CIRED 19th International Conference on Electricity Distribution Vienna, 21-24 May 2007

Paper 0383

TEST EXPERIENCES WITH A NEW GENERATOR CIRCUIT BREAKER

BASED ON VACUUM TECHNOLOGY

Ren SMEETS Jrgen JGER Nils ANGER Joop HOOIJMANS

KEMA TDT Netherlands Siemens TD Testing Siemens R&D - Germany KEMA TDT Netherlands
[email protected] [email protected] [email protected] [email protected]

wish to make clear that certain requirements will not

ABSTRACT automatically be covered in this case, and case study is
necessary.
This contribution highlights the testing of generator circuit
breakers, in particular the testing of a 17.5 kV vacuum To illustrate this, several TRV waveshapes of generator
generator circuit breaker that was type-tested up to 72 kA in breaker duties (intended for a 15 kV generator rated 200
a direct circuit. Differences in the requirements of MVA) are combined in fig. 1, and compared to the T30 TRV
generator circuit breakers with general purpose breakers of a general purpose breaker with a rated voltage of 15 kV.
are highlighted. As can be seen, rate-of-rise and peak value of TRV of the
generator breaker system source - and out-of-phase fault duty
INTRODUCTION are far higher than for general purpose breakers, even those
Generator circuit breakers have to fulfil a number a special qualified for line systems (S2 class), as defined recently in
requirements that differ greatly from "general purpose" circuit IEC 62271-100 ed. 1.2 (2006). In addition, in contrast to
breakers. Apart from the very high nominal current carrying general purpose breakers, generator breakers have to deal
capacity, the faults that have to be dealt with by generator with very steep TRV following very high current, whereas
breakers have a peculiar nature. In the system source fault normally for general purpose breakers the steepest TRVs
(fault between generator breaker and generator), very steep result following moderate to low current interruption.
transformer-dominated transient recovery voltages (TRV) are In applications with lower power, the step-up transformer is
produced, whereas generator source faults (faults between often connected with a cable to the generator breaker (in stead
generator breaker and transformer) can result in missing
40 15 kV CB T30
current zeroes that prolong the arcing time. In addition,
out-of-phase fault S2 S1
the out-of-phase requirements are very severe. 35
This contribution focusses on a new development in this 30
respect: the application of a vacuum circuit breaker as
voltage (kV)

25
generator breaker. system generator
20 source
Testing has been performed with this device, as yet in a source fault
fault
direct circuit, in which currents and (transient) voltages 15
are produced by a single source. In a further stage of 10 load current
development, it is necessary to use a synthetic circuit, in 5
which an auxiliary circuit is used to produce the special generator rating 150 MVA, 15 kV
0
values of the TRV.
0 5 10 15 20
time (us)
STANDARDISATION STATUS
Fig. 1: TRV shapes of some generator circuit breaker (rated 15 kV, 200
In IEEE Std. C37-013 [1] relevant test-requirements are MVA) duties compared to general purpose 15 kV circuit breaker T30
laid down for generator circuit breakers, intended for TRV (S1 intended for cable systems, S2 for line systems)
generators ranging from 100 - 1000 MVA. Recently
(Nov. 2006) draft Amendment 1: Supplement for use with of a bus). In this case, the peak value of the system source
Generators 10 - 100 MVA was released [2]. The main topic fault TRV increases, whereas the TRV rate-of-rise decreases.
of this draft is to include TRV requirements in the generator From the examples of IEEE PC37.013a/D6 [2], proposed to
power range 10 - 100 MVA and to provide guidance on how be included as Annex B in a future version of the IEEE
cables between transformer and breaker can affect TRV. generator standard [1], it follows that only considerable length
In IEC, a separate generator breaker standard does not exist of cable (> 20 m) combined with small MVA generator
and the IEC 62271-100 circuit breaker standard excludes ratings (< 20 MVA) can be effective to reduce TRV rate-of-
generator breakers. However, practice is often to employ rise strongly, but still at the cost of increase of the TRV peak
"general purpose" breakers (type tested per IEC 62271-100) value [3].
in a range 20 - 50 MVA as generator breakers. The authors In fig. 2, TRV rate-of-rise of generator breaker fault
interruption duties are compared to the TRVs belonging to the

CIRED2007 Session X Paper No 0383 Page 1 / 4

CIRED 19th International Conference on Electricity Distribution Vienna, 21-24 May 2007

Paper 0383

maximum fault current (T100 duty) of a general purpose 15 tests, only one LC circuit is used to produce the injection
kV breaker. From this, it becomes clear that a breaker, type current, in separate tests on the first and last poles to clear
tested in accordance with IEC 62271-100, even with very respectively. The other LC unit is needed for arc prolongation.
high rated breaking current, e.g. 63 kA, generally falls short A new method for arc prolongation is developed that can
to fulfil the very severe generator TRV requirements. produce far higher current than the arc prolongation circuits
normally used in high-voltage synthetic testing. Such a circuit
TESTING is necessary to produce a realistic arc duration in synthetic
tests.
6 The auxiliary breaker for isolation of the current source
system source fault from the HV-source, must also be a generator circuit
5
TRV rate of rise (kV/us)

breaker of similar rating as the tested one.

out-of-phase fault
4 Experience with this circuit is up 120 kA at 25.3 kV [4]
3 Single-phase synthetic testing
generator source fault
2 This method is applied for the most powerful units ( up to
1
15 kV general purpose CB T100 S2 2000 MVA). The main problem is in the magnitude of
S1 current involved (up to 210 kARMS, 600 kApk). This
0 requires accurate control of the extreme forces on the
0 200 400 600 800 1000 conductors supplying the main current. Specially designed
generator rating (MVA)
heavy buswork is in use, yet allowing a certain degree of
flexibility. This is connected to the main 400 kA busbar
Fig. 2 :Rate of rise of TRV of generator circuit breaker duties [2] as a system of the laboratory which is directly connected to the
function of generator rating compared to general purpose 15 kV
current source: four generators in parallel.
circuit breaker T100 (maximum current) TRV (S1 intended for cable
systems, S2 for line systems)
The second problem is in the very steep values of TRV
which can not be reached with the standard MV TRV
High-power testing of generator circuit breakers sets extreme circuitry that is located some tens of metres from the test-
demands to test laboratories. It is the combination of high object. Also here, a synthetic circuit is employed, triggered
current and steeply rising TRV that is generally a challenging with a special making device, allowing triggering at relatively
combination to test-engineers. KEMA has the following low voltage (order of 35 kV).
solutions for generator breaker testing: Experience with this circuit is up to 210 kA at 25.3 kV [5].
Direct three-phase testing
For general purpose circuit breakers, IEC 62271-100 requires
This is a method guaranteeing the highest degree of equival- a power frequency recovery voltage during a period of at least
ence to the service situation, and can generally be applied up 300 ms from interruption. Synthetic circuits by their nature do
to 100 kA at nominal output voltage of the generators (15 not produce these ac recovery voltages, so special circuits
17 kV, depending on the power frequency). In a typical three- must be developed. For the case a breaker has to be type
phase direct circuit, the output of the generators is directly fed tested according to IEC 62271-100 and IEEE Std. C37.013
into the test-bay (without transformer) through a 17 kV (breaker can then double as generator- and high-current
busbar system designed for 400 kARMS. The testing described general purpose breaker), KEMA has synthetic circuits
in this contribution was performed in this manner. available that produce transient recovery - as well as power
Direct single-phase testing frequency recovery voltage in a single circuit.
In this method, the output voltage of the short-circuit
VACUUM GENERATOR CIRCUIT BREAKER
generators is transformed to the desired voltage level. Having
a total available power of 4800 MVA single phase (8400
MVA three-phase), KEMAs test facility is able to supply General
current up to 120 kARMS at a voltage level of 31 kV. TRV By its specific features, vacuum as an interrupting medium
requirements are met through suitable networks. behaves differently from SF6 or compressed air, that are
usually employed in generator breakers. Thanks to its well-
Synthetic three-phase testing known excellent dielectric recovery, vacuum is by nature
This method is used for testing the larger SF6 units typically better suited than gasses to cope with the fast rising TRV
25 kV with 100 120 kA (for generators in the order of 1000 involved in the generator breaking application.
MVA). In this case KEMAs synthetic installation is added to The challenges of the designer of vacuum generator breaker
the high-current circuit. Hereby, the appropriate values of then are located in the following areas :
TRV can be produced using the current injection method. 1. Due its straightforward 'butt-type' contacts, very high
KEMAs synthetic installation is a double LC circuit (energy nominal current carrying capacity of vacuum breakers can
2*1.7 MJ), designed for three-phase synthetic testing. In these only be achieved with very high contact force;

CIRED2007 Session I Paper No 0383 Page 2 / 4

CIRED 19th International Conference on Electricity Distribution Vienna, 21-24 May 2007

Paper 0383

The final breakthrough of the vacuum

generator circuit breaker came in the
90s with rated short-circuit breaking
current of 50 kA at 15 kV rated
voltage in cooperation with ENEL.
This was demonstrated with tests in
both direct and synthetic test circuits.

Actual status
The modular high-current and gene-
rator vacuum circuit breakers 3AH37
/3AH38 from Siemens was type tested
up to 72 kA in accordance with IEEE
Std. C37-013. The breaker has rated
voltage of 17.5 kV and is available for
nominal current up to 6300 A and
short-circuit current up to 72 kA.
Fig. 3: Oscillograms of 10 kA tests by Siemens (early-eighties of last century) regarding
During the design work, simulations
missing current zero with vacuum breaker (left) and oil circuit breaker (right). Upper trace:
in mechanics, electrical and thermal
current, lower trace: arc voltage.
fields were necessary. The
2. Long arcing times and high asymmetry such as occur in optimisation of contact material, contact system of the
the generator-source faults need a very careful design of interrupter as well as the mechanical properties of the breaker
(radial magnetical field) arc control; are the basis of success. Existing subassemblies of the 3AH3
3. The arc voltage of vacuum breakers is much lower than of family were used.
gasfilled breakers. The advantage is a lower thermal stress Three-phase direct tests up to 72 kA were done in the
of the breaking chambers, the disadvantage is the Siemens test laboratory and at KEMA. All development
impossibility to interact with the circuit in order to testing was done at Siemens. The final type tests were done at
advance current zero in the case of very high asymmetry KEMA.
during a generator source fault. This is illustrated in fig. 3 The increase of rated voltage up to 24 kV will be the next step
of early tests, showing that the relatively high arc voltage in the development. Therefore, an understanding of the
of a minimum oil circuit breaker can "advance" current plasma physics and the influence of the arc voltage on the
zero, whereas "vacuum" has to wait for current zero breaking process is necessary.
because of its low arc voltage. Based on the results and experience in high-current and
generator vacuum circuit-breakers breaking capabilities up to
100 kA in vacuum are possible.
Circuit breaker parameters
In fig. 4, a typical oscillogram of an interruption of 80.1 kA

History of Siemens development

The development started with the application of general 1 00

current: 50 kA/div
cs
voltage: 10 kV/div
purpose circuit breakers for extreme service situations
time: 20 ms/div
such as occur for example in the off-shore industry. In 50

such applications, short-circuit currents arise that have

an asymmetry that far exceeds the values, obtained with 0

test-circuits having 45 ms dc time constant, as the IEC

56 and its successors prescribed. Already early in the - 50

development of vacuum breakers, Siemens investigated

the limits of interruption of vacuum technology, such as - 1 00

TRV
for example occur with short-circuit currents having an
asymmetry higher than 75%. Around 1980, compara- - 1 50

tive tests were performed at 10 kA level between

minimum oil breakers and vacuum breakers with - 2 00

respect to their behaviour towards missing current zero current

(see fig. 3). This leads to unusually large, but - 80 - 60 - 40 - 20 0 20 40 60 8 0

controllable arcing times. With the 3AH38 breaker, an Fig.4: Interruption of 80.1 kARMS current with 53% asymmetry by a
arcing time of 50 ms (at a level of 31.5 kA) was vacuum genererator circuit breaker. cs: contact separation.
demonstrated. The figure shows one phase of a direct three-phase test.

CIRED2007 Session I Paper No 0383 Page 3 / 4

CIRED 19th International Conference on Electricity Distribution Vienna, 21-24 May 2007

Paper 0383

current is shown. Very high asymmetry is produced here, Another example is single-phase vs. three-phase testing. In
leading to a first asymmetrical current peak of 228 kA and the single-phase case, great caution must be exercised in
asymmetry (at contact separation) of 53%. order to simulate the phase-to-phase (dynamical, mechanical)
Fig. 5 shows the object in the KEMA test-station. In fig. 6 the interaction when testing single phase.
circuit breaker and interrupter are shown. A relative newcomer in the field of generator breakers is the
vacuum circuit breaker. Direct three-phase tests up to 72 kA
DISCUSSION AND CONCLUSION are described that confirm the performance of both test-circuit
and test-object.
Due to specific network requirements, such as a very steep
rise of transient recovery voltage after fault current inter- REFERENCES
[1] IEEE Std. C37.013-1997: "Standard for AC-High
Voltage Generator Circuit Breakers Rated on a
Symmetrical Current Basis", IEEE New York
[2] IEEE PC37.013a/D6, Nov. 2006, "Standard for AC
High Voltage Generator Circuit Breakers Rated on a
Symmetrical Current Basis - Draft Amendment 1:
Supplement for use with Generators Rated 10 - 100
MVA", IEEE New York
[3] D. Dufournet, G.F. Montillet, 2002, "Transient Recovery
Voltage Requirements for System-Source Fault
Interruption by Small Generator Circuit Breakers", IEEE
Trans. Pow. Del., vol.17, no.2, 474-478.
[4] R.P.P. Smeets, W.A. van der Linden, 1998, "The Testing
of SF6 Generator Circuit-Breakers", IEEE Trans. Pow.
Del., vol.13, no.4, pp. 1188 - 1193
[5] R.P.P. Smeets, H.D. Barts, L. Zehnder, 2006, "Extreme
Stresses of Generator Circuit Breakers", CIGRE
Conference, paper A3-306

Fig. 5: Siemens vacuum generator circuit breaker

under test at KEMA
ruption and a very high degree of asymmetry of short-circuit
current, general purpose circuit breakers are generally not
suited to fulfil generator circuit breaker duties.
Therefore, special standards have been developed (the main
points of which are described in this contribution) and special
breakers are developed.
As a consequence, test laboratories have to develop test
circuits, in order to verify the various duties required.
Because of the unusually severe requirements, also such tests
demand great efforts from test laboratories, and sometimes
there is no other choice than to perform partial tests, each part
directed towards a specific duty.
Thorough technical knowledge and understanding of the Fig. 6: Circuit breaker
background of standards is then required in order to allow type 3AH37-6300 A
subdivision of the stresses - combined in service - into several (without phase barriers)
part-tests.
An example of this is the combination of transient recovery
voltage and power frequency recovery voltage, that can in
principle be realized in a single test circuit.

CIRED2007 Session I Paper No 0383 Page 4 / 4