WO1997048159A1

WO1997048159A1 - Method of protecting a system or an installation from surge and voltage clamping circuit

Info

Publication number: WO1997048159A1
Application number: PCT/CH1997/000165
Authority: WO
Inventors: Felix Greuter
Original assignee: Abb Research Ltd.
Priority date: 1996-06-13
Filing date: 1997-04-24
Publication date: 1997-12-18
Also published as: DE19623541A1; AU2502297A

Abstract

A first MOV surge arrester (4) e.g. for a high-voltage system, has a conventional voltage limit, at which it becomes conductive, that lies just above the phase voltage of the system and forms a first current path with a series-connected impedance (6), the value of which (Z) in an initial state is negligible. If the first MOV surge arrester (4) is threatened with overload owing to a prolonged surge and consequent excessive energy absorption by the surge arrester, a further current path which runs through the first MOV surge arrester (4) and a second MOV surge arrester (5) parallel to the impedance (6) and which is connected in series with it will be switched over by raising the impedance (6) value. In this way leakage current through the first MOV surge arrester (4) and thus the energy absorption of the arrester are reduced to normal values. The circuit enables substantial reduction in the level of protection and system insulation may therefore be designed to substantially lower values. The impedance (6) may be formed as a PTC element, a safety fuse or a switch controlled by the energy absorption or temperature of the first MOV surge arrester (4), possibly with a parallel high pass filter and a parallel spark gap.

Description

METHOD FOR PROTECTING A NETWORK OR SYSTEM FROM OVERVOLTAGE AND VOLTAGE LIMITING CIRCUIT

The invention relates to a method for protecting a network or a system against overvoltages and a voltage limiting circuit.

From EP-A-0 651 491 a method according to the preamble of claim 1 is known, in which the switchover between a first current path containing only a varistor in addition to a switch by opening the

Switch to another current path, which contains a gas discharge arrester in series with the varistor, is carried out when the resistance of the varistor has irreversibly collapsed due to overloading and it has therefore failed. The process, which is apparently intended for the low-voltage range, is not suitable for preventing the failure of the varistor, but is only used to at least partially absorb the consequences of such a failure.

It has also been known for a long time to protect networks of all voltage ranges from overvoltages by means of MOV surge arresters composed of varistors, in particular MOV varistors. Most of these are simple series connections or series-parallel connections of MOV varistors. Such MOV surge arresters have a strongly non-linear characteristic and become abruptly conductive when a fixed limit voltage is reached, while they isolate below the limit voltage, so that the voltage between the two connections between them the overvoltage conductor is limited to the limit voltage, while practically no leakage currents occur below it. For general information on MOV surge arresters, see For example, see the article 'MOV surge arresters without spark gaps enable optimal surge protection' from Brown Boveri Technique No. 12 (1985).

While such MOV surge arresters are excellent for the derivation of short-term, even high overvoltages, such as those caused by switching operations and

Lightning strikes are caused or are suitable for fast transient overvoltages, their interpretation with regard to longer lasting overvoltages is not without problems. Even relatively low temporary overvoltages, none for the isolation of the network or the system itself

Pose a danger, can cause larger leakage currents, so that the MOV surge arrester absorbs too much energy and is thermally overloaded, especially if the temporary surge voltage is prolonged, which can lead to destruction of the MOV surge arrester.

For this reason, it was previously necessary to work with a considerable safety margin when protecting networks and systems and to design the MOV surge arresters so that the limit voltage is above the expected temporary surge voltages and the maximum continuous operating voltage is correspondingly high. However, this has the consequence that the network or the system must be isolated to correspondingly high voltages, which increases the costs significantly. In itself, a protection would be that the normal mains voltage as the maximum continuous Voltage is based on the cheapest with a corresponding limit voltage just above the same, but this was not feasible due to the risk of overloading the MOV surge arrester.

To give a numerical example:

In the case of a network with a network voltage U _m of 144 kV, ie a voltage between one phase and earth of U _m / V3 = 83 kV, temporary overvoltages of

to be calculated, where c _e <1.4 is the earth fault factor and c _x «1.2-1.5 is the load shedding factor. This leads to U _τ = 157 kV if c _λ = 1.35 is assumed.

If one assumes that the MOV surge arrester can withstand an overvoltage of a maximum of TU _c with T = 1.45, for a few seconds from experience, whereby T is a known factor taking into account the duration of the load and U _{c is} the maximum continuous operating voltage, so U _c > U _τ / T must be chosen, since overvoltages up to a voltage level of U _τ and a duration of a few seconds can be expected.

In the context of the numerical example above, this leads to U _c ~ 112 kV, ie the MOV surge arrester must be designed for a maximum continuous operating voltage which is 112 kV, i.e. approx. 35% above the phase voltage of 83 kV. With the usual protection factors of approx. 3 means this means that the insulation of the network must be designed for 336 kV.

In principle, it would be desirable to protect the network in such a way that the network voltage cannot be exceeded, or only barely. However, this was previously not possible due to the risk of overloading and destruction of the MOV surge arrester.

The invention is intended to remedy this. The invention is based on the object of specifying a method for protecting networks and systems against overvoltages, which makes it possible to keep the limit voltage effective in normal operation low. A suitable voltage limiting circuit should also be specified.

The invention makes it possible to move the maximum continuous operating voltage close to the mains voltage. As a result, lower voltage values can be assumed when designing the insulation of the network in question or the system, which makes the network or the system significantly cheaper.

In the numerical example above, U _{c is} reduced to 83 kV in normal operation, so that the insulation of the network can be designed for 249 kV instead of 336 kV.

By means of the voltage limiting circuit according to the invention, the voltage is increased by a MOV surge arrester both in normal operation when the latter assumes a basic state and when temporary overvoltages occur when it has switched to a protective state limited. The advantages associated with these, such as low leakage current and very quick response, are maintained in both states.

The invention is explained in more detail below with reference to drawings, which only represent exemplary embodiments. Show it

1 shows the general circuit diagram of a voltage limiting circuit according to the invention,

Fig. 2a-c examples of the impedance in Fig. 1 and

Fig. 3a, b examples of the switch in Fig. 2c.

In its basic structure, the one between z. B. a high-voltage network connected first terminal 1 and a grounded second terminal 2 lying voltage limiting circuit 3 of Fig. 1, a first MOV surge arrester 4 and a second MOV surge arrester 5 in series. Each of the two MOV surge arresters 4, 5 is conventionally z. B. constructed as a series parallel connection of MOV varistors, as can be found in the article mentioned above. They can also contain other elements that do not impair their function, such as parallel spark gaps or resistors in series.

Also in series with the first MOV surge arrester 4, but parallel to the second MOV surge arrester 5 is an impedance 6, the properties of which are represented by the value Z. The switching voltage at which the first MOV surge arrester 4 becomes conductive corresponds to a first limit voltage, which, for. B. is at the normal mains voltage. In the numerical example mentioned at the beginning, this would correspond to 83 kV. The second MOV surge arrester 5 has a switching voltage which, together with the first limit voltage, gives a further limit voltage which corresponds to that determined according to known principles, i.e. 112 kV in the numerical example, i.e. the switching voltage of the second MOV surge arrester 5 is 29 kV corresponding to the Difference between these two values.

The impedance 6 can be implemented in a wide variety of ways. However, it will always contain a switch-like element, which in the event of impending overload of the first MOV surge arrester 4 due to longer-term temporary overvoltages and excessive energy consumption caused thereby, at least below a cut-off frequency, which in an AC network is not lower than the network frequency of z. B. may be 50 Hz, at least up to a blocking voltage, which is higher than the above-mentioned further limit voltage, so that on the further current path, which by another MOV overvoltage arrester - in the shown embodiment of the voltage limiting circuit of the series connection of the first MOV surge arrester 4 and the second MOV surge arrester 5 is formed - is switched over.

The impedance 6 can e.g. B. be implemented as a purely passive element. In the embodiment according to FIG. 2a, this is a PTC element 7, which is a good conductor at low currents and whose resistance when a certain one is reached Limit current and corresponding temperature increases suddenly, so that it almost completely blocks. The PTC element 7 is matched to the first MOV overvoltage arrester 4 in such a way that an excessive leakage current flowing through it and the PTC element 7 causes it to switch, ie causes its resistance to rise suddenly before it leads to an overload of the first MOV surge arrester 4. Another purely passive element that can be used instead of the PTC element is a fuse. However, this would have to be replaced after each switching.

2b, a separator 8 is arranged in series with the PTC element 7, which, for. B. controlled by the leakage current through the PTC element 7, is opened after the same has switched and the leakage current has dropped to very low values. Providing a disconnector connected in series, which is opened after the impedance 6 has been blocked, is possible with each configuration thereof.

2c, the impedance 6 is a parallel connection of a switch 9, a high pass 10 and one

Spark gap 11 formed. The high pass 10 is designed in such a way that it is practically transparent to frequencies above a certain cut-off frequency, which may not be less than the mains frequency. It is used to derive rapidly increasing overvoltages, such as those caused by lightning or switching operations, and from rapid transient overvoltages. It can be made up of purely passive elements - capacitors, resistors, linear or non-linear inductors - in the simplest Case it is a capacitor. If necessary, the spark gap 11 becomes active when temporary overvoltages are superimposed with overvoltages caused by lightning or switching operations.

The switch 9 can be designed as a mechanical switch or also as a semiconductor switch or as a series connection or series parallel connection of such. So he's z. B. as a parallel connection of two series connections of opposite polarity, each of which (FIG. 3a) comprise three jointly controlled GTOs 12a, b, c, or as a corresponding parallel connection of IGBTs 13 (FIG. 3b). In addition, the use of photoconductive switches, MCTs, BODs etc. is also possible. Another option is a piezoresistor.

In any case, the switch 9 generally forms an active element - even if the use of a passive element such as e.g. B. a PTC element in its place in Fig. 2c is quite possible - the opening of which must be triggered from the outside. Various parameters can be used for tripping, provided that they reflect the load on the first MOV surge arrester 4.

In order to do this more precisely: In the long term, of course, there must be a balance between the absorption of electrical energy by the first MOV surge arrester 4 and its conversion into heat on the one hand and the dissipation of heat by heat conduction, radiation, convection at an operating temperature at which the first MOV transmitter ¬ voltage arrester 4 also functions safely in the long term. This balance may be disturbed in the short term are so that the MOV varistors in the first MOV surge arrester 4 heat up above the normal operating temperature. However, such faults must not be too severe and long-lasting, since otherwise the temperature will reach values that result in a change in the material properties of the MOV varistors, which affects their electrical properties. This can lead to a positive feedback between an increase in the leakage current and the temperature, which results in a total failure of the first MOV surge arrester 4 with a short circuit.

The decisive factor is the temperature of the varistor, which can be measured directly or indirectly, but often only with some effort. For this reason, it is usually preferable to measure a quantity that represents a measure of the energy consumption of the first MOV surge arrester 4, for. B. the leakage current through the first current path or one of the same dependent parameter as the magnetic field, the z. B. can be determined by a Hall probe. It is also possible to measure the voltage instead of the current, e.g. B. by voltage tap on one of the varistors. The energy consumption can be calculated from current or voltage using the known characteristic.

In addition, a direct or indirect measurement of the temperature is of course not excluded, e.g. B. directly through a T-element, an IR sensor, an element

Bimetal or a memory alloy or indirectly via a membrane switch which responds to the expansion of gas heated by the first MOV surge arrester 4. The basic idea behind all designs is the switchover from a first current path with a first limit voltage to a further current path with a higher further limit voltage in the event of impending overload of the first MOV surge arrester 4. This is done by changing the impedance value of the first current path in such a way that it reaches to a blocking voltage which is higher than the further limit voltage, in any case up to a limit frequency which may be above the mains frequency. As a result, the excessive leakage current, which has led to excessive energy consumption by the first MOV surge arrester 4, is practically completely prevented. The additional current path, which also contains a MOV surge arrester or a series connection of such surge arresters, with a total of a higher limit voltage than the first MOV surge arrester, still ensures that short-term overvoltages are derived.

Even if the further current path, as shown in the exemplary embodiments, also passes through the first MOV surge arrester 4, the energy consumption of the latter is throttled to very low values so that it can cool down again. It can also be completely relieved by the additional current path completely avoiding the first MOV surge arrester 4. However, this is usually not economically viable and technically not necessary.

The principle described can also be staggered, i. h In addition to the second current path, a third current path with an increased limit voltage can be provided, etc. The voltage limiting circuit 3 can usually be designed such that the voltage in most operating states is largely absorbed by the first MOV surge arrester 4. The other elements, the second MOV overvoltage arrester 5 and the impedance 6 are only exposed to relatively low voltages and can therefore be designed as medium voltage components in the above numerical example, which significantly reduces their price and thus that of the entire voltage limiting circuit.

Claims

claims

1. A method for protecting a network or a system against an overvoltage, whereby to limit the voltage between a first connection (1) and a second connection (2) of a first connecting the first connection (1) to the second connection (2) Current path with a first MOV surge arrester (4), which becomes conductive at a first limit voltage, on a further current path which likewise connects the first connection (1) to the second connection (2) and becomes conductive at a further limit voltage, depending on at least one parameter dependent on the state of the first MOV surge arrester (4) is switched from the first current path to the further current path, characterized in that the switchover to the further current path, the further limit voltage of which is higher than the first limit voltage, takes place precisely when the value of the at least one parameter is an impending overload of the first MOV surge arrester (4) d by a long-term overvoltage.

2. The method according to claim 1, characterized in that the network or the system is a medium or high voltage network or a medium or high voltage system.

3. Voltage limiting circuit (3) for protecting a network or a system, with, between a first connection and a second connection, a first current path, which has a first MOV overvoltage contains arrester (4), which in a basic state of the voltage limiting circuit (3) becomes conductive at a first limit voltage, characterized in that it likewise has a further current path between the first connection (1) and the second connection (2), which contains a further MOV overvoltage arrester and which becomes conductive at a higher additional limit voltage and that in the event of impending overload of the first MOV overvoltage arrester (4), the voltage limiting circuit switches over to a protective state in which the first current path is at least below a limit frequency, which may be above the mains frequency, blocks up to a blocking voltage above the further limit voltage.

4. Voltage limiting circuit (3) according to claim 3, characterized in that the first current path in series with the first MOV surge arrester (4) contains a variable impedance (6), the value of which is low in the basic state and in the protective state at least for below the limit frequency lying frequencies up to at least the blocking voltage is high.

5. voltage limiting circuit (3) according to claim 4, characterized in that the variable impedance (6) one of the following components or a

Parallel connection of several such components includes: switching device (9), fuse, PTC resistor (7), piezoresistor, high pass (10), spark gap (11). 6. voltage limiting circuit (3) according to claim 5, characterized in that the switching device (9) at least one mechanical switch or disconnector or a semiconductor switch such. B. GTO (12a, b, c), IGBT (13) or photoconductive switch.

₇ . Voltage limiting circuit (3) according to one of _A nsprüche 4 to 6, characterized in that the further current path also, and a direction parallel to the variable impedance (6) second MOV surge arresters includes the first MOV Ueber¬ arresters (4) (5).

₈ . _S pannungsbegrenzungsschaltung (3) according to one of claims 3 to 7, characterized in that possibly impending overload of the first MOV surge arrester (4) by means of electrical monitoring of the recording power is determined by the same.

9. _S voltage limiting circuit (3) according to claim 8, characterized in that the absorption of electrical energy by the first MOV surge arrester (4) is determined from a measured variable associated with the same, such as current, magnetic field, voltage.

_10th _S pannungsbegrenzungsschaltung (3) according to one of _A nsprüche 3 to 7, characterized in that possibly impending overload of the first MOV Ueber¬ arrester is determined by means of monitoring the temperature of the same (4).