Cb Chapter 3

8/14/2019 Cb Chapter 3

1/19

ABB Switchgear AB: 2000 3-1

1999-01-21

3 Circuit-breaker design

3.1 Interrupting modes

When a circuit-breaker is tripped to interrupt a short-circuit current, the contact parting can takeplace anywhere in the current loop. The current will then flow between the contacts through an arc,which consists of a core of extremely hot gas with a temperature of 10 000 to 20 000 K. Thiscolumn of gas is fully ionized (plasma) and has an electrical conductivity comparable to that ofcarbon.

To interrupt the current, the circuit-breaker has to wait for a current zero. When the currentapproaches zero the arc diameter will decrease. The cross section is approximately proportional tothe current. When the current reaches zero, the arc has decreased to a tiny filament or thread ofionized gas. The natural current zero is used to attack the arc region by rapidly taking away the

ionized particles and cooling the arc region.

The current interruption process is a complex matter due to simultaneous interaction of severalphenomena. Based on the nature of the dominant phenomena, two distinct regimes can beidentified in the interruption process: the thermal and the dielectric regime. These regimes areslightly separated in time (see Figure 3-1). The choice of an extinguishing medium and the circuit-breaker characteristics are dependent on the behavior of this medium in the two regimes.

Figure 3-1 - Stresses on a circuit-breaker at interruption

3.1.1 Thermal regime

At current zero the hot arc channel between the circuit-breaker contacts has to be rapidly cooleddown to such a low temperature that it is no longer electrically conducting. The difficulty to interruptis related to the rate of decrease of the current towards zero, d i/dt and to the rate of rise of the

recovery voltage after current zero, du/d t.

There exists a certain inertia between the current and electrical conductivity of the arc (see Figure3-2). When the current approaches zero, there is still a certain amount of electrical conductivity leftin the arc path. This gives rise to what is called a "post-arc current". The fact, whether or not theinterruption is going to be successful is determined by a race between the cooling and the energyinput in the arc path by the transient recovery voltage. When the scales of the energy balance tipover in favor of the energy input, the breaker fails thermally. Thermal failures typically occur lessthan 20 s following current zero.


2/19


1999-01-21

Figure 3-2 - Stresses on a circuit-breaker at interruption (times are in the microsecond range)

The decisive regime for current interruption is the thermal region, where the interaction between thearc and the network is of paramount importance.

The thermal regime is especially critical for short-line fault interruption (see 2.3). The circuitparameters directly affecting this regime are the slope of the current to be interrupted (di/dt) and theinitial rate of rise of the transient recovery voltage immediately after current zero (d u/dt). The higher

the values of any of these two parameters, the more severe the interruption.

The thermal interruption regime for SF6 circuit-breakers corresponds to the period of time starting

some s before current zero up to the vanishing post-arc current (2 - 4 s after current zero).

3.1.2 Dielectric regime

After having passed the thermal region, the voltage rises relatively fast at the circuit-breakerterminals and the success of the interruption is now depending on the voltage withstand capabilityof the contact gap.

In the dielectric regime, located some hundreds of microseconds after current zero, the circuit-breaker has to withstand the full peak value of the TRV without dielectric breakdown.

In the dielectric regime the extinguishing/isolating medium is not electrically conducting anymore,but it still has a much higher temperature than the ambient. This reduces the voltage withstand

capacity of the contact gap.

The cases with short-circuit currents close to the circuit-breaker (terminal faults) as dealt with insection 2.2 mainly give TRV's with relatively low rates of rise after current zero. The predominantstress on the circuit-breaker will thus be caused by the TRV peak (a stress in the dielectric region).

An SF6 breaker decides whether it is going to interrupt within the first few s after current zero.

Even if the thermal interrupting capability is sufficient there is no guarantee that the interruption willbe successful. At current zero the circuit-breaker attempts to interrupt. When this attempt issuccessful, (no thermal failure), the transient recovery voltage across the contacts rises rapidly andto very high values when the breaking unit is well utilized. For example in a single unit 245 kVcircuit-breaker the contact gap may be stressed by 400 kV or more 70 to 200 s after the current

zero.


3/19


1999-01-21

Figure 3-3 - Dielectric interruption mode

The recovery of the voltage withstand of the contact gap must always be higher than the recoveryvoltage, see Figure 3-3, otherwise a dielectric reignition will occur (dielectric failure). This requires

an extremely good dielectric withstand of the gas, that is still rather hot (and thin)

3.2 Interruption of short-circuit currents

The interruption of short-circuit currents is usually the most critical duty and therefore determinesthe design of the circuit-breaker.

As described in 3.1, the plasma is fully ionized and has an electrical conductivity comparable to thatof graphite. This is illustrated in Figure 3-1.

Figure 3-4 - Electrical conductivity of most gases as a function of temperature


4/19


1999-01-21

As long as the current is high, the arc will have negligible influence on the current flow. An arcvoltage of only some hundred volts may exist. Blowing onto the arc to cool it at this stage, will only

result in a slight increase in arc voltage to compensate for the increased losses by the cooling.

As described in 3.1, the circuit-breaker has to wait for a current zero to interrupt. At the very

moment of current zero no energy is fed into the arc. Would the arc react instantaneously to thecurrent, it would disappear and the current would then be interrupted. However, the arc has a timeconstant, that depends on how much heat is stored in the arc and how quickly this heat can beremoved by cooling. This means that at current zero the arc still has a certain conductance and thatthe network can inject more energy into the arc after current zero (see also 3.1.1). A successfulinterruption requires that the cooling is so efficient relative to the energy input after current zerothat the temperature drops quickly.

Figure 3-4 shows that cooling a gas from 5 000 K to 1 500 K decreases the conductivity by 12orders of magnitude. The gas is transformed from a conductor comparable to graphite to an

insulator comparable to porcelain.

3.3 Characteristics of extinguishing media

The arc can be cooled by various processes, e.g. radiation, thermal conduction, convection andturbulent mixing of cold gas into the arc. Radiation is dominant at high currents and temperatures,whereas turbulent mixing is generally considered the most important cooling mechanism at currentzero. Different extinguishing media have different properties in this respect. As an example Figure3-4 shows the thermal conductivity of the most interesting media for circuit-breakers:

- N2, nitrogen, which is the main constituent of the gas in air-blast circuit-breakers;

- H2, hydrogen, dominating in the gas bubble in which the arc is burning in oil circuit-breakers;

- SF6, sulfur-hexafluoride, in SF6 circuit-breakers.

(vacuum circuit-breakers, where the extinction mechanism is different from those given above, arenot considered in this document)

The three types of circuit-breakers stated above do not cover the whole range of circuit-breakers.There are circuit-breakers that use different extinguishing principles:

- air magnetic circuit-breakers, where the arc is stretched by the magnetic forces. Cooling of thearc is by the surrounding air;

- vacuum c ircuit-breaker;

- fuses may be considered as well, since their task is to interrupt short-circuit currents.


5/19


1999-01-21

Figure 3-5 - Thermal conductivity of H2, N2 and SF6 as function of temperature

It would seem favorable to use a gas that has a high thermal conductivity in the temperature rangewhere the transition from conductor to insulator takes place.

Nitrogen has a high thermal conductivity but at too high a temperature to be effective. This wouldexplain the inferior capability of air-blast circuit-breakers to cope with short-line-faults.

Hydrogen has a high thermal conductivity in the region of interest. This would explain the very goodshort-line fault capability of oil circuit-breakers.

SF6 has a couple of peaks in the conductivity curve in the temperature interval where the gas is

transformed to an insulator. This would explain the good short-line performance.

However, a closer analysis indicates that thermal conduction is insufficient for the extinction of thearc. In this case, the specific heat vs. temperature is important and it is indirectly related to the

thermal conductivity (Figure 3-5) in the following way:

The peaks in Figure 3-5 appear at temperatures where the dissociation of the gas moleculesconsumes a lot of energy, which means that the specific heat should be high too. It is an advantageif this happens already at a low temperature, since this promotes a low final temperature and a lowelectrical conductivity (Figure 3-4).


6/19


1999-01-21

A simplified explanation of the good extinguishing properties of SF 6 is given below.

Figure 3-6 - The SF6 molecule

The SF6 molecule consists of six Fluorine (F) atoms bound to a central Sulfur (S) atom. The SF 6molecule has the shape of an octahedron, with the F atoms located at the corners (see Figure 3-6).These 6 bindings are equal, which explains the high stability of this gas. Below 800 K, the SF 6molecule is stable. Above this temperature the SF6 molecule starts to dissociate, i.e. F atoms areknocked away leaving smaller and smaller molecular fragments the higher the temperature.

This process is reversible, i.e. the atoms recombine to form SF6 molecules when the temperature isdecreased. This dissociation requires energy and this energy is returned when the atoms recombine

when the temperature is lower.

Consider now a hot arc column along which a cold SF6 stream is blown (Figure 3-7). Some

turbulence in the boundary between the hot and cold gas may cause some cold SF 6 molecules tomix with the hot gas. If the temperature in the arc column is high, the cold molecules will be

dissociated. This means consumption of a lot of energy and hence effective cooling of the hot gas.

Figure 3-7 - Circuit-breaker arc in axial blast nozzle

This process is very efficient. A numerical example may illustrate this: When mixing 1.0 liter of hotSF6 gas with a temperature of 10 000 K with 0,01 liter of cold SF6 gas at a temperature of 1 000 Kat the same pressure, the temperature of the mixture will be 3 000 K. See Figure 3-8. (For an idealgas the resulting temperature would have been > 9 000 K)


7/19


1999-01-21

Figure 3-8 - Cooling effect of SF6 gas

This model explains why the dissociation temperatures (that appear as peaks in the thermalconductivity curve ofFigure 3-5) should be in the lower end of the temperature range of transition

from conductor to insulator for maximum performance.

Therefore the excellent dielectric characteristic of the SF 6 gas (depending on the electronegativeproperties) is a very important feature. The electronegative properties can be explained by the factthat the Fluorine atoms miss an electron in the outer shell which means that free electrons are

effectively captured by the Fluorine atoms, thus forming negative ions.

The dielectric withstand of cold SF6 at a pressure of 0,3 MPa (abs) is comparable to transformer oil.See Figure 3-9.

In an oil circuit-breaker Hydrogen (H2) is the main constituent of the gaseous products that areformed when the arc interacts with the oil. H2 does not have good dielectric properties. This ishowever compensated by the very high pressure in the contact region after arc extinction. Apressure of 10 MPa or more at rated breaking current is not unusual.

In an air-blast circuit-breaker the pressure is also rather high, 2 to 3 MPa.

Figure 3-9 - Breakdown as a function of pressure

Air, SF6 and transformer oil


8/19


9/19


1999-01-21

Figure 3-11 - Nozzle arrangements for air-blast circuit-breakers

3.4.3 Oil circuit-breaker

The principle of the oil circuit-breaker is relatively simple as shown in Figure 3-12. This shows one

of the first types of oil circuit-breakers that was manufactured, the bulk oil circuit-breaker.

An oil circuit-breaker basically consists of two contacts that separate in oil. The intense heat of thearc that is drawn between the contacts while opening causes the oil to disintegrate mainly into

hydrogen gas. The hydrogen is used as a cooling medium.

In the case of a bulk oil circuit-breaker, the oil is used both for interruption and also as insulationmedium (between the contacts, the phases and to the earthed tank).

Figure 3-12 - Bulk oil circuit-breaker

Most modern oil circuit-breakers are of the so called minimum-oil type. These circuit-breakers use,compared to the bulk oil circuit-breaker, only a fraction of the oil by using the live tank design. Thecontacts in these types of circuit-breakers are separated in so called extinguishing chambers, highpressure resistant cylindrical constructions of insulating material. These extinguishing chambers areused to increase the pressure in the contact region and to achieve a more intensive contact

between the expanding gas and the arc (see Figure 3-13).


10/19


1999-01-21

Figure 3-13 - Simple extinguishing chamber

Modern extinguishing chambers consist of a number of these chambers in series. When the highpressure (up to 20 MPa) can escape in a symmetrical way, this is called an axial blast extinguishingchamber (Figure 3-14 a). Asymmetrical designs with a single expansion opening are called

transversal or cross-blast extinguishing chambers, illustrated by Figure 3-14 b. In the cross-blastextinguishing chamber the escaping gas drives the arc into the expansion openings that areuncovered one by one as the contact moves towards the open position.

Figure 3-14 - Basic designs of extinguishing chambersa) axial blast extinguishing principleb) cross-blast extinguishing principle

Because the cooling only takes place along the circumference of the arc, the cross-blast causesmainly an axial cooling (see Figure 3-14).In some circuit-breakers both principles are combined.The blast pressure may amount to 10 MPa or more, which together with the thermal properties of

hydrogen explains the extremely good thermal interrupting properties of oil circuit-breakers.

It is mainly the ability to withstand the TRV peak that sets the limit of the minimum-oil circuit-breaker.


11/19


1999-01-21

The interrupting capability expressed in MVA (breaking current times recovery voltage) is usuallyfairly constant over a considerable range. A modern minimum-oil circuit-breaker may thus have aninterrupting capacity of 8 000 MVA with one unit per phase. This means that the interruptingcapacity at 145 kV is 31,5 kA and 37,5 kA at 123 kV.

As the blast energy is taken from the arc, it takes some time to build up the blast pressure.Therefore, a break time of 45 to 50 ms is a practical minimum (2,5 cycles at 50 Hz or 3 cycles at 60Hz is usually stated). This interrupting principle has the advantage that the operating energyrequired is small and more or less independent of the breaking currents. Most of the energy is in

fact used at closing of the circuit-breaker.

The lifetime of the arcing contacts is 6-8 operations at rated breaking current. Not only the contactswill be affected by the arc but also the interrupting medium, oil. The oil will get carbonized and willalso set the limit for maintenance periods. This is especially valid for circuit-breakers with frequentoperation at low currents.

However, it should be noted that the oil blackens even after a few operations. This does not mean

that it has to be changed. Only after chemical analysis (acidity, water etc.) and dielectric testingindicating bad quality, should the oil be filtered or changed.

3.4.4 SF6 circuit-breakers

3.4.4.1 SF6-two pressure circuit-breakers

The first SF6 circuit-breakers in use were in principle air-blast circuit-breakers converted to workwith SF6. But instead of letting the gas out in free air (costly, environmental problems) the circuit-breakers were encapsulated and the exhaust gas was collected and compressed in the highpressure SF6 reservoir again. Due to the liquefaction of high pressure SF 6 gas at low temperatures,it was necessary to heat the high pressure reservoir. In spite of this, condensed SF 6 has caused

problems by contaminating insulators.

3.4.4.2 SF6 puffer circuit-breakers

In the SF6 puffer, the gas pressure for the cooling blast is created during the opening stroke in acompression cylinder. The piston of the compression cylinder is connected to the moving contact,so that the gas is compressed at the same time as the contacts are opened. The compressed gas isblown out through an insulating nozzle in which the arc is burning. Figure 3-15 shows the functionof a dual blast puffer, which is the most commonly used design. The insulating nozzle is made ofPTFE or PolyTetraFluorEthene.


12/19


1999-01-21

Figure 3-15 - Function of a single blast puffer

One important feature of the puffer design is the current dependent build-up of extinguishingpressure. At a no-load operation (without arc) the maximum pressure in the puffer cylinder is

typically twice the filling pressure. See the no-load curve in Figure 3-16.

A heavy arc burning between the contacts blocks flow of the gas through the nozzle. When thecurrent decreases towards zero, the arc diameter also decreases, letting more and more outlet areafree. A full gas flow is thus established at the current zero resulting in maximum cooling whenneeded. The blocking of the nozzle (nozzle clogging) during the high current interval, gives a furtherpressure build-up in the puffer cylinder that may be several times the maximum no-load pressure

(see Figure 3-16).

In other words: the decreasing puffer volume, nozzle clogging and heating of the gas by the arcinteract to create a high pressure.

The high pressure in the puffer requires a high operating force to prevent stopping or even reversalof the contact movement. The blast energy is therefore mainly delivered by the operating

mechanism.


13/19


1999-01-21

Figure 3-16 - Pressure in the puffer volume at no-load operation and during interruption of anasymmetrical short-circuit current of 40 kA

To extinguish the arc a certain blast pressure is required. It is determined by the rate-of-change ofcurrent at current zero (di/d t) and the rate of rise of the recovery voltage immediately after currentzero (du/dt). A high value of d i/d tresults in a hot arc with a large amount of stored energy at currentzero, which makes interruption more difficult. High values of d u/dt will result in an increase of theenergy to the post arc current. The following expression gives an idea of the relative importance of

the stresses:

am

pkdt

du

dt

di=

1

Where: m is in the range 2,5 to 4;

is a constant with a value 0,5;p is the gas pressure in the puffer (MPa).

The curve showing the thermal interrupting capability of a particular design of SF 6 circuit-breaker isgiven in Figure 3-17. This curve shows the maximum combination of current derivative and voltagerise that can be interrupted successfully by this circuit-breaker. The most severe thermal stress

occurs when the circuit-breaker has to interrupt a short-line-fault (SLF). Then

dt

diZ

dt

due =

where Ze is the equivalent surge impedance of the line.


14/19


1999-01-21

The stress subjected to the circuit-breaker during an IEC-type test with Ze = 450 can be

represented by a straight line in Figure 3-17. The intersection with the capability curve then givesthe maximum current derivative (or 50 Hz breaking current) that is possible to interrupt with theactual design.

In this way each puffer circuit-breaker has a rather well defined current limit. Note that the ratedvoltage has no influence on this.

A circuit-breaker with two series connected breaks will have a higher current limit, since the voltagestress is then shared by the two breaks. Three units will give a still higher limit, etc.

Figure 3-17 - Typical capability of an SF6 puffer circuit-breaker

Besides the current limit there also exists a voltage limit. This limit is independent of the breakingcurrent at low currents. During the dielectric recovery, the gas between the contacts is stillconsiderably heated, as a result of the heating by the arc an instant earlier. The increased gastemperature strongly affects the dielectric withstand capability of the gas, mainly by the decrease indensity. See Figure 3-18. In addition the gas is also slightly ionized in the upper temperature range,too little to conduct any current, but enough to distort the electrical field between the contacts ascompared to that in a cold gas. The voltage distribution during dielectric recovery is therefore

difficult to treat analytically.


15/19


1999-01-21

Figure 3-18 - Breakdown voltage, hot SF6. Pressure 0,2 MPa (abs)

Figure 3-19 gives an example of the influence of the number of series connected breaking units onthe circuit-breaker interrupting capacity. The horizontal lines give the dielectric limit of the numberof series connected breaking units and the vertical lines give the thermal limit of the series

connected breaking units.

The decreasing part towards the end of the horizontal line is caused by the effect of the reduction indielectric withstand due to an increasing amount of hot gas.

Figure 3-19 - Principal performance curves of SF6 puffer circuit-breakers


16/19


1999-01-21

3.4.4.3 SF6 self-blast circuit-breakers

Service experience has shown that circuit-breaker failures due to insufficient interrupting capacityare rare. The majority of the failures reported are of a mechanical nature. That is why there is adrive to improve the overall reliability of the operating mechanisms. Because of the fact that puffer

circuit-breakers require a lot of operating energy, manufacturers were forced to use pneumatic orhydraulic mechanisms, and the spring mechanisms used have been exposed to high mechanicalstresses. For this reason, the move towards a reduction of the energy requirement of the operatingmechanisms has been one of the main targets of recent development.

In a normal puffer circuit-breaker a small part of the pressure rise is caused by heating of the gasthrough the arc. The ideal situation would be to let the arc produce the blast pressure in the sameway as in a minimum-oil circuit-breaker. In this way the operating mechanism only needs to deliverenergy necessary for the movement of the contact.

This ideal situation can, however, not be reached at the moment. Problems will arise wheninterrupting small currents, since there is only a limited amount of energy accessible for the

pressure rise. For this reason a compromise has been reached: a self blast circuit-breaker with pre-compression.

Because the blast pressure required for interruption of small currents/current derivatives (see3.4.4.2) is moderate, a small pressure rise independent of the current is sufficient. For highercurrents, the energy producing the blast pressure is taken from the arc through heating of the gas.By using suitable check valves, the braking of the movement of the contact system can be

prevented.

Figure 3-20 shows the working principle of such a high voltage circuit-breaker.

Figure 3-20 Self-blast SF6 circuit-breaker with pre-compression

When interrupting small currents (up to some kA), the circuit-breaker operates as a pure puffercircuit-breaker: gas is compressed in the auxiliary puffer cylinder V 2 and flows through volume V1and the nozzle. In the case of a short-circuit current, the pressure rise necessary for the

extinguishing of the arc is built up in the self-blast volume V 1, through heating by the arc. A checkvalve between the volumes V1 and V2 prevents the high pressure to escape to the auxiliary puffer.


17/19


1999-01-21

The pressure in V2 is relatively independent of the current. It is limited to a moderate level bymeans of a spring loaded valve (overpressure valve), which means that the compression energyrequired from the operating mechanism is limited. Figure 3-21 shows how the energy from theoperating mechanism is used.

Compared with a conventional puffer circuit-breaker of the same rating, the energy requirements ofthe operating mechanism can be reduced to 30 %.

Figure 3-21 - Utilization of operating energy at a breaking operation

Even circuit-breakers with rotating arcs, treated in the next section, are often self-blast circuit-breakers. Current development is directed towards a further reduction of the energy requirements

from the operating mechanism, but it is doubtful whether the pre-compression can be totally omittedfor high voltage circuit-breakers with their high TRV requirements. For medium voltage there ismore potential, due to the lower requirements for the voltage derivative after current zero.

3.4.4.4 SF6 circuit-breakers with rotating arcs

Instead of blowing on a stationary arc, one can think of obtaining cooling by moving the arc througha stationary gas. This can be done by means of a magnetic field.

Maxwell's equation F = I x B says that a force is applied to a current that flows perpendicular to amagnetic field. By an arrangement as shown in Figure 3-22 it is possible to obtain a rotation of anarc between two contacts. The magnetic field is produced by letting the short-circuit current pass

through a magnet. The speed of the arc can be 100 m/s or higher depending on current andmagnetic field. When the current reaches its zero crossing, the speed decreases and at currentzero the speed is zero. Expressing the speed in the number of arc diameters per unit of time willresult in a limit value that is not zero. This in combination with the instability of low current arcs andthinning of the gas along the axis due to centrifugal force, can possibly explain why this interrupting

principle is working only in the medium voltage range (at the moment).

In order to obtain a better and safer interrupting capacity, it is possible to combine arc rotation withself blast function, possibly with pre-compression in an auxiliary puffer. Arc rotation isadvantageous for a self blast circuit-breaker, since it gives a homogenous heating of the gas.Otherwise there is a risk that the arc creates local "hot spots", that de-ionize slowly and can lead to

voltage breakdown at the peak of the TRV.


18/19


1999-01-21

Besides the very low operating energy requirement, a further advantage of arc rotation is thatcontact wear is reduced because of the fact that the foot point of the arc is forced to move. A

principle of a combined rotating arc/self blast circuit-breaker is given in Figure 3-23.

Figure 3-22 - Principle of rotating arc Figure 3-23 - Circuit-breaker with rotating arc andself blast function

3.4.5 Vacuum circuit-breakers

The heart of the vacuum circuit-breaker is the vacuum "bottle". This is an evacuated absolutevacuum-tight container made of ceramic and metal parts fused together in permanent seals. SeeFigure 3-24.

The bottle contains two "butt" contacts which carry the current. One of the contacts is movable inaxial direction. To secure the vacuum it is welded to a metal bellows (any type of gasket would not

be sufficiently tight).

Figure 3-24 - Vacuum-bottle

1 Flexibl e metallic bell ows 4 Fixed cont act

2 Moveable contact 5 Tubular metal shield

3 Insulating vacuum envelope


19/19


When the contacts open an arc is drawn in the metal vapor from the contacts (usually a copperalloy, e.g. CuCr or CuBi).

When the current goes to zero no new metal vapor is produced for a short moment. The "old" vaporis quickly absorbed by the surfaces (each atom will hit a surface within a few s and stick to it),

hence no arc can exist.

This rapid process gives the vacuum circuit-breaker a very good breaking capacity, in factsometimes too good, as mentioned in 2.7.

To prevent the arc roots to produce too much metal vapor they are kept moving by designing thecontacts so as to produce a transversal magnetic field close to the surface by forcing the current tohave a component parallel to the surface. See Figure 3-25.

Figure 3-25 - Two different contact shapes in vacuum circuit-breakers

Vacuum circuit-breakers are mostly used in the medium voltage range up to 36 kV. But designs upto 84 kV in one bottle exist and up to 145 kV with two bottles in series. The high voltage types arevery expensive and therefore they cannot compete with e.g. SF 6.Besides the good breaking capability the most important feature is the electrical endurance. Up to100 full short-circuit current interruptions are claimed possible without exchanging the bottles.

Cb Chapter 3

Documents