Total 4731 registered members
Current Switching with High Voltage Air Disconnector

Current Switching with High Voltage Air Disconnector

In the paper are presented results of switching overvoltages investigations, produced by operations of air disconnector rated voltage 220 kV. Measurements of these switching overvoltages are performed in the air-insulated substation HPP Grabovica on River Neretva, which is an important object for operation of electric power system of Bosnia and Herzegovina.

Investigations of operating of air disconnector type Centre-Break were performed in order to determine switching overvoltage levels that can lead to relay tripping in HPP Grabovica. During operations of disconnector (synchronization or disconnecting of generator from network) malfunctions of signalling devices and burning of supply units of protection relays were appeared. Also, results of computer simulations using EMTP-ATP [1] are presented.

I. INTRODUCTION

Switching operation in power stations and substations, highvoltage faults and lightning cause high levels of high frequency overvoltages that can be coupled with low voltage secondary circuits and electronic equipment unless they are suitably protected. The function of high-voltage air-break disconnectors is to provide electrical isolation of one part of the switchgear.

Disconnector’s standards define a negligible current interrupting capability (≤0.5 A) or a voltage between the contacts if it is not significantly changed. These values of currents include the capacitive charging currents of bushing, bus bars, connectors, very short lengths of cables and the current of voltage instrument transformers. Disconnector’s contacts in air-insulated substations (AIS) are moving slowly causing numerous strikes and restrikes between contacts.

When the contacts are closed, the capacitive charging current flowing through the contacts ranges from 0.017×10-3 to 1.1×10-3 A/m for voltage levels 72.5 – 500 kV [2], depending on the rated voltage and length of bus, which is switched.

Strikes and restrikes occur as soon as the dielectric strength of the air between contacts is exceeded by overvoltage. The distance between contacts, the contacts geometry and relative atmospheric condition defines the overvoltage at the instant of strike. Every strike causes high-frequency currents tending to equalize potentials at the contacts. When the current is interrupted, the voltages at the source side and the loading side will oscillate independently. The source side will follow the power frequency while the loading side will remain at the trapped voltage. As soon as the voltage between contacts exceeds the dielectric strength of the air, at that distance the restrike will occur, and so on. Successive strikes occurring during the closing and opening operations of the off-loaded bus by the disconnector are shown in Fig. 1 a and b, respectively.

When closing takes place, the first strike will occur at the maximum value of the source voltage. Its values can be positive or negative. As the time passes a series of successive strikes will keep occurring at reduced amplitude, until the contacts touch. The highest transient overvoltage therefore occurs during the initial pre-arc, Fig.1 a. When the disconnector opening, restrikes occur because of the very small initial clearance between the contacts. At the transient beginning, the intervals between particular strikes are on the order of a millisecond, while just before the last strike; the period can reach about one half of cycle at power frequency, Fig. 1 b.

Fig. 1. The voltage due to the disconnector switching a)	Disconnector closing, b)	Disconnector opening 1-source side voltage, 2- load side voltage

Fig. 1. The voltage due to the disconnector switching a) Disconnector closing, b) Disconnector opening 1-source side voltage, 2- load side voltage

During the switching time of operations of disconnectors at HPP Grabovica up to 500 restrikes were registered. In paper [3] there are up to 5000 restrike registered during switching operation of the disconnector. The maximum value of voltages and maximum value of the wave front increasing will take place at the maximum distance between contacts. For the purpose of the investigation of the insulation strength and induction of electromagnetic interferences (EMI), the most important are the first few strikes during the closing operation or the last few strikes during the opening operation. Each individual strike causes a travelling wave with the basic frequency on the order 0.5 MHz (330 kHz-600 kHz). Very fast transient overvoltage due to the closing operation of the disconnector at the load side of the test circuit is shown in Fig.2.

Fig. 2. Very fast transient overvoltage due to the closing operation Channel 1- source side voltage Channel 2-load side voltage

Fig. 2. Very fast transient overvoltage due to the closing operation Channel 1- source side voltage Channel 2-load side voltage

These high-frequency phenomena are coupled with the secondary circuits as a result of various mechanisms. The strongest interference is exerted by the stray capacities between the high-voltage conductors and the grounding system, followed by the metallic link between the grounding system and the secondary circuits.

High-frequency transient current flowing in the grounding system generates potential differences, every time when a strike occurs between disconnector’s contacts. In large secondary circuits, the potential differences are in the form of longitudinal voltages between the equipment inputs and the equipment enclosures.

Depending on the type of secondary circuits used and the way they are laid, differential voltages may also occur. Such a coupling mechanism has a special effect on the secondary circuits of instrument transformers, and particularly on the connected instruments, since these circuits are always galvanically linked to the grounding system. Another factor, which cannot be discounted, is the linking of these circuits to the primary plant via the internal capacities of the instrument transformers [4].

Interference levels in secondary circuits of air-insulated substations during switching disconnectors depend on following parameters:

  • The transient voltages and currents generated by the switching operation;
  • The voltage level of the substation;
  • The relative position of the source of disturbances and susceptor;
  • The nature of the grounding network;
  • The cable type (shielded or unshielded);
  • The way the shields are grounded.

There are two main modes of coupling secondary circuits with primary circuits [3, 5]:

  1. Electromagnetic or EM coupling, which can be split into three sub-categories; inductive, capacitive and radiative. The most important source of EM coupling is the propagating current and voltage waves on bus bars and power lines during high-voltage switching operations by disconnectors;
  2. Common impedance coupling, as a result of coupling caused by the sharing of a lumped impedance common to both the source and susceptor circuits.

Common mode voltages, i.e., voltages measured between conductors and local ground, represent the main parameter used for assessing equipment immunity. The difficulty of comparing data comes from the fact that different authors performed measurements at different places (some measurements were made at the closest point to the disconnector being operated whereas others made measurements in the vicinity of the auxiliary equipment, i.e. in the relay room). Little information is available about the grounding practice of the neutral conductor in CT or VT circuits, the quality and grounding of the sable shields as well as how the measurements have been performed. Therefore, the measured levels have to be analyzed very carefully before comparison and drawing any conclusions [5]. Results of up to date measured common mode voltages at secondary circuits of CVT, CT and VT are presented in the paper [5]. There are maximum levels of the common mode voltages ranging from 100 Vpeak up to 2.5 kVpeak in the shields of the secondary circuits cables of the CT and VT. Results show that measured values of the common mode voltages at CT/CV secondary circuits, 220 kV ratings, range from Ucm=0.32 kVpeak [6] up to Ucm=0.85 kVpeak [7].

Results shown in paper [3] are for measured common mode voltages from 3-4 kV during switching operation by disconnector in 150 kV switchgear up to 6-10 kV at 400 kV switchgear.

II. RESULTS OF EXPERIMENTAL MEASUREMENTS ON SITE

The last ten years of extensive analysis of disconnector and circuit breakers generated EMI measurements that have confirmed that disconnector operation with off-loaded busbar is the most important and typical source of interference in secondary circuits of substations. Measurements of switching overvoltages generated during disconnector operation in the air insulated substation HPP Grabovica on the river Neretva were performed. HPP Grabovica is an important object for operating of electric power system of Bosnia and Herzegovina. Investigations of operating of air disconnector type Centre-Break were performed in order to determine switching overvoltage levels that can lead to relay tripping in HPP Grabovica [8].

During operations of disconnector (synchronization or disconnecting of generator from network) malfunctions of signalling devices and burning of supply units of protection relays were appeared. Malfunctioning of auxiliary circuits were manifested by tripping relay of differential protection of the generator, phase ’4′- signalization on relay box ‘ZB I‘ and signalling ‘fire’ in 35 kV control panel.

At the same time sparking between primary terminals of the current transformer (CT) was occurred. Malfunctioning of
signalling circuits were lower (not eliminated) with installing shielded cables. Also, independent of switching operation of air insulated disconnectors, during synchronization of generator AG1 on network, it’s happened that one of the pole of 220 kV circuit breaker failures. In this case generator AG1 worked in motor regime. Because of that, HPP Grabovica plans to install circuit breakers on generator’s voltage (10,5 kV) [9].

The field tests were performed at the test circuit at HPP Grabovica, Fig. 3.

Fig. 3. The considered test circuit VT-voltage transformer (220/√3/0.1/√3/0.1/√3 kV), CT-current transformer (200/1/1 A), CVD-capacitive voltage divider, CB-circuit breaker with two interrupting chambers and parallel capacitors (SF6 220 kV, 1600 A), Dc- disconnector (220 kV, 1250 A), MOSA-metal oxide surge arrester (Ur=199,5 kV, 10 kA), PT-power transformer (64 MVA, 242/10,5±5% kV, YD5), AG1- generator 1 (64 MVA, 10,5±5% kV)

Fig. 3. The considered test circuit VT-voltage transformer (220/√3/0.1/√3/0.1/√3 kV), CT-current transformer (200/1/1 A), CVD-capacitive voltage divider, CB-circuit breaker with two interrupting chambers and parallel capacitors (SF6 220 kV, 1600 A), Dc- disconnector (220 kV, 1250 A), MOSA-metal oxide surge arrester (Ur=199,5 kV, 10 kA), PT-power transformer (64 MVA, 242/10,5±5% kV, YD5), AG1- generator 1 (64 MVA, 10,5±5% kV)

The recorded wave shape of the overvoltage at the load side is shown in Fig. 4. The overvoltage factors at busbar, k, were recorded up to 1.16 p.u. with the dominant frequency of considered transient fd equal to 0.536 MHz. Common mode voltages, Ucm, at VT were up to 708 Vpeak, with dominant frequency equal to 1.31 MHz.

Fig. 4. Waveshape of the overvoltage Channel 1-voltage at CVD; ch 1 (2.5 V/div), probe 1x100, ratio 455 Channel 2-voltages at secondary of VT; ch 2 (5 V/div), probe 1x100

Fig. 4. Waveshape of the overvoltage Channel 1-voltage at CVD; ch 1 (2.5 V/div), probe 1x100, ratio 455 Channel 2-voltages at secondary of VT; ch 2 (5 V/div), probe 1x100

III. MODELING OF THE TEST CIRCUIT

Computer simulations were performed on the model of test circuit containing elements drawn in Fig. 5. Overvoltages at busbars were calculated during disconnector closing operations, for the same substation layout on which measurements were carried out.

Fig. 5. Model of the test circuit Arc-4 Ω; stray-200 pF; connection tube Z=370 Ω; CVD-R=300 Ω, C=1 nF; VT-500 pF; CB-2 capacitors, each C≅2 nF, (capacitance of open contacts, each C≅20 pF), Ccb=100 pF; CT-500 pF; MOSA-100 pF; connection wire Z=440 Ω; PT-3.5 nF

Fig. 5. Model of the test circuit Arc-4 Ω; stray-200 pF; connection tube Z=370 Ω; CVD-R=300 Ω, C=1 nF; VT-500 pF; CB-2 capacitors, each C≅2 nF, (capacitance of open contacts, each C≅20 pF), Ccb=100 pF; CT-500 pF; MOSA-100 pF; connection wire Z=440 Ω; PT-3.5 nF

The waveshape of simulated overvoltage surge at load side is given in Fig. 6. The difference between magnitudes of measured and simulated overvoltages is 5 %. The dominant frequency of simulated overvoltage is 0.620 MHz. Comparison between results of measured and calculated overvoltages certified a good agreement of obtained values.

Fig. 6. Waveshape of simulated overvoltage surge

Fig. 6. Waveshape of simulated overvoltage surge

When the Capacitive Voltage Divider (CVD) was excluded, there were higher values of calculated overvoltages (15% higher on amplitude and 6 % on frequency). Capacitive divider due to primary resistor equal to 300 W and primary capacitance equal to 1 nF influences on overvoltage at the same measurement point causing attenuation and damping of transient overvoltrages. In order to reduce EMI in secondary circuits the best way is to reduce sources of interference emission during switching of air insulated disconnector.

One of the ways of reducing is to install disconnecting circuit breakers. Substation disconnectors isolate circuit breakers from rest of the system during maintenance and repair. The maintenance requirements for modern SF6 high voltage circuit breakers are lower than maintenance demands made on disconnectors, which means one of reasons for disconnectors removed. Installing disconnecting circuit breaker there are no needs for switching operation of disconnectors. With disconnecting circuit breakers it is still possible to isolate the line, but low maintenance requirements means it is no longer necessary to isolate the circuit breaker. The disconnecting breaker had to be designed to safety lock in the open position, and to meet all voltage withstanding capabilities and safety requirements of disconnectors.

Another way of reducing sources of interference emission is to install circuit breaker without parallel capacitors to contacts. This suggestion is based on analyses performed on three circuit models:

  1. Model of CB with two breaking chambers and paralel capacitors and VT on netvork side of CB;
  2. Model of CB with two breaking chambers and without paralel capacitors and VT on netvork side of CB
  3. Model of CB with two breaking chambers and without paralel capacitors and VT on generator side of CB

Magnitudes of simulated overvoltages are presented in Table I. Voltages are measured in point of connection of VT, CT and PT.

TABLE I - MAGNITUDES OF SIMULATED OVERVOLTAGES

TABLE I - MAGNITUDES OF SIMULATED OVERVOLTAGES

Overvoltages on generator side of 220 kV CB during switching of disconnectors could be up to 320 V in the case of installing instrument voltage transformer (VT) on generator side of CB without parallel capacitors (near instrument current transformer CT). This case causes installing of circuit breaker at generator’s voltage (10,5 kV) for synchronization of generator to network (better conditions for synchronization). This solution of installing circuit breakers on generator’s voltage resulted from problems have occurred during synchronization of generatror with current 220 kV CB.

IV. CONCLUSION

Switching overvoltages due to disconnector operations have been analysed on the existing 220 kV AIS on HPP Grabovica. Measurements and calculations were conducted on the characteristic points in AIS, in order to determine the level of the EMI.

The result of measurements has shown that high frequency voltages on busbars occur with amplitudes up to 1.16 p.u. (233 kVpeak) and the dominant frequencies up to 0.6 MHz. The difference between magnitudes of measured and calculated overvoltages is 5 % and 15.6 % on frequency. Measured common mode voltages at secondary circuits were from 430 V up to 708 V. CVD influences on overvoltages at the same measurement point on busbars causing attenuation and damping of transient overvoltages.

Comparison of the transient computer simulations with field measurements showed that calculations could be used for
assessment of the transient overvoltages due to disconnector switching. In order to reduce EMI in secondary circuits, it is suggested to install switching modules and disconnecting circuit breakers [10] or to install circuit breakers without parallel capacitors to contacts.

AUTHORS: Salih Carsimamovic, Zijad Bajramovic, Miroslav Ljevak, Meludin Veledar, Nijaz Halilhodzic

.

Related articles

Figure 1: Three generations of transformer substations

Figure 1: Three generations of transformer substations

Early transformers were located at the top of pylons and could achieve powers of up to 1000kVA. Column-type transformer substations provided the interface between overhead and underground networks. These were equipped essentially with air-insulated MV switchgear, a liquid- insulated transformer and a low voltage distribution switchboard.

These were fabricated from bricks, and thanks to the chimney effect provided by the column the substations had good airflow, and consequently there were no problems with overheating of the equipment.

However the next generation of substations for underground networks also built from brick, had a reduced height and no chimney effect, and for the first time the equipment designers had to confront overheating problems.

This second generation of transformer substations was also the subject of the first internal fault tests, intended to provide operating personnel and the general public with greater levels of safety. The next step was the introduction of factory assembled prefabricated transformer substations.

Figure 2: 630kVA qualified by EDF, validated against internal faults in the air and easily integrated into buildings.

Figure 2: 630kVA qualified by EDF, validated against internal faults in the air and easily integrated into buildings.

This third generation of transformer substations were subject to the first international regulations. These substations were characterised by the use of more compact, environmentally insensitive equipment, factory assembled and standardised which means that Utilities can be supplied with series produced products, the production and performance of which are guaranteed by the tests carried out by the manufacturers in accredited laboratories.

For the third generation substations the first Utility specifications demanded small surface areas, leading to a standardised layout of the substations (Figure 1).

This enclosure can be metallic, GRC (Glass Reinforced Concrete), GRP (Glass Reinforced Plastic), SFRC (Steel Fibres Reinforced Concrete). The world wide trend is for reinforced concrete enclosures for the following reasons :

  • Improved mechanical strength
  • Reduced effect of solar radiation
  • Reduced condensation
  • Improved fire behaviour
  • Weathering
  • Improved aesthetic.

Finally, a fourth generation of transformer substations (Figure 2) has recently appeared, where the detailed technical specification has been replaced by functional specification.

AUTHORS: Thierry CORMENIER, ALSTOM-France; Robert DIDES.

.

Related articles