Transient overvoltage is an overvoltage with a duration of a few milliseconds or less, oscillatory or non-oscillatory, usually highly damped [source: IEC 60050-614-2016].
Note 1 to entry: Transient overvoltages may be immediately followed by temporary overvoltages. In such cases the two overvoltages are considered as separate events.
IEC 60071-1 defines three types of transient overvoltages, namely slow-front overvoltages, fast-front overvoltages and very fast-front overvoltages according to their time to peak, tail or total duration, and possible superimposed oscillations.
Slow-front overvoltage (SFO) is a transient overvoltage, usually unidirectional, with time to peak 20 μs < Tp ≤ 5 000 μs, and tail duration T2 ≤ 20 ms
Slow-front overvoltages have front durations of some tens to some thousands of microseconds and tail durations in the same order of magnitude, and are oscillatory by nature.
They generally arise from:
- line energization and re-energization;
- faults and fault clearing;
- load rejections;
- switching of capacitive or inductive currents;
- distant lightning strokes to the conductor of overhead lines.
The representative voltage stress is characterized by:
- a representative voltage shape, and
- a representative amplitude which can be either an assumed maximum overvoltage or a probability distribution of the overvoltage amplitudes.
The representative voltage shape is the standard switching impulse (time to peak 250 μs, and time to half-value on the tail 2 500 μs). The representative amplitude is the amplitude of the overvoltage considered independently from its actual time to peak. However, in some systems in range II, overvoltages with very long fronts may occur and the representative amplitude may be derived by taking into account the influence of the front duration upon the dielectric strength of the insulation.
The probability distribution of the overvoltages without surge arrester operation is characterized by its 2 % value, its deviation and its truncation value. Although not perfectly valid, the probability distribution can be approximated by a Gaussian distribution between the 50 % value and the truncation value, above which no values are assumed to exist.
Fast-front overvoltage (FFO) is a transient overvoltage, usually unidirectional, with time to peak 0,1 μs < T1 ≤ 20 μs, and tail duration T2 < 300 μs.
Lightning overvoltages are caused by direct strokes to the phase conductors or by backflashovers, or are induced by lightning strokes to earth close to the line. Induced lightning surges generally cause overvoltages below 400 kV on the overhead line and are, therefore, of importance only for systems in the lower system voltage range. Owing to the high insulation withstand, back-flashovers are less probable in range II than in range I and are rare on systems at 500 kV and above.
The representative shape of the lightning overvoltage is the standard lightning impulse (1,2/50 μs). The representative amplitude is either given as an assumed maximum or by a probability distribution of peak values usually given as the peak value dependent on the overvoltage return rate.
The lightning overvoltages in substations and their rates of occurrence depend on
– the lightning performance of the overhead lines connected to it,– the substation layout, size and in particular the number of lines connected to it, and
– the instantaneous value of the operating voltage (at the moment of the stroke).
The severity of lightning overvoltages for the substation equipment is determined from the combination of these three factors, and several steps are necessary to assure adequate protection. The amplitudes of the overvoltages (without limitation by surge arrester) are usually too high to base insulation co-ordination on these values. In some cases, however, in particular with cable connected substations, the self-protection provided by the low surge impedance of the cables may reduce the amplitudes of the lightning overvoltages to suitably low values.
For the phase-to-phase and the longitudinal insulation, the instantaneous power frequency voltage value on the opposite terminals shall be considered. For the phase-to-phase insulation, it can be assumed that the effects of power-frequency voltage and coupling between the overhead line conductors cancel each other and the opposite terminal can be considered as earthed. For the longitudinal insulation, however, such cancelling effects do not exist and the power-frequency voltage shall be taken into account.
Shielding penetrations occur at a random point on the power-frequency wave. The effect of the power-frequency at the opposite terminal of a longitudinal insulation has to be taken into account by:
- calculating the lightning overvoltage return rates for different instantaneous values of the operating voltage,
- evaluating the insulation failure probability for the various subdivisions into components – usually the sum of the two components is the decisive parameter,
- determining the insulation failure rate dependent on the sum of the lightning overvoltage and of the instantaneous value of power-frequency, and
- applying the performance criterion to this expected failure rate to obtain the necessary sum of the two components.
If this sum is subdivided into a lightning impulse component equal to the representative lightning overvoltage phase-to-earth and a power-frequency component, the power-frequency voltage component will be smaller than the operating voltage phase-to-earth peak. It has been found that a factor of 0,7 may be considered suitable. This means that, for shielding penetration, the longitudinal representative overvoltage should be composed of the representative lightning overvoltage to earth at one terminal and 0,7 times the operating voltage phase-to-earth peak with opposite polarity at the other.
Back flashovers are most likely to occur on the phase which has the highest instantaneous power-frequency voltage value of opposite polarity. This means that, for substations, the representative longitudinal lightning overvoltage shall be equal to the sum of the representative lightning overvoltage to earth at one terminal and of the operating voltage peak at the other (opposite polarity).
Very-fast-front overvoltage (VFFO) is a transient overvoltage, usually unidirectional with time to peak Tf ≤ 0,1 μs, and with or without superimposed oscillations at frequency 30 kHz < f < 100 MHz.
Very-fast-front overvoltages (VFFO) originate from disconnector operations or faults within GIS due to the fast breakdown of the gas gap and the nearly undamped surge propagation within the GIS. Their amplitudes are rapidly dampened on leaving the GIS, for example at a bushing, and their front times are usually increased into the range of those of fast-front overvoltages.
The overvoltage shape is characterized by a fast increase of the voltage nearly to its peak value resulting in a front time below 0,1 μs. This front is usually followed by an oscillation with frequencies above 1 MHz. Dominant frequency components may reach up to some tens of MHz.
For disconnector operations the VFFO amplitude depends on the disconnector construction and on the substation configuration. In the most unfavourable case, the maximum VFFO amplitude can reach 2,8 p.u.
VFFO due to faults within GIS can have amplitudes up to 1,6 times the breakdown voltages.
VFFO in UHV systems may be dangerous to the insulation of GIS, transformers and voltage transformers. The shape and the amplitude of the overvoltage depend on the kind of connection of the equipment to the GIS.
A representative overvoltage cannot be established because suitable standardizations are not available. Experience show that very-fast-front overvoltages have no influence on the selection of rated withstand voltages up to system voltages of 800 kV.
Special care has to be taken for very-fast transients in GIS of UHV systems. Due to the decreasing ratio of lightning impulse withstand voltage to the system voltage, VFFO can become the limiting dielectric stress defining the dimensions of GIS.
Where:
- T2 – time to half-value of a decreasing voltage;
- Tp – time to peak value;
- T1 – front time.
Transient overvoltages, which are also referred to as voltage surges, are brief and sudden deviations or fluctuations that arise within the electrical power network, and may inflict harm on highly-sensitive electrical apparatus. These overvoltages can be instigated by a variety of circumstances, including:
- Lightning strikes. Lightning is considered a principal factor behind the emergence of transient overvoltages within electrical power networks. In situations where lightning strikes near a power line, it can produce a voltage surge that may journey through the line and inflict harm on the apparatus.
- Switching operations. When switching operations occur, such as the activation or deactivation of a circuit breaker or the separation of a power line, an abrupt alteration arises within the electrical current flow, which leads to a transient overvoltage.
- System faults. Electrical faults, such as short circuits or ground faults, may result in overvoltages within the electrical system.
- Electromagnetic interference. Electromagnetic interference (EMI) originating from nearby electrical devices, such as motors or transformers, can result in overvoltages within the electrical system.
- Capacitive coupling. Capacitive coupling arises when two conductors, such as power lines or cables, are in proximity to each other but are not connected. The capacitance between the two conductors may give rise to a voltage surge.
- Power quality issues. Poor power quality, such as voltage dips or sags, may result in overvoltages when the voltage returns to normal.
Protection against transient overvoltages, according to IEC 60364-4-44-2018, is provided by the installation of surge protection devices (SPDs).
