination — Part 1: Definitions, principles and rules, classification of voltage ranges.
Frequently asked questions
What does 250/2500 µs mean?
It is the shape of the standard switching impulse: 250 µs to the peak and 2500 µs to fall to half of the peak value on the tail. It is the shape used to represent slow-front overvoltages in insulation co-ordination.
Do earth-fault overvoltages matter in every network?
No. In range I they matter in networks with isolated or resonant earthed neutral, where the earth fault factor is close to the square root of 3. In range II they need careful examination when energization overvoltages are already controlled below 2 p.u. and fault overvoltages are not controlled to the same degree.
Does a metal-oxide arrester work against slow-front overvoltages?
Yes, against those from energization, reclosing and switching of inductive and capacitive currents. In general no against those from earth fault inception and clearing, because their expected amplitude is too low for the arrester to operate.