-
5/19/2018 VFTO documentation.doc
1/119
CHAPTER 1
INTRODUCTION
1.1 What is Gas Insulated Switchea!"
A compact, multi component assembly enclosed inside a grounded
metallic
encapsulation, which shields all energized parts from the
environment. The primary
insulating medium is compressed SF6gas.
It generally consists of
a. !us"bars
b. #ircuit"brea$ers
c. %isconnecting switches
d. &arthing switches
e. #urrent transformers
f. 'oltage transformers
g. #able and bo(es
h. )as supplying and gas monitoring e*uipment
i. %ens meters+. ocal control
)as Insulated Substations -)IS have found a broad range of
applications in power
systems over the last three decades because of their high
reliability, easy maintenance, small
ground space re*uirement etc.. In our country also, a few )IS
units have been in operation
and a large number of units are under various stages of
installation.
)IS is based on the principle of operation of complete enclosure
of all energized or
live parts in a metallic encapsulation, which shields them from
the e(ternal environment.
#ompressed SF6 gas, which has e(cellent electrical insulating
properties, is employed as the
insulating medium between the encapsulation and the energized
parts. )as Insulated
Substations have a grounded outer sheath enclosing the high
voltage inner conductor unli$e
conventional e*uipment whose closest ground is the earth
surface.
/
-
5/19/2018 VFTO documentation.doc
2/119
The !asic Insulation evel -!I re*uired for a )as Insulated
Substation -)IS is
different from that of the conventional substation because of
certain uni*ue properties of the
former. )as insulated bus has a surge impedance -01 2hm more
than that of the
conventional oil filled cables, but much less than that of a
over head line -311 4 511 2hms.
In addition, the )IS is totally enclosed and therefore is free
from any atmospheric
contamination. ence, in general the )IS permit lower !I rating
than the conventional one.
A )IS re*uires less number of lightning arresters than a
conventional one. This is mainly
because of its compactness. The basic consideration for
insulation co"ordination is '"t
characteristic. The '"t characteristic of SF6 is considerably
flat compared to that of air. Air
can withstand to very high voltages for very short time. 2n the
other hand SF6 e(hibits a flat
characteristic. Thus the ratio of basic switching impulse level
to basic lightening impulse
level is close to unity for )IS, where as for the conventional
substations this ratio varies
between 1.6 and 1.76.
1.# Ad$antaes %& GIS %$e! the C%n$enti%nal' O(en Ai!
Su)stati%ns"
/ 'ery much reduced area and volume re*uirements resulting in
lower costs.
8 )reatly improved safety and reliability due to earthed metal
housing of all high
voltage parts and much higher intrinsic strength of SF6 gas as
insulation.3 9ore optimal life cycle costs because of lesser
maintenance, down time and
repair costs.
5 &limination of radio interference with the use of earthed
metal enclosures.
: It is not necessary that high voltage or e(tra high voltage
switchgear has to be
installed outdoors.
6 They offer saving in land and construction costs.
0 These substations can be located closer to load centers
thereby reducing
transmission losses and e(penditure in the distribution
networ$.
1.* Disad$antaes %& GIS"
Although )IS has been in operation for several years, a lot of
problems encountered
in practice need fuller understanding. Some of the problems
being studied are
8
-
5/19/2018 VFTO documentation.doc
3/119
/. Switching operations generate 'ery Fast Transient 2ver
voltages -'FT2.
8. 'FT2 may cause secondary brea$down inside a )IS and Transient
&nclosure
'oltages -T&' outside the )IS.
3. Field non"uniformities reduce withstanding levels of a
)IS.
5. ;rolonged arcing may produce corrosive < to(ic
by"products.
:. Support spacers can be wea$ points when arc by"products and
metallic particles are
present.
For these reasons, 'FT2 generated in a )IS should be considered
as an important
factor in the insulation design. For designing a substation it
is essential to $now the
ma(imum value of 'FT2. 9oreover, this 'FT2 in turn generates
Transient &nclosure
'oltages -T&' outside the )IS. ence studies are carried out
on estimation of the 'FT2
and T&' levels. For this purpose ;S;I#& can be used.
In )IS, 'ery Fast Transient 2ver voltages -'FT2 are caused by
two ways, due to
switching operations, line to enclosure faults and internal
insulation flashover.
The internal FT2=s generated have traveling wave behavior of a
surge. Since FT2=s
have the characteristics of traveling wave, they can change
significantly at different points
within )IS. These FT2=s travel to the e(ternal system through
enclosures, gas"air bushings,
cable +oints, current transformers etc. and may cause damage to
the outside e*uipments li$ehigh voltage transformers connected to
the )IS.
FT2=s can also lead to secondary brea$down in )IS. Further they
may give rise to
electro"magnetic interference.
Since the contact speed of the dis"connector switches is low,
re"stri$ing occurs many
times before the interruption is completed. &ach re"stri$e
generates 'FT2=s with different
levels of magnitude.
%is"connector Switches -%S are used primarily to isolate the
operating sections of an
' installation from each other as a safety measure. !eyond this,
they must also be able to
perform certain switching duties, such as load transfer from one
busbar to another or
disconnection of bus bar, circuit brea$er etc.. Step shaped
traveling wave generated between
the dis"connector switch contacts propagates in both directions,
reflecting at the components
of )IS, thus resulting in a comple( waveform.
3
-
5/19/2018 VFTO documentation.doc
4/119
1.+ The ,ain P!%)le,s Ass%ciated with the -TO a!e as
&%ll%ws"
/ Flashover to )round at the dis"connector switch contacts.
8 Failure of electronic control circuits connected to )IS,
because of electromagnetic
interference of 'FT2.
3 %ielectric strength is reduced under 'FT2, if non"uniform
electric field is formed
by the particles -mainly metallic.
5 &ffect on components such as bushing and transformer.
: Transient &nclosure 'oltage -T&' on e(ternal surface
of the sheath. This may
cause flashover to near by grounded ob+ects.
For these reasons, 'FT2 generated in )IS should be considered as
an important
factor in the insulation design of not only gas insulated
components, but the entire substation.
The 'FT2 generated due to switching operation, the brea$down may
occur if a sharp
protrusion e(ists within the )IS. The over voltage pattern and
the 'FT2 level changes after
the 'FT2 brea$down. This type of brea$down is $nown as Secondary
!rea$down. This type
of brea$down is also possible at the switching contacts during
the current interruption. From
the insulation design point of view, this new 'FT2 level and
amplitudes of the high
fre*uency components are also important.For designing a
substation it is essential to $now the ma(imum value of 'FT2.
ence studies are carried out on estimation of the 'FT2 levels.
For this purpose ;S;I#& can
be used. In ;S;I#& simulation a suitable e*uivalent circuit
is necessary for each component
of the substation.
From the above it can be seen that the estimation of magnitudes
of 'FT2=s are
essential for the design of a )IS. This has been the scope of
this pro+ect.
1./ Ai, and Sc%(e %& the P!esent Stud0"
The present wor$ is aimed at calculating magnitude of fast
transient over voltages in
)IS due to Switching 2perations and ine"to &nclosure faults
by suitably modeling a typical
)IS system. A comparison is made for different lengths of )IS.
For better understanding of
5
-
5/19/2018 VFTO documentation.doc
5/119
the transients, they are calculated with Fi(ed Arc >esistance
and with 'ariable Arc
>esistance. Attempts have been made to compare the transients
with oad and without oad.
Therefore in the present study, the following wor$ has been
carried out.
/. The ma(imum possible 'FT2 level for 85:?' substation is
estimated.
8. The effect of each component of )IS on the 'FT2 level is
estimated separately.
3. The length of the cable termination depends on station
configuration. From 'FT2
point of view, minimum length of the cable is estimated by
considering different
switching operations.
5. A model of the spar$ channel development is proposed for
estimating the 'FT2
level.
In #hapter"8, iterature Survey, ;rinciple and )eneration of
'FT2, Secondary
!rea$down, Surges, >e"stri$es and ;re"stri$es, Trapped #harge
and #urrent #hopping in
)IS are discussed.
In #hapter"3, 9odelling of )IS #omponents, ;S;I#& models,
9odelling details,
#alculation of ;arameters and &*uivalent circuit of )IS
components are presented.
In #hapter"5, The transients due to switching operations and
line"to"enclosure faults
with Fi(ed Arc >esistance for different lengths of )IS and
also the transients due to fault
along with load and without load are described and analyzed.In
#hapter":, The transients due to switching operations and faults
with 'ariable Arc
>esistance for different lengths of )IS and also the
transients due to fault along with load
and without load are dealt with.
In #hapter"6, Suppression of fast transient over voltages is
discussed.
In #hapetr"0, #omparison between the transients due to switching
operations with
Fi(ed and 'ariable Arc >esistance for different lengths of
)IS. #omparison between the
transients due to fault for different lengths, with fi(ed and
variable arc resistance, with and
without oad, and suggestions for the further wor$ are
presented.
:
-
5/19/2018 VFTO documentation.doc
6/119
CHAPTER #
2ITERATURE SUR-E3
#.1 Int!%ducti%n"
%uring the current operation of dis"connector switch in a )IS,
re"stri$es -pre"stri$es
occur because of low speed of the dis"connector switch moving
contact, hence 'ery Fast
Transient 2ver voltages are developed. These 'FT2=s are caused
by switching operations
and line"to"enclosure faults.
@hen a dis"connector switch is opened on a floating section of
switchgear, a trappedcharge may be left on the floating section. In
the opening operation of dis"connector switch,
transients are produced and the magnitude of these transients
and rise times depends on the
circuit parameters. @hen there is a fault occurs, there is a
short circuit in the system.
Transients are also produced due to the faults in the system.
%ue to this 'FT2=s are caused
by switching operation can also lead to secondary brea$down with
in )IS. >e"stri$ing surges
generated by the dis"connector switches at )IS generally possess
e(tremely high fre*uencies
ranging from several hundred ?z to several 9z.
In this chapter, the general layout of 85:?' )IS is given in
section 8.8. The literature
survey is presented in section 8.3. )eneration of 'FT2 is
discussed in section 8.5. ;rinciple
of FT2 generation is described in section 8.:. Secondary
brea$down in )IS is e(plained in
section 8.6. The occurrences of Surges, >e"stri$es and
;re"stri$es in )IS are presented in
sections 8.0, 8.7 respectively. Trapped charge condition in )IS
is also discussed in section
8., and necessity of current chopping is described in section
8./1.
#.# Gas Insulated Su)stati%ns"
The general layout of 85:?' )as Insulated Substation comprises
the following
components
#ircuit !rea$er
Isolator
6
-
5/19/2018 VFTO documentation.doc
7/119
%is"connector Switch
&arthing Switch
#urrent Transformer
'oltage Transformer
!us bar B #onnectors
;ower Transformer
!ushing B #able
@hen designing the )IS, space"associated costs are reduced,
resulting in a substantial
reduction in overall station costs, as )IS occupies only roughly
/1C of the space re*uired by
a conventional substation. Typical cases for which )IS is
undoubtedly the more economicsolution -along with areas of ma+or
cost savings are given below
/. Drban and Industrial areas -space, pollution
8. 9ountain areas -site preparation, altitude, snow and ice
3. #oastal areas -salt"associated problems
5. Dnderground substations -site preparation
:. Areas where aesthetics are a ma+or concern -andscaping
etc.
#.* 2ite!atu!e Su!$e0"
@or$ing )roup 33
-
5/19/2018 VFTO documentation.doc
8/119
S. Ganabu E5, has e(perimentally estimated fast transient over
voltages in )IS. The
ma(imum FT2 estimated from observation was 8.0 p.u. This was
observed infre*uently and
occurred only at the open end of the bus bars.
S. A. !ogss E:, carried out field tests for measurement of
dis"connector switch
operation induced transients and indicated that transients do
not e(ceed 8.1p.u. Further it
gives that the trapped charge left during dis"connector switch
opening depends on the design
of the switch.
S. 2gawa E6, proved that re"stri$ing surge of dis"connector
switches can estimated
by conducting calculations with considerably high accuracy than
measured waveforms.
Accuracy of as low as 3C to :C has been achieved for measured
and calculated values.
H. aznadar E0, >. @itzmann E7, has developed models for
different )IScomponents and conducted e(periments with regard to
waveform distortion on various
models consisting of spacers, bushing etc..
Amir 9ansour 9iri E, presented numerical and e(perimental
evaluation of the
transient behavior of )IS. @ith the help of electrical
e*uivalent circuits of )IS components,
the generation and propagation of transients inside )IS have
been evaluated.
obuhiro Shimoda E/1, J. 2zawa E//, describes the method of
suppression of
transient over voltages caused by dis"connector switch. This is
obtained by insertion of
resistor with appropriate value during switching operation.
T. ). &ngel E/8, determined the resistance of high"current
pulsed arc by various
formulae. The results indicate that in the initial stages of
discharges -t K 1.:s, e*uation
developed by Toepler and some other authors are identical.
). &c$lin and %. Schlicht E/3, describes the operation and
switching procedures with
isolators occurring in )IS and the principle operation of FT2=s
generated in )IS.
Tohei itta E/5, describes surge propagation in )IS. Traveling
velocity of surges is
e*ual to the velocity of light. Any component, which adds e(tra
ground capacitance to the
system should be properly included in the calculation model.
Small inductance plays
important in the surge propagation performance of a given
system.
7
-
5/19/2018 VFTO documentation.doc
9/119
;. 2smo$rovic E/6, describes the formative times and Toepler=s
constant approach to
modeling the brea$down event and it depends on the macroscopic
parameters of the
insulation.
#.+ Gene!ati%n %& -e!0 ast T!ansient O$e! $%ltaes 4-TO5 in a
GIS"
%uring the current operation of dis"connector switch in a )IS,
re"stri$es -pre"stri$es
occur because of the low speed of the dis"connector switch
moving contact, due to the very
fast voltage collapse within a few nano seconds -ns and the
subse*uent traveling waves,
'ery Fast Transient 2ver"voltages are developed. The main
oscillation fre*uency of the fast
transients depends on the configuration of )IS. 9oreover, the
effect of comple(ity of the
configuration of a )IS on the pea$ value of the transients has
been studied in this thesis.
For the development of e*uivalent circuits, low voltage step
response measurements
of the main )IS components have been made. Dsing the ;S;I#&
the e*uivalent electrical
models are developed. The pea$ value of the fast transients
often occurs when circuit
structure is relatively simple, but more fre*uently if the
structure is rather complicated. The
propagation velocity of traveling wave generated during
dis"connector switch operation is
about 31cm < ns.
The representation of bushing is important for simulating the
fast transients.
)enerally, the transit time through a bushing is comparable to
or greater than the rise time of)IS generated transients. For this
reason, bushings cannot be considered as a lumped element
in estimating the 'FT2 level.
The generation of fast transients can be classified into two
types. They are due to the
following
a %is"connector switch operation
b Faults between !us bar and &nclosure
In case of line"to"earth fault, the voltage collapse at the
fault location occurs in a
similar way as in dis"connector gap during re"stri$ing. !y this
event, step shape traveling
surges are in+ected. For such a surge source inside )IS, two
surges traveling in opposite
directions are generated. owever, if voltage collapse occurs at
the open end of )IS, only
single surge propagates on the bus.
-
5/19/2018 VFTO documentation.doc
10/119
Spar$ collapse time is defined as the time to bridge the gap
with the spar$ after the
initiation of brea$down. A longer spar$ length causes longer
spar$ collapse time. It was also
observed that with a constant SF6gas pressure, a higher inter
electrode brea$down voltage
causes longer spar$ collapse time. @ith the same voltage, a
lower gas pressure also causes
longer spar$ collapse time.
@hen SF6brea$down occurs it re"combines very *uic$ly, since it
has a high electro"
negative property. %ue to this property, re"stri$ing voltages of
the order of nanoseconds rise
time are produced. ence FT2=s are mainly because of SF6. As a
conse*uence of
characteristics of brea$down in electro"negative gases and short
traveling wave times in )IS
resulting from short overall length, transient over"voltages
with steeper voltage rise and
higher fre*uencies are produced.
!rea$down in SF6starts initially by avalanche, starting with
initiatory electron due to
cosmic radiation, field emission or several other phenomena
producing electrons. These
electrons are accelerated by electric field thereby increasing
its $inetic energy. As a result,
number of electrons increases because of collisions. According
to streamer criteria, first
avalanche occurs followed by chain of avalanches bridging the
gap between the electrodes
and thus forming a streamer. Thus, to have brea$down there
should be sufficient electric field
to produce se*uence of avalanches and there should be atleast
one primary electron to initiate
first avalanche.In the above se*uence of events there e(ists a
time lag for initiating electron to be
available in the gap after the voltage is applied. This time lag
is termed as the Statistical Time
ag. Similarly the formation of spar$ channel ta$es definite time
$nown as Formative Time
ag -Tf and is defined below E/0.
=D
?l5.5T
Tf
@here l L Spar$ ength
?TL Toepler=s #onstant
D L Ignition 'oltage
This time lag is of the order of nanoseconds. Therefore the rise
time of FT2=s will be
of the order of nanoseconds. The above phenomenon suggests that
the FT2=s are generated
/1
-
5/19/2018 VFTO documentation.doc
11/119
due to voltage collapse, which occurs when spar$ is produced.
This spar$ is produced after a
time lag of Tf.
%is"connector Switches -%S are designed to interrupt small
charging current that
flows through the short lines as fast as the circuit brea$er. In
this case, since the contact speed
of %S is generally slow, re"stri$ing occurs a number of times
before interruption is
completed, resulting in generation of high fre*uency surge
voltage each time re"stri$e ta$es
place. %S operation in )IS generates the largest line"to"ground
voltage transients imposed on
the switchgear during normal operation.
#./ P!inci(le %& TO Gene!ati%n"
%uring opening operation of %is"connector Switch -%S, transients
are produced due
to internal oscillations. The magnitude of these transients and
rise times depends on the
circuit parameters li$e Inductance, #apacitance and #onnected
oad. Assuming that some
trapped charge is left during opening operation, transients can
be calculated during closing
operation of %S.
Fast Transient 2ver voltages generated during %is"connector
Switch operation are a
se*uence of voltage steps created by voltage collapse across the
gap at re"stri$ing. Specific
over voltage shape is formed by multiple reflections and
refractions. 2peration of %is"
connector Switch -%S can be shown by using the below figure
i #.1 Elect!ic Ci!cuit &%! e6(lainin !est!i7es
//
-
5/19/2018 VFTO documentation.doc
12/119
@here /L Inductance of Source
#/L #apacitance of Source
#8L #apacitance of )IS 2pen ;art
D/L ;ower Fre*uency 'oltage
D8L 'oltage of )IS Section
The more fre*uent service situation of the isolator is its use
to connect or dis"connect
unloaded parts of the installation as is shown in figure 8./.
For e(ample, a part of the )IS is
dis"connected by an isolator from a generator or from an
overhead supply line, where by the
self"capacitance #8of this part of circuit can be upto several
nF, depending on its length.
First re"stri$e across the gap occurs when voltage across the
gap e(ceeds the
brea$down voltage. The occurrence of se*uence of re"stri$es is
described with the following
figure 8.8.
i #.# -%ltae %& the %(enended GIS side %& the
Is%lat%!
/8
-
5/19/2018 VFTO documentation.doc
13/119
The voltage across the gap is the difference between D/and D8.
If it is assumed that
the brea$down voltage D!of the gap increases with increasing
separation and therefore with
time as shown in figure 8.8. Then the curve D8can be constructed
as follows.
At the instant of mechanical contact separation, D/and D8have
the same value, the
voltage D8 continues to retain this value, while D/ changes with
power fre*uency. The
voltage -D8" D/ across the gap of the isolator also changes. As
soon as, -D 84 D/ e(ceeds
the dielectric strength D!of the gap, a brea$down and thus a
first re"stri$e occurs. !oth
electrodes are there by electrically connected by a conducting
spar$, whereby )IS section
with initial voltage D8is very rapidly charged to instantaneous
value of D /. The transient
current flowing through the spar$ then interrupts as soon as the
)IS have been charged to D /
and spar$ e(tinguishes.
The voltage D8now remains constant with time, while the voltage
D /, on the side of
supply $eeps changing. This continues until the second re"stri$e
occurs with an increased
brea$down voltage D!as a conse*uence of larger separation. ence
D 8 follows D/, until
finally at the end of the switching process the gap no longer
can be bro$en down. Transients
are also produced due to faults in the system. @hen there is a
fault, there will be short circuit
in the system. %ue to this, oscillations occur due to presence
of inductance and capacitance
on both sides of the fault section causing transients.
#.8 Sec%nda!0 9!ea7d%wn in a GIS"
'ery Fast Transient 2ver voltages -'FT2 caused by switching
operations can lead to
Secondary !rea$downs within )as Insulated Substations.
In the first type, the flashover to ground at the dis"connector
switch contacts is due to
the streamer generated during re"stri$e or pre"stri$e between
the dis"connector switch
contacts. Secondly, inside the )IS, li$e particles or fi(ed
protrusions cause an
inhomogeneous field distribution and insulation can fail. In
these two types of earth faults,
'FT2 are developed. The flashover voltages under these two
conditions are appreciably
lower than the normal withstand voltages to the ground.
;ractically, it can be observed that,
/3
-
5/19/2018 VFTO documentation.doc
14/119
/. Streamers are generated from several locations over a
contact. Apparently one of
these streamers develops a flashover between the contacts, while
the flashover to ground is
caused by the development of the other streamers.
8. The flashover voltage to ground is lower when the spar$ is
generated between the dis"
connector switch contacts by an impulse voltage than when the
spar$ is simulated with a
piece of wire. This is because of the e(istence of
streamers.
;ractically, it can be observed that the 'FT2 induced earth
faults are possible at the
dis"connector switch contacts during its operation. This is
because of the development of the
enhanced field gradient to earth and later 'FT2 will be
generated in the )IS.
The brea$down from the live conductor to the outer conductor is
possible under
'FT2 or impulse voltages. Thus it is important to develop a
simulation model for the
brea$down and the characteristics of the spar$ channel. The time
varying process during
voltage brea$down and the resulting 'FT2 can be measured. The
computer simulation
model for this brea$down can be developed. The results obtained
with ;S;I#& are compared
with measured values. The time varying process during the
building of the spar$ will be
simulated by using the Toepler=s spar$ law.
#. : Su!es in GIS"
The discharge process during each individual re"stri$e begins
with a voltage collapse
across the contact gap, which because of the particular
brea$down mechanism in
electronegative gases ta$es place within only appro(imately /1
"7 sec. This voltage collapse is
directly related to the formation of the spar$ channel. @ith a
typical voltage decrease rate of
/1/3v
-
5/19/2018 VFTO documentation.doc
15/119
towards the open end of the )IS and is there again reflected.
For this reason, the discharge
transient shows a periodicity of double the traveling time of
the wave in the )IS.
The amplitudes of the voltage and current surges depend on the
re"stri$ing voltage
and on the parameters of the circuit. Therefore very different
amplitudes can occur depending
on the comple(ity of the installation.
#. ; ReSt!i7es and P!eSt!i7es in GIS"
%is"connector Switch -%S operation typically involves slow
moving contacts which
results in numerous discharges during operation. For e(ample, a
floating section of
switchgear between a disconnect switch and an open brea$er -load
side may be disconnected
from an energized )as Insulated System -supply side.
For capacitive currents below M / amp, a re"stri$e occurs every
time the voltage
between the contacts e(ceeds the dielectric strength of the
gaseous medium between them.
&ach re"stri$e generates a spar$, which e*ualizes the
potential between the switch
contacts. Following spar$ e(tinction, the supply and load side
potentials will deviate
according to the A# supply voltage variation and the discharge
characteristics of the load
side respectively. Another spar$ will result when the voltage
across the electrode gap
dependent brea$down voltage D!and the potential difference of
the load and supply side, D.&ach %is"connector Switch -%S
operation generates a large number of ignitions
between the moving contacts. The number of ignitions depends on
the speed of the contacts.
The largest and steepest surge voltages are generated only by
those brea$downs at the largest
contact gap. Therefore, only a few brea$downs -/1 4 :1 need be
considered for dielectric
purpose.
The slow operation and very rapid brea$down give rise to
T>A;;&% #A>)& and
traveling wave surges within )as Insulated Substation -)IS.
#. < T!a((ed Cha!e in GIS"
@hen a %isconnect Switch is opened on a floating section of
switchgear, a Trapped
#harge may be left on the floating section. The potential caused
by this charge will decay
/:
-
5/19/2018 VFTO documentation.doc
16/119
very slowly as a result of lea$age through spacers. A trapped
charge near /.1 p.u -pea$ can
levitate particles.
;article motion under %.# conditions is much more severe than
that for A.#
e(citation and may lead to scattering of particles onto
insulating surfaces. owever, such
particle motion leads to appreciable -A %.# currents, which will
normally discharge the
floating section in a relatively short time.
A trapped charge of / p.u implies that the first brea$down upon
closing the disconnect
switch will occur at 8 p.u across the switch contacts and may
lead to conductor4to4ground
over voltages of upto 8.: p.u. Thus the magnitude of trapped
charge left after operation of a
disconnect switch may be of some conse*uence to switchgear
reliability.
%uring recent field tests on a :11 ?' sub station, measurements
were made of the
trapped charge left when a %S was opened onto a floating section
of switchgear. umerous
measurements led to the conclusion that for this switch, a
potential of 1./ 4 1.8p.u is left on
the floating section and that this result is consistent. The
reason for this consistent result is
that the negative brea$down occurs at appro(imately /:C greater
potential difference than
the positive brea$downs for this switch.
The asymmetry in brea$down voltages leads to the NfallingO
pattern near the end of
operation which continues until the potential is low enough that
brea$downs can occur
during the rising portion of a power fre*uency cycle as shown in
below figure 8. 3.
i #. * 2%ad side $%ltae wa$e&%!, du!in %(enin %&
disc%nnect switch
/6
-
5/19/2018 VFTO documentation.doc
17/119
Two such brea$downs bring the potential bac$ to a large positive
value after which
the falling pattern is re"established. The end point of this
process is inevitably a transition
from a large negative potential to a slightly positive potential
at a gap distance for which the
positive brea$down potential is /./ p. u -pea$ and the negative
brea$down potential is
/.8 p. u -pea$. At this point another positive and negative
brea$down cannot occur, as a
result 1./ " 1.8 p. u -pea$ is left on the floating
switchgear.
The salient features which lead to this small trapped charge are
the asymmetry in
brea$down potential and relatively long arcing time. This
trapped charge can be controlled
through careful design of contact geometry. For the purpose of
calculating transient
magnitudes, a trapped charge of /.1 p. u -pea$ prior to closing
of %is"connector Switch -%S
is assumed. 2ne of the methods suggested to suppress these over
voltages is by insertion of a
resistor with an appropriate value during switching.
#. 1= Cu!!ent Ch%((in"
@hen a #ircuit !rea$er -#.! is made to interrupt low inductive
currents such as
currents due to no load magnetizing current of a transformer, it
does so even before the
current actually passes through zero value, especially when the
brea$er e(erts the same de"
ionizing force for all currents within its short circuit
capacity. This brea$ing of current before
it passes through the natural zero is termed as N#urrent
#hoppingO.
The energy contained in the electro"magnetic field cannot become
zero
instantaneously. The only possibility is the conversion from
electro"magnetic to electro"static
of energy.
i.e. #'8
/,I
8
/ 88 =
I#
,' =
/0
-
5/19/2018 VFTO documentation.doc
18/119
)enerally in 'acuum or SF6circuit brea$ers the currents chopped
are of the order of
: Amps. @hen a constant de"ionizing force is applied by a
brea$er for arc interruption, then
force must be high enough to interrupt highest value of short
circuit current.
i #. + wa$e&%!, %& %$e! $%ltae with cu!!ent ch%((in
ow, if the brea$er is called upon to brea$ a load current which
is less than thehighest short circuit current, then the de"ionizing
force would be sufficient enough to force
the arc from its high value straight to zero before the same
actually reaches to natural zero.
This results a tremendous amount of over voltage as shown in the
above figure 8.5. This
phenomenon is termed as N#urrent #hoppingO.
/7
-
5/19/2018 VFTO documentation.doc
19/119
#.11 C%nclusi%n"
Switching operations in a )as Insulated Switchgear lead to very
fast transient
phenomena. These 'FT=s stress the e*uipment in )IS as well as
the secondary e*uipment.
Switchgear reliability can be improved by assuring that
dis"connectors minimize the trapped
charge left on the switchgear. >educed trapped charge carries
two benefits. Firstly, the
magnitude of dis"connector operation induced transients is
reduced and Secondly, the
tendency for free conducting particles to be scattered onto
spacers is reduced.
ence it is essential to $now the ma(imum value of 'FT2=s
produced in the
switching operation. For this reason ;S;I#& is used. In
pspice simulation a suitable
e*uivalent circuits is necessary for each component of the
substation. The designed
e*uivalent circuit of each component in the substation using
pspice simulation is used in the
5thand :th#hapters.
/
-
5/19/2018 VFTO documentation.doc
20/119
CHAPTER*
>ODE22ING O GIS CO>PONENT OR
CA2CU2ATION O TRANSIENTS
*.1 Int!%ducti%n"
For accurate analysis of transients, it is essential to find the
'FT2=s and circuit
parameters. %ue to the traveling nature of the transients the
modelling of )IS ma$es use of
electrical e*uivalent circuits composed by lumped elements and
especially by distributed
parameter lines, surge impedances and traveling times. The
simulation depends on the *uality
of the model of each individual )IS component. In order to
achieve reasonable results in )IS
structures highly accurate models for each internal e*uipment
and also for components
connected to the )IS are necessary.
The dis"connector spar$ itself has to be ta$en into account by
transient resistance
according to the Toepler=s e*uation and subse*uent arc
resistance of a few ohms. The waveshape of the over voltage surge
due to dis"connector switch is affected by all )IS elements.
Accordingly, the simulation of transients in )IS assumes an
establishment of the models for
the !us, !ushing, &lbow, Transformers, Surge Arresters,
!rea$ers, Spacers, %is"connectors,
and &nclosures and so on.
In this chapter, the modeling concept of )IS is given in section
3.8. The ;S;I#&
models are developed in section 3.3. #alculation of parameters
of )IS is described in section
3.5. &(perimental apparatus of )IS is described in section
3.:. Single"line diagram
dimensions of 85:?' )IS are given in sections 3.6 B 3.0
respectively. The e*uivalent circuit
of )IS components is given in section 3.7.
81
-
5/19/2018 VFTO documentation.doc
21/119
*.# >%dellin C%nce(t"
A )IS system comprising of an Input #able, Spacer, %is"connector
Switch, !us bar
of :mts length and load has been considered for modeling into
electrical networ$ and
analysis.
The Fast Transient 2ver voltage waveform generated during
#losing and 2pening
operation of %is"connector Switch and ine"to"&nclosure
faults has been considered for
calculations.
Spacers are simulated by lumped #apacitance. The Inductance of
the busduct is
calculated from the diameters of #onductor and &nclosure.
#apacitances are calculated on
the basis of actual diameters of inner and outer cylinders of
central conductor and outer
enclosure. #one Insulators used for supporting inner conductor
against outer enclosure are
assumed to be dis$ type for appro(imate calculation of spacer
capacitance.
The busduct can be modeled as a series of ;i"networ$ or as
se*uence parameters.
owever in this model, it is considered as distributed
;i"networ$. The Schematic %iagram of
a Typical )as Insulated System -)IS is shown in below figure
3./.
i *.1 Sche,atic dia!a, %& a t0(ical Gas Insulated
Su)stati%n
Assuming that some trapped charge is left on the floating
section of switchgear during
opening operation of dis"connector switch, a voltage of certain
value is considered during
simulation.
8/
-
5/19/2018 VFTO documentation.doc
22/119
*.* PSPICE >%dels"
To simulate the 'ery Fast Transient 2ver voltages in )IS,
;S;I#& is used. The
e*uivalent circuit of )IS is shown in below figures 3.8 B
3.3.
i *.# E?ui$alent ci!cuit %& GIS
@here,
H/L Surge Impedance of )as Insulated !us duct w.r.to
&nclosure Interior surface
H8L Surge Impedance of 2verhead Transmission ine w.r.to
&arth Surface
H3L Surge Impedance of &nclosure &(terior Surface w.r.to
&arth Surface
#bL #apacitance of the !ushing
# L #apacitance of the #urrent Transformer
i *.* E?ui$alent ci!cuit %& GIS
88
-
5/19/2018 VFTO documentation.doc
23/119
@here l/, l8be the length of the source side bus bar, length of
the load side bus bar, #/
and #8are source side capacitance and the load side capacitance
respectively. et H cand lcbe
the surge impedance and length of the cable respectively.
For estimating these voltages, the e*uivalent impedance networ$s
for the components
li$e #apacitance, Inductance of the )round @ire, )rounding )rid,
Spar$ #hannel, and the
>esistance of )round )rid, Switch -@hich follows Toepler=s
Spar$ aw are re*uired.
*.+ Calculati%n %& Pa!a,ete!s"
*.+.1 Calculati%n %& Inductance"
The inductance of the bus duct can be calculated by using the
formula E/6
given below
@here r/, r8, r3, r5, are the radii of the conductors in the
order of decreasing
magnitude and Pl= is the length of the section.
+
+
+
= /
r
rln
r
r"/
r
r
8r
rln
r
rln
r
rln1.11/,
8
/
8
/
8
8
/
8
3
5
/
8
3
/l
i *.+ C!%ss secti%n %& t0(ical GIS S0ste,
83
-
5/19/2018 VFTO documentation.doc
24/119
*.+.# Calculati%n %& Ca(acitance"
The #apacitance is calculated with the assumption that the
conductors are
#ylindrical. #apacitance is calculated by using the standard
formulae given below
@here oL 7.7:5 Q /1"/8, r L /
b L 2uter #ylinder >adius
a L Inner #ylinder >adius
l L ength of the Section
*.+.* Calculati%n %& Ca(acitance due t% S(ace!"
Spacers are used for supporting the inner conductor with
reference to the outer
enclosure. They are made with Allumina filled epo(y material
whose relative
permittivity -r is 5. The thic$ness of the spacer is assumed to
be the length of the
capacitance for calculation.
*.+.+ Calculati%n %& Sh%!t Ci!cuit Inductance @
Resistance"
Assuming a short circuit fault level of /1119'A for /38?' system
voltage,
Inductance and >esistance are calculated as follows
S'Qph
=phI
ph'
S=phI
And'
IQRHC =
85
-
5/19/2018 VFTO documentation.doc
25/119
I
'QCHR =
!ut ,QfQQ8R =
fQQ8
R,
=
And it is assumed that > L R
*.+./ Calculati%n %& Inductance due t% 2%ad
4T!ans&%!,e!5"
For 6119'A, /38?' transformer with /1C impedance and 1.7 power
factor
the inductance is calculated as follows
;#osQIQ'Q3 =
=#osQ'Q3
;I
And'
IQRHC =
I
'
QCHR =
!ut ,QfQQ8R =
fQQ8
R,
=
*.+.8 Calculati%n %& -a!ia)le A!c Resistance"
!ased on earlier studies in SF6 gas, Toepler=s Spar$ aw is valid
for
calculation of 'ariable Arc >esistance. The 'ariable Arc
>esistance due to Toepler=s
formulae E: is given below
8:
-
5/19/2018 VFTO documentation.doc
26/119
> L( )+
t
1o
T
ti*
l?
dt
@here ? TL Toepler=s #onstant L 1.11: volt.sec-t, is calculated
until it reaches a
value of / to 3 ohms. The integral in the denominator sums up
the absolute value of
current Pi= through the resistance >-t over the time
beginning at brea$down inception.
Thus, it corresponds to the charge conducted through the spar$
channel upto timePt=.
Initial charge *ois an important parameter while considering the
non"uniform
fields. !ut the field between the dis"connector contacts is
almost uniform. Therefore
*ois very small.
*./ E6(e!i,ental A((a!atus 4>%dellin details5"
A )IS unit with the following arrangement is assumed for
developing the model as
shown in below figure 3.:.
86
-
5/19/2018 VFTO documentation.doc
27/119
i *./ S7etch %& E6(e!i,ental A((a!atus
The apparatus has a dis"connector with an earthing switch, four
dis$"type spacers, a
load bus bar about /1m long with three post"type spacers and a
::1?' gas bushing
containing stress capacitor.
The / )z surge sensor mentioned in the diagram is located at a
distance of /.6m
from the dis"connector. Further, holding the load side bus bar
at zero potential, dc voltage
80
-
5/19/2018 VFTO documentation.doc
28/119
was applied from the high voltage dc power supply to the bushing
via a / 9 resistor and
'FT2 waveform of the closing operation was observed.
The dc voltage applied was positive and moving contact of the
dis"connector was
located on the load side.
*.8 Sinle2ine Dia!a, %& #+/ - Su)stati%n"
87
-
5/19/2018 VFTO documentation.doc
29/119
i *.8 Sinleline dia!a, %& #+/7$ GIS
*.: Di,ensi%ns %& a #+/ - Gas Insulated Su)stati%n"
8
-
5/19/2018 VFTO documentation.doc
30/119
#omponents of a )IS %istance in meters
/. 2verhead Transmission ine
8. #able
3. #able to ightening Arrester -A
5. #able to 'oltage Transformer -'T
:. 'T to #urrent Transformer -#T
6. #T to &arthing Switch -&S
0. &S to %is"connector Switch -%S/
7. %S/ to !DS"II
. !DS"II
/1. !DS"II to %S3
//. %S3 to &arthing Switch -&S/
/8. &S/ to #ircuit !rea$er -#!3
/3. #!3
/5. #!3 to #!:
/:. %S: to ;ower Transformer -;T
/6. ;T to %S6
/0. %S6 to &arthing Switch -&S8
/7. &S8 to #!5
:111
7111
/.3
8.1:
/.8
1.3:
/.:
1.:
/1
1.:
3.3:
1.5
8.:
1.
//
/:
1.:
1.:
*.; E?ui$alent ci!cuit %& GIS c%,(%nents"
31
-
5/19/2018 VFTO documentation.doc
31/119
E2E>ENT >ODE2EBUI-A2ENT
CIRCUITCHARACTERISTIC
!DS %D#T
Transmission line with
distributed parameters.
oss in transmission line
because of s$in effect.
S;A#&>umped #apacitance
towards the ground.# 81pf
&!2@
Transmission line with
distributed parameters
and capacitance added in
between the line.
;arameters depending on
the ratio between conduct
and enclosure radius. 'alu
of the capacitance #
depending on the system
topology.
#A!&
Transmission line with
distributed parameters.
&ach end of cable is
terminating with a
lumped capacitance.
#D>>&T
T>ASF2>9&>
umped capacitance
towards the ground
#A;A#ITI'&
'2TA)&
T>ASF2>9&>
umped capacitance
towards the ground
!DSI)
-#apacitively
)raded !ushing
Transmission line of
varying surge impedances
are connected in series
Hg/, Hg8, U are variable
surge impedance in SF6
side. Ha/, Ha8, U are
variable surge impedance
air side.
3/
-
5/19/2018 VFTO documentation.doc
32/119
SD>)&
A>>&ST&>
Arrester capacitance is
considered. ;rotection
characteristic connected
in parallel with arrester
capacitance
In case of 'FT -1.:Vs the
protection characteristic i
corrected in reference to t
characteristic for the surge
7
T>ASF2>9&>
umped capacitance
towards the ground
'alue of capacitance
depends on the transforme
type, voltage level, windin
connection and winding
type.
%IS"#2T2>
#2S&%
Transmission line with
distributed parameters.
#apacitance of the
switching contacts
towards the ground is
considered.
;arameters depending on
the ratio between conduct
and enclosure radius. 'alu
of capacitance # depends
on the system topology.
%IS"#2T2>
2;&&%
Inter electrode
capacitanceof the
switching contacts
towards the ground is
considered.
# includes spacer
capacitance also.
&A>T
S@IT#I)
umped capacitance
towards the ground.
S;A>?
>&SISTA#& -in
case of %S operation
It is a non"linear function
of time. It varies
according to the Toepler=s
Spar$ aw
if t K /Vs, > L 1
if t /Vs, > varies
from 1 to :
38
-
5/19/2018 VFTO documentation.doc
33/119
S;A>?
-earth fault
Spar$ resistance varies
according to Toepler=s
Spar$ aw. is the
inductance of the spar$
channel.
> is in the range of
/ to 3
#I>#DIT
!>&A?&> -#.!
#2S&%
Transmission line with
distributed parameters
e*uivalent capacitance of
switching contacts
towards the ground is
considered.
The surge impedance of
#.! bus duct is less than
because of additional
capacitance.
#I>#DIT
!>&A?&> -#.!
2;&&%
The capacitance between
switching contacts is
considered. #.! bus duct
is represented with
distributed parameters on
both sides of the contacts.
The length of bus duct o
both sides of contacts is
e*ual. The inter electrod
capacitance incase of #.!
high, because of large arc
the contacts.
T>ASF2>9&>@here r L /V
L /8.7 m
*.< C%nclusi%ns"
A model is developed for the prediction of the 'FT2 phenomena in
the circuit of
voltage and current transformers in )IS. The main advantage of
such model is to enable the
transient analysis of )IS. A spar$ collapse time was correctly
simulated by the variable
33
-
5/19/2018 VFTO documentation.doc
34/119
resistor. !y this spar$ collapse time, resistance of the 'FT2 is
e(tended, and the component
caused by short surge impedance discontinuities such as spacers,
dis"connectors and short
bus branches were damped.
A )IS system comprising of spacers, bus bar and dis"connectors
has been considered
for modeling into electric networ$. The inductance of the bus
bar is calculated from
diameters of conductors and enclosure using standard formulae.
#one insulators used for
supporting inner conductor against outer enclosure are assumed
to be dis$ type for
appro(imate calculation of spacer capacitance. The busduct
capacitance is calculated using
formulae for concentric cylinders. The entire bus length is
modeled as distributed pi"networ$.
CHAPTER+
TRANSIENTS DUE TO SWITCHING @ AU2TS WITH
IED ARC RESISTANCE
35
-
5/19/2018 VFTO documentation.doc
35/119
+.1 Int!%ducti%n"
%uring the switching operation of the circuit, the transients
are developed. !y the
calculated values of the circuit parameters in previous chapter,
the e*uivalent circuits are
constructed by using ;S;I#& software. !y using the circuits
the transients are calculated for
different lengths of )as insulated substation. The transients
are also calculated during the
faults with and without load at different distances.
#onsider a circuit with the elements as shown in below figure
5./.
i +.1 Elect!ic ci!cuit &%! e6(lainin Rest!i7es
I/, I8are Isolators and
#! is #ircuit !rea$er
In this chapter, the transients due to switching operations for
:mts and /1mts length
)IS are given in section 5.8. The transients due to faults for
:mts and /1mts length )IS
without load are presented in sections 5.3./ B 5.3.8
respectively. The results of transients due
to faults for :mts and /1mts length )IS with load are described
in sections 5.3.3 B 5.3.5
respectively.
+.# T!ansients due t% switchin %(e!ati%n"
+.#.1 Sinle Phase e?ui$alent ci!cuit &%! /,ts lenth GIS"
The bus duct is divided into three sections of length 8.:mts,
/.:mts, and /.1mts
respectively from load side. The )IS bushing is represented by a
capacitance of 811pf. A
3:
-
5/19/2018 VFTO documentation.doc
36/119
Fi(ed >esistance of 8ohms of the spar$ channel is connected
in series with the circuit
brea$er. The e*uivalent circuit is shown in figure 5.8.
%ue to trapped charge some voltage remains on the floating
section which can create
severe conditions because the first re"stri$e can occur at the
pea$ of power fre*uency voltage
giving a voltage of 8. 1 p.u. 2n re"stri$e the voltages on each
side will collapse initially zero
and hence creating two /.1 p.u voltage steps of opposite
polarities. In this, it is assumed that
re"stri$ing is created at /.1 p.u and "/.1p.u respectively on
either side of dis"connector Switch
-%S. The transients due to different switching operations are
observed.
i +.# Sinle Phase e?ui$alent ci!cuit &%! /,ts lenth GIS due
t% Switchin %(e!ati%n
Dsing the circuit given in Fig 5.8, transients due to closing of
the circuit brea$er are
calculated as given in Fig 5.3. 9a(imum voltage obtained is
3.18p.u with a rise time of 31ns.
The graphs are obtained from ;S;I#& simulations and software
is given in Appendi("/.
In figure 5.8, the voltages before and after circuit brea$er is
ta$en to be /.1 p. u and
" /.1 p.u as the most onerous condition. !ut depending on the
time of closing of #.!, the
magnitude of the voltage on the load side changes.
36
-
5/19/2018 VFTO documentation.doc
37/119
i +.* T!ansient $%ltae wa$e&%!, du!in Cl%sin %(e!ati%n
%& C9 &%! /,ts GIS
36
-
5/19/2018 VFTO documentation.doc
38/119
For different values of voltages on the load side, the
magnitudes and rise times of the
voltage waveform are calculated $eeping source side voltages as
constant at /.1 p.u. The
values are tabulated as shown in Table 5./.
S. N%
2%ad Side -%ltae
4(.u5
>anitude %&
-%ltae 4(.u5
Rise Ti,e
4ns5/ "/.1 8.5:
8 "1. 8.33 /8
3 "1.7 8./0
5 "1.0 8./5 /1
: "1.6 /.6 /8
6 "1.: /.70 /1
0 "1.5 /.01 /8
7 "1.3 /.61 /8
"1.8 /.50 /8
/1 "1./ /.37 /1Ta)le +.1 T!ansients due t% $a!iati%n %&
$%ltae %n l%ad side
Similarly by changing the magnitudes of the voltage on the
source side, $eeping
voltage on load side constant at "/.1 p.u. Then the transients
due to variation of voltage on
source side obtained. The values are tabulated as shown in Table
5.8.
S. N%S%u!ce Side -%ltae
4(.u5
>anitude %&
-%ltae 4(.u5
Rise Ti,e
4ns5
/ /.1 8.5:
8 1. 8.37 //
3 1.7 8.86 //5 1.0 8.1 /8
: 1.6 8.18 //
6 1.: /.75 /8
0 1.5 /.03 /1
7 1.3 /.63 //
1.8 /.:1 /1
/1 1./ /.37 /1
Ta)le +.# T!ansients due t% $a!iati%n %& $%ltae %n s%u!ce
side
%uring closing operation, the current through the resistance of
the circuit brea$er is
shown in Fig 5.5. From the graph, it was found the ma(imum
current is 31mA at a rise time
of /8ns.
30
-
5/19/2018 VFTO documentation.doc
39/119
To introduce current chopping, the circuit brea$er is opened.
ence to calculate
transients due to opening operation the #.! is opened at /1ns
-say. The transients are
obtained and as shown in Fig 5.:. From the graph, the ma(imum
voltage obtained is 8.1/p.u
with rise time of /7ns.
Assuming that there is a second re"stri$e, another switch is
connected in parallel to
the circuit brea$er for simulation in ;S;I#& modeling.
Transients are calculated by closing
this switch when voltage difference across the contacts of the
circuit brea$er reaches
ma(imum value. Transients calculated due to second re"stri$e
gives the pea$ voltage of
8.88p.u at a rise time of /6ns as shown in Fig 5.6. The values
are tabulated as shown in
below Table 5.3.
>%de %& O(e!ati%n>anitude %&
-%ltae 4(.u5
Rise Ti,e
4nan% sec%nds5
%uring #losing
2peration3.18 31
%uring 2pening
2peration8.1/ /7
%uring Second
>e"Stri$e8.88 /6
Ta)le +.* T!ansients due t% switchin %(e!ati%ns &%! /,ts
lenth GIS
37
-
5/19/2018 VFTO documentation.doc
40/119
i +.+ Cu!!ent wa$e&%!, du!in Cl%sin %(e!ati%n %& C9
&%! /,ts GIS
3
-
5/19/2018 VFTO documentation.doc
41/119
i +./ T!ansient $%ltae wa$e&%!, du!in O(enin %(e!ati%n
%& C9 &%! /,ts GIS
51
-
5/19/2018 VFTO documentation.doc
42/119
i +.8 T!ansient $%ltae wa$e&%!, du!in Sec%nd Rest!i7e
&%! /,ts GIS
5/
-
5/19/2018 VFTO documentation.doc
43/119
+.#.# Sinle Phase e?ui$alent ci!cuit &%! 1=,ts lenth
GIS"
The circuit is divided into three sections of /mt, 5mt, and :mts
respectively from load
side and by using the below circuit shown figure 5.0. The
transients due to closing of the
circuit brea$er are calculated as shown in Fig 5.7. From this
graph, the pea$ voltage obtained
is 8.5: p.u at a rise time of 0/ns.
i +.: SinlePhase e?ui$alent ci!cuit &%! 1=,ts lenth GIS due
t% switchin %(e!ati%n
To introduce current chopping, the circuit brea$er is opened.
The transients are
obtained during opening operation is shown in Fig 5.. From the
graph, the ma(imum
voltage obtained is /.85 p.u at a rise time of 6:ns.
Assuming a second re"stri$e transients are calculated by closing
another switch at the
time ma(imum voltage difference occurs across the circuit
brea$er. The transient obtained
due to second re"stri$e is shown in Fig 5./1. From the graph,
the ma(imum voltage obtained
is 8.:/ p.u at a rise time of /80ns.
>%de %& O(e!ati%n>anitude %&
-%ltae 4(.u5
Rise Ti,e
4nan% sec%nds5
%uring #losing
2peration8.5: 0/
%uring 2pening
2peration/.85 6:
%uring Second
>e"Stri$e8.:/ /80
Ta)le +.+ T!ansients due t% switchin %(e!ati%ns &%! 1=,ts
lenth GIS
58
-
5/19/2018 VFTO documentation.doc
44/119
i +.; T!ansient $%ltae wa$e&%!, du!in Cl%sin %(e!ati%n
%& C9 &%! 1=,ts GIS
53
-
5/19/2018 VFTO documentation.doc
45/119
i +.< T!ansient $%ltae wa$e&%!, du!in O(enin %(e!ati%n
%& C9 &%! 1=,ts GIS
55
-
5/19/2018 VFTO documentation.doc
46/119
i +.1= T!ansient $%ltae wa$e&%!, du!in Sec%nd Rest!i7e
&%! 1=,ts GIS
5:
-
5/19/2018 VFTO documentation.doc
47/119
+.* T!ansients due t% aults"
+.*.1 GIS %& /,ts lenth t% calculate t!ansients due t%
&aults"
i +.11 GIS %& /,ts lenth t% calculate t!ansients due t%
&aults
The e*uivalent circuit of :mts length )IS is shown in figure
5.//. This circuit is
divided into three sections of /mt, /.:mt and 8.:mts lengths
respectively from the load side.
The transients are obtained without fault is shown in Fig 5./8.
From this graph, the ma(imum
voltage is obtained at 8.1 p.u at rise time of 5311ns.
Fast transient over voltages are generated not only due to
switching operations but
also due to single"line"to"ground faults. A fault at a
particular point is e*uivalent to a short"
circuit at that location. This situation can be simulated by
connecting a switch at a particular
point and closing it at the pea$ of the voltage.
P!%cedu!e &%! calculati%n %& t!ansients at
di&&e!ent distances"
Case 4i5" -%istance of 8.:mts
56
-
5/19/2018 VFTO documentation.doc
48/119
i +.1# T!ansient $%ltae wa$e&%!, &%! /,ts GIS with%ut
ault' with%ut 2%ad
50
-
5/19/2018 VFTO documentation.doc
49/119
i +.1* A &ault %ccu!s at a distance %& #./,ts lenth
&!%, l%ad side
The circuit is shown in above figure 5./3. From this circuit,
the ma(imum voltageacross the circuit brea$er can be found. The
transients that are obtained in this case is shown
in Fig 5./5. From this graph, the ma(imum voltage is obtained at
8.18p.u at a rise time of 83
ns.
Case 4ii5" -%istance of 5mts
i +.1/ A &ault %ccu!s at a distance %& +,ts lenth
&!%, l%ad side
The circuit is shown in above figure 5./:. From this circuit,
the ma(imum voltage
across the circuit brea$er can be found. The transients that are
obtained in this case is shown
in Fig 5./6. From this graph, the pea$ voltage is obtained at
8.7p.u at a rise time of :8 ns.
57
-
5/19/2018 VFTO documentation.doc
50/119
Case 4iii5" -%istance for :mts
i +.1: A &ault %ccu!s at a distance %& /,ts lenth
&!%, l%ad side
The circuit is shown in above figure 5./0. From this circuit,
the ma(imum voltage
across the circuit brea$er can be found. The transients that are
obtained in this case is shown
in Fig 5./7. From this graph, the pea$ voltage is obtained at
3./8 p.u at a rise time of /18 ns.
The magnitudes and rise times of :mts length )IS due to faults
are tabulated in the
Table 5.:.
S. N%Distance in
4,ts5
>anitude %&
-%ltae 4(.u5
Rise Ti,e
4nan% sec5
/ 1.1 1 1
8 8.: 8.18 83
3 5.1 8.7 :8
5 :.1 3./8 /18
Ta)le +./ T!ansients due t% &aults &%! /,ts lenth GIS
with%ut 2%ad
Form the above table, it is clear that as the length of the bus
bar between faulted point
and load is increasing, higher degree of oscillations are
obtained in the circuit.
5
-
5/19/2018 VFTO documentation.doc
51/119
i +.1+ T!ansient $%ltae wa$e&%!, at a distance %& #./,ts
&!%, l%ad side' &%! /,ts GIS' with%ut 2%ad
:1
-
5/19/2018 VFTO documentation.doc
52/119
i +.18 T!ansient $%ltae wa$e&%!, at a distance %& +,ts
&!%, l%ad side' &%! /,ts GIS' with%ut 2%ad
:/
-
5/19/2018 VFTO documentation.doc
53/119
i +.1; T!ansient $%ltae wa$e&%!, at a distance %& /,ts
&!%, l%ad side' &%! /,ts GIS' with%ut 2%ad
:8
-
5/19/2018 VFTO documentation.doc
54/119
+.*.# GIS %& 1=,ts lenth t% calculate t!ansients due t%
&aults"
i +.1< GIS %& 1=,ts lenth t% calculate t!ansients due t%
&aults
The e*uivalent circuit for /1mts length )IS is shown in above
figure 5./. The above
circuit is divided into three sections of :mt, 5mt and /mts
respectively from load side.
The transients are obtained without fault is shown in Fig 5.81.
From this graph, the
ma(imum voltage is obtained at /. p.u at a rise time of :701 ns.
The transients are
calculated at different distances by short circuiting at their
respective distances are given
below.
P!%cedu!e &%! calculati%n %& t!ansients at
di&&e!ent distances"
Case 4i5" -%istance of /mts
i +.#1 A &ault %ccu!s at a distance %& 1,ts lenth
&!%, l%ad side
:3
-
5/19/2018 VFTO documentation.doc
55/119
i +.#= T!ansient $%ltae wa$e&%!, &%! 1=,ts GIS with%ut
ault' with%ut 2%ad
:5
-
5/19/2018 VFTO documentation.doc
56/119
The circuit is shown in above figure 5.8/. From this circuit,
the ma(imum voltage
across the circuit brea$er can be found. The transients that are
obtained in this case is shown
in Fig 5.88. From this graph, the ma(imum voltage is obtained at
/.: p.u at a rise time of 78
ns.
Case 4ii5 -%istance of :mts
i +.#* A &ault %ccu!s at a distance %& /,ts lenth
&!%, l%ad side
The circuit is shown in above figure 5.83. From this circuit,
the ma(imum voltage
across the circuit brea$er can be found. The transients are
obtained in this case is shown in
Fig 5.85. From this graph, the ma(imum voltage is obtained at
8.17 p.u at a rise time of 75
ns.
Case 4iii5" -%istance of /1mts
i +.#/ A &ault %ccu!s at a distance %& 1=,ts lenth
&!%, l%ad side
The circuit is shown in above figure 5.8:. From this circuit,
the ma(imum voltage
across the circuit brea$er can be found. The transients that are
obtained in this case is shown
in Fig 5.86. From this graph, the ma(imum voltage is obtained at
8.65p.u at a rise time of
/88 ns.
::
-
5/19/2018 VFTO documentation.doc
57/119
i +.## T!ansient $%ltae wa$e&%!, at a distance %& 1,ts
&!%, l%ad side' &%! 1=,ts GIS' with%ut 2%ad
:6
-
5/19/2018 VFTO documentation.doc
58/119
i +.#+ T!ansient $%ltae wa$e&%!, at a distance %& /,ts
&!%, l%ad side' &%! 1=,ts GIS' with%ut 2%ad
:0
-
5/19/2018 VFTO documentation.doc
59/119
i +.#8 T!ansient $%ltae wa$e&%!, at a distance %& 1=,ts
&!%, l%ad side' &%! 1=,ts GIS' with%ut 2%ad
:7
-
5/19/2018 VFTO documentation.doc
60/119
The magnitudes and rise times of /1mts length )IS due to faults
are tabulated in the
Table 5.6.
S. N%Distance in
4,ts5
>anitude %&
-%ltae 4(.u5
Rise Ti,e
4nan% sec5
/ 1.1 1 1
8 /.1 /.: 78
3 :.1 8.17 75
5 /1 8.65 /88
Ta)le +.8 T!ansients due t% &aults &%! 1=,ts lenth GIS
with%ut 2%ad
+.*.* GIS %& /,ts lenth t% calculate t!ansients due t%
&aults with 2%ad"
In this analysis, it has been carried out by connecting a
transformer as oad. The load
is represented as a capacitance and short"circuit inductance
connected at the end of )IS.
i +.#: GIS %& /,ts lenth t% calculate t!ansients with 2%ad
due t% &ault
The e*uivalent circuit for :mts length )IS with load is shown in
above figure 5.80.
The transients are obtained without fault is shown in Fig 5.87.
From this graph, the ma(imum
voltage is obtained at /.3 p.u at a rise time of :178ns. The
transients are calculated at
different distances by short circuiting at their respective
distances are given below.
:
-
5/19/2018 VFTO documentation.doc
61/119
i +.#; T!ansient $%ltae wa$e&%!, with%ut &ault &%!
/,ts GIS' with 2%ad
61
-
5/19/2018 VFTO documentation.doc
62/119
P!%cedu!e &%! calculati%n %& t!ansients at
di&&e!ent distances"
Case 4i5" -%istance of 8.:mts
From the above circuit, the ma(imum voltage across the circuit
brea$er can be found.
The transients that are obtained in this case is shown in Fig
5.8. From this graph, the
ma(imum voltage is obtained at /.05 p.u at a rise time of 66
ns.
Case 4ii5" -%istance of 5mts
From the above circuit, the ma(imum voltage across the circuit
brea$er can be found.
The transients that are obtained in this case is shown in Fig
5.31. From this graph, the
ma(imum voltage is obtained at /.0: p.u at a rise time of 68
ns.
Case 4iii5" -%istance of :mts
From the above circuit, the ma(imum voltage across the circuit
brea$er can be found.
The transients that are obtained in this case is shown in Fig
5.3/. From this graph, the
ma(imum voltage is obtained at /.7/ p.u at a rise time of 67
ns.
The magnitudes and rise times of :mts length )IS due to faults
with load are
tabulated in the Table 5.0.
S. N%Distance in
4,ts5
>anitude %&
-%ltae 4(.u5
Rise Ti,e
4nan% sec5
/ 1.1 1 1
8 8.: /.05 66
3 5.1 /.0: 68
5 :.1 /.7/ 67
Ta)le +.: T!ansients due t% &aults &%! /,ts lenth GIS
with 2%ad
6/
-
5/19/2018 VFTO documentation.doc
63/119
i +.#< T!ansient $%ltae wa$e&%!, at a distance %&
#./,ts &!%, l%ad side' &%! /,ts GIS with 2%ad
68
-
5/19/2018 VFTO documentation.doc
64/119
i +.*= T!ansient $%ltae wa$e&%!, at a distance %& +,ts
&!%, l%ad side' &%! /,ts GIS with 2%ad
63
-
5/19/2018 VFTO documentation.doc
65/119
i +.*1 T!ansient $%ltae wa$e&%!, at a distance %& /,ts
&!%, l%ad side' &%! /,ts GIS with 2%ad
65
-
5/19/2018 VFTO documentation.doc
66/119
+.*.+ GIS %& 1=,ts lenth t% calculate t!ansients due t%
&aults with 2%ad"
i +.*# GIS %& 1=,ts lenth t% calculate t!ansients with 2%ad
due t% &ault
The e*uivalent circuit for /1mts length )IS with load is shown
in above figure 5.38.
The transients are obtained without fault is shown in Fig 5.33.
From this graph, the ma(imum
voltage is obtained at /.5 p.u at a rise time of 6188ns. The
transients are calculated at
different distances by short circuiting at their respective
distances are given below.
P!%cedu!e &%! calculati%n %& t!ansients at
di&&e!ent distances"
Case 4i5" -%istance of /mts
From the above circuit, the ma(imum voltage across the circuit
brea$er can be found.
The transients that are obtained in this case is shown in Fig
5.35. From this graph, the
ma(imum voltage is obtained at /.58 p.u at a rise time of //8
ns.
Case 4ii5" -%istance of :mts
From the above circuit, the ma(imum voltage across the circuit
brea$er can be found.
The transients that are obtained in this case is shown in Fig
5.3:. From this graph, the
ma(imum voltage is obtained at /.8 p.u at a rise time of /85
ns.
Case 4iii5" -%istance of /1mts
From the above circuit, the ma(imum voltage across the circuit
brea$er can be found.
The transients that are obtained in this case is shown in Fig
5.36. From this graph, the
ma(imum voltage is obtained at /.36 p.u at a rise time of /08
ns.
6:
-
5/19/2018 VFTO documentation.doc
67/119
i +.** T!ansient $%ltae wa$e&%!, with%ut &ault &%!
1=,ts GIS' with 2%ad
66
-
5/19/2018 VFTO documentation.doc
68/119
i +.*+ T!ansient $%ltae wa$e&%!, at a distance %& 1,ts
&!%, l%ad side' &%! 1=,ts GIS with 2%ad
60
-
5/19/2018 VFTO documentation.doc
69/119
i +.*/ T!ansient $%ltae wa$e&%!, at a distance %& /,ts
&!%, l%ad side' &%! 1=,ts GIS with 2%ad
67
-
5/19/2018 VFTO documentation.doc
70/119
i +.*8 T!ansient $%ltae wa$e&%!, at a distance %& 1=,ts
&!%, l%ad side' &%! 1=,ts GIS with 2%ad
6
-
5/19/2018 VFTO documentation.doc
71/119
The magnitudes and rise times of /1mts length )IS due to faults
with load are
tabulated in the Table 5.7.
S. N%Distance in
4,ts5
>anitude %&
-%ltae 4(.u5
Rise Ti,e
4nan% sec5
/ 1.1 1 1
8 /.1 /.58 //8
3 :.1 /.8 /85
5 /1.1 /.36 /08
Ta)le +.; T!ansients due t% &aults &%! 1=,ts lenth GIS
with 2%ad
+.+ C%nclusi%ns"
The transients due to switching operations and line to enclosure
faults with fi(ed arc
resistance for different lengths of )IS was made. Transients are
calculated along with load
also. It was observed that the transients obtained due to
switching operations and faults in
:mts length )IS will affect the system more than that obtained
in /1mts length )IS. It was
also found that during fault analysis, as the distance between
the fault point and load
increases the magnitudes and rise times of the transients also
increase. @hen load is
connected at the open end of )IS, the pea$ voltages and rise
times that are obtained due to
short"circuit do not follow a definite pattern.
01
-
5/19/2018 VFTO documentation.doc
72/119
CHAPTER/
TRANSIENTS DUE TO SWITCHING @ AU2TS WITH
-ARIA92E ARC RESISTANCE 93 USING TOEP2ERS
SPAR 2AW
/.1 Int!%ducti%n"
In previous chapter, transient over voltages calculated on the
basis of fi(ed arc
resistance have been presented. It is however, $nown that the
resistance of the spar$ channel
varies with current. At the instant of initiation of arc the
resistance is very high. As the
current in the arc increases the value of resistance starts
decreasing until it saturates at very
low value. In general, the arc resistance appears to be
inversely proportional to some function
of current.
Several authors have given arc resistance e*uations which can be
divided into two
groups as given below.
/. Inverse integral e*uation reported by Toepler et al. E/8
8. Inverse e(ponential e*uation reported by %emeni$ et al.
E/8
These e*uations were numerically evaluated for a given arc
current and then
normalized with the e(perimental arc resistance at t L 1.:Vs
-appro(imate time of ma(imum
current. 2f all these e*uations, one e*uation has been used for
the analysis in this thesis.
!ased on earlier studies in SF6gas, Toepler=s Spar$s aw is valid
for calculation of
variable arc resistance. The variable arc resistance due to
Toepler=s formulae E/8 is
calculated as given below.
> -t L+
t
1
1
T
-
?
dttiq
l
@here ?TL Toepler=s #onstant
L 1.11: volt.sec
-
5/19/2018 VFTO documentation.doc
73/119
L Spar$ ength in meters
*1 L Initial #harge
t L Spar$ #ollapse Time in sec.
The value of time varying spar$ resistance > -t is calculated
until it reaches a value
of / ohm. Initial charge *1 is an important parameter while
considering the non"uniform
fields. !ut the field between the dis"connector contacts is
almost uniform. Therefore, initial
charge *1is very small and can be neglected.
@hen a circuit brea$er operates a conducting spar$ channel is
established with time
lag of few nanoseconds after the brea$down channel is connected
the electrodes. %uring this
time only the spar$ resistance changes from a very large value
to very small value. For
homogeneous fields, this time is given by
tzL/3.3 Q1E
KT
@here &1L!rea$down field strength
L 7.6 Q /16volt
-
5/19/2018 VFTO documentation.doc
74/119
/.# T!ansients due t% switchin %(e!ati%n"
/.#.1 Sinle Phase e?ui$alent ci!cuit &%! /,ts lenth GIS"
i /.1 /,ts lenth GIS with -a!ia)le A!c !esistance due t%
switchin %(e!ati%n
Dsing the e*uivalent circuit of :mts length )IS given in Fig
:./, transients due to
closing operation of the circuit brea$er are calculated as given
in Fig :.8. From this graph,
the ma(imum voltage obtained is 3.37p.u with a rise time of
50ns. The difference between
ma(imum value for Fi(ed and 'ariable Arc >esistance is found
to be insignificant.
!y using the above circuit, the transients due to opening
operation of the circuit
brea$er is shown in Fig :.3. From this graph, the ma(imum
voltage obtained is /.37p.u at a
rise time of 3/ns. The difference between ma(imum value for
Fi(ed and 'ariable Arc
>esistance is found to be significant.
Assuming that there is a second re"stri$e, another switch is
connected in parallel to
the circuit brea$er for simulation in ;S;I#& modeling.
Transients are calculated by closing
this switch when voltage difference across the contacts of the
circuit brea$er reaches
ma(imum value. Transients calculated due to second re"stri$e
gives the pea$ voltage of
8.::p.u at a rise time of /3ns as shown in Fig :.5.
03
-
5/19/2018 VFTO documentation.doc
75/119
i /.# T!ansient $%ltae wa$e&%!, du!in Cl%sin %(e!ati%n
%& C9 &%! /,ts GIS' with -a!ia)le A!c Resistance
05
-
5/19/2018 VFTO documentation.doc
76/119
i /.* T!ansient $%ltae wa$e&%!, du!in O(enin %(e!ati%n
%& C9 &%! /,ts GIS' with -a!ia)le A!c Resistance
0:
-
5/19/2018 VFTO documentation.doc
77/119
i /.+ T!ansient $%ltae wa$e&%!, du!in Sec%nd Rest!i7e
&%! /,ts GIS' with -a!ia)le A!c Resistance
06
-
5/19/2018 VFTO documentation.doc
78/119
The magnitudes and rise times of :mts length )IS are tabulated
in the Table :.3.
>%de %& O(e!ati%n>anitude %& -%ltae
4(.u5
Rise Ti,e
4nan% sec5
%uring #losing
2peration3.37 50
%uring 2pening
2peration/.37 3/
%uring Second
>e"Stri$e8.:: /3
Ta)le /.1 T!ansients due t% switchin %(e!ati%n &%! /,ts
lenth GIS with -a!ia)le
A!c Resistance
/.#.# Sinle Phase e?ui$alent ci!cuit &%! 1=,ts lenth
GIS"
i /./ 1=,ts lenth GIS with -a!ia)le A!c !esistance due t%
switchin %(e!ati%ns
The e*uivalent circuit of /1mts length )IS is given in Fig :.:,
transients due to
closing operation of the circuit brea$er are calculated as given
in Fig :.6. From this graph,
the ma(imum voltage obtained is 8.30p.u with a rise time of
06ns.
!y using the above circuit, the transients due to opening
operation of the circuit
brea$er is shown in Fig :.0. From this graph, the ma(imum
voltage obtained is /.17p.u at a
rise time of 08ns.
00
-
5/19/2018 VFTO documentation.doc
79/119
i /.8 T!ansient $%ltae wa$e&%!, du!in Cl%sin %(e!ati%n
%& C9 &%! 1=,ts GIS' with -a!ia)le A!c Resistance
07
-
5/19/2018 VFTO documentation.doc
80/119
i /.: T!ansient $%ltae wa$e&%!, du!in O(enin %(e!ati%n
%& C9 &%! 1=,ts GIS' with -a!ia)le A!c Resistance
0
-
5/19/2018 VFTO documentation.doc
81/119
i /.; T!ansient $%ltae wa$e&%!, du!in Sec%nd Rest!i7e
&%! 1=,ts GIS' with -a!ia)le A!c Resistance
71
-
5/19/2018 VFTO documentation.doc
82/119
Assuming that there is a second re"stri$e, another switch is
connected in parallel to the
circuit brea$er for simulation in ;S;I#& modeling.
Transients calculated due to second re"
stri$e gives the pea$ voltage of /.76p.u at a rise time of :1ns
as shown in Fig :.7. The
magnitudes and rise times are tabulated as shown in below Table
:.8.
>%de %& O(e!ati%n>anitude %&
-%ltae 4(.u5
Rise Ti,e
4nan% sec5
%uring #losing
2peration8.30 06
%uring 2pening
2peration/.17 08
%uring Second
>e"Stri$e /.76 :1
Ta)le /.# T!ansients due t% switchin %(e!ati%n &%! 1=,ts
lenth GIS with -a!ia)le
A!c Resistance
/.* T!ansients due t% &ault"
/.*.1 GIS %& /,ts lenth t% calculate t!ansients due t%
&aults"
i /.< GIS %& /,ts lenth t% calculate t!ansients with
-a!ia)le A!c Resistance
due t% &aults
P!%cedu!e &%! calculati%n %& t!ansients at
di&&e!ent distances"
7/
-
5/19/2018 VFTO documentation.doc
83/119
Case 4i5" -%istance of 8.:mts
The e*uivalent circuit is shown in above figure :.. From this
circuit, the ma(imum
voltage across the circuit brea$er can be found. The transients
that are obtained in this case is
shown in Fig :./1. From this graph, the ma(imum voltage is
obtained at /.75 p.u at a rise
time of //8 ns.
Case 4ii5" -%istance of 5mts
From the above circuit shown in fig :., the ma(imum voltage
across the circuit
brea$er can be found. The transients that are obtained in this
case is shown in Fig :.//. From
this graph, the ma(imum voltage is obtained at 8.:8 p.u at a
rise time of 68 ns.
Case 4iii5" -%istance for :mts
From the above circuit shown in fig :., the ma(imum voltage
across the circuit
brea$er can be found. The transients that are obtained in this
case is shown in Fig :./8. From
this graph, the pea$ voltage is obtained at 8.78 p.u at a rise
time of :6 ns.
The magnitudes and rise times of :mts length )IS due to faults
with variable arc
resistance are tabulated in the Table :.3.
S. N%Distance in
4,ts5
>anitude %&
-%ltae 4(.u5
Rise Ti,e
4nan% sec5
/ 1.1 1 1
8 8.: /.75 //8
3 5.1 8.:8 68
5 :.1 8.78 :6
Ta)le /.* T!ansients due t% &aults &%! /,ts lenth GIS
with $a!ia)le a!c !esistance'
with%ut 2%ad
78
-
5/19/2018 VFTO documentation.doc
84/119
i /.1= T!ansient $%ltae wa$e&%!, at a distance %& #./,ts
&!%, l%ad side &%! /,ts GIS' with%ut 2%ad
73
-
5/19/2018 VFTO documentation.doc
85/119
i /.11 T!ansient $%ltae wa$e&%!, at a distance %& +,ts
&!%, l%ad side &%! /,ts GIS' with%ut 2%ad
75
-
5/19/2018 VFTO documentation.doc
86/119
i /.1# T!ansient $%ltae wa$e&%!, at a distance %& /,ts
&!%, l%ad side &%! /,ts GIS' with%ut 2%ad
7:
-
5/19/2018 VFTO documentation.doc
87/119
/.*.# GIS %& 1=,ts lenth t% calculate t!ansients due t%
&aults"
The e*uivalent circuit of /1mts length )IS with variable arc
resistance is shown in
below figure :./3.
i /.1* GIS %& 1=,ts lenth t% calculate t!ansients with
-a!ia)le A!c Resistance
due t% &ault
P!%cedu!e &%! calculati%n %& t!ansients at
di&&e!ent distances"
Case 4i5" -%istance of /mts
From this circuit, the ma(imum voltage across the circuit
brea$er can be found. The
transients that are obtained in this case is shown in Fig :./5.
From this graph, the ma(imum
voltage is obtained at /.73 p.u at a rise time of / ns.
Case 4ii5" -%istance of :mts
From the above circuit, the ma(imum voltage across the circuit
brea$er can be found.
The transients that are obtained in this case is shown in Fig
:./:. From this graph, the
ma(imum voltage is obtained at 8.5: p.u at a rise time of /33
ns.
76
-
5/19/2018 VFTO documentation.doc
88/119
Case 4iii5" -%istance for /1mts
From the above circuit, the ma(imum voltage across the circuit
brea$er can be found.
The transients that are obtained in this case is shown in Fig
:./6. From this graph, the pea$
voltage is obtained at 8.73 p.u at a rise time of /38 ns.
70
-
5/19/2018 VFTO documentation.doc
89/119
i /.1+ T!ansient $%ltae wa$e&%!, at a distance %& 1,t
&!%, l%ad side &%! 1=,ts GIS' with%ut 2%ad
77
-
5/19/2018 VFTO documentation.doc
90/119
i /.1/ T!ansient $%ltae wa$e&%!, at a distance %& /,ts
&!%, l%ad side &%! 1=,ts GIS' with%ut 2%ad
7
-
5/19/2018 VFTO documentation.doc
91/119
i /.18 T!ansient $%ltae wa$e&%!, at a distance %& 1=,ts
&!%, l%ad side &%! 1=,ts GIS' with%ut 2%ad
1
-
5/19/2018 VFTO documentation.doc
92/119
The magnitudes and rise times of /1mts length )IS due to faults
with variable arc
resistance are tabulated in the Table :.5.
S. N%Distance in
4,ts5
>anitude %&
-%ltae 4(.u5
Rise Ti,e
4nan% sec5
/ 1.1 1 1
8 /.1 /.73 /
3 :.1 8.5: /33
5 /1 8.73 /38
Ta)le /.+ T!ansients due t% &aults &%! 1=,ts lenth GIS
with $a!ia)le a!c !esistance'
with%ut 2%ad
/.*.* GIS %& /,ts lenth t% calculate t!ansients due t%
&aults with 2%ad"
i /.1: GIS %& /,ts lenth t% calculate t!ansients with
-a!ia)le A!c Resistance
due t% &ault
The e*uivalent circuit for :mts length )IS with load is shown in
above figure :./0.
The transients are calculated at different distances by short
circuiting at their respective
distances are given below.
P!%cedu!e &%! calculati%n %& t!ansients at
di&&e!ent distances"
/
-
5/19/2018 VFTO documentation.doc
93/119
Case 4i5" -%istance of 8.:mts
From the above circuit, the ma(imum voltage across the circuit
brea$er can be found.
The transients that are obtained in this case is shown in Fig
:./7. From this graph, the
ma(imum voltage is obtained at /.51 p.u at a rise time of 03
ns.
Case 4ii5" -%istance of 5mts
From the above circuit, the ma(imum voltage across the circuit
brea$er can be found.
The transients that are obtained in this case is shown in Fig
:./. From this graph, the
ma(imum voltage is obtained at /.83 p.u at a rise time of 63
ns.
Case 4iii5" -%istance of :mts
From the above circuit, the ma(imum voltage across the circuit
brea$er can be found.
The transients that are obtained in this case is shown in Fig
:.81. From this graph, the
ma(imum voltage is obtained at /.35 p.u at a rise time of 0
ns.
The magnitudes and rise times of :mts length )IS due to faults
with load are
tabulated in the Table :.:.
S. N%Distance in
4,ts5
>anitude %&
-%ltae 4(.u5
Rise Ti,e
4nan% sec5
/ 1.1 1 1
8 8.: /.51 03
3 5.1 /.83 63
5 :.1 /.35 0
Ta)le /./ T!ansients due t% &aults &%! /,ts lenth GIS
with $a!ia)le a!c !esistance'
with 2%ad
8
-
5/19/2018 VFTO documentation.doc
94/119
i /.1; T!ansient $%ltae wa$e&%!, at a distance %& #./,ts
&!%, l%ad side &%! /,ts GIS' with 2%ad
3
-
5/19/2018 VFTO documentation.doc
95/119
i /.1< T!ansient $%ltae wa$e&%!, at a distance %&
+,ts &!%, l%ad side &%! /,ts GIS' with 2%ad
5
-
5/19/2018 VFTO documentation.doc
96/119
i /.#= T!ansient $%ltae wa$e&%!, at a distance %& /,ts
&!%, l%ad side &%! /,ts GIS' with 2%ad
:
-
5/19/2018 VFTO documentation.doc
97/119
/.*.+ GIS %& 1=,ts lenth t% calculate t!ansients due t%
&aults with 2%ad"
i /.#1 GIS %& 1=,ts lenth t% calculate t!ansients with
-a!ia)le A!c Resistance
due t% &ault
The e*uivalent circuit for /1mts length )IS with load is shown
in above figure :.8/.
The transients are calculated at different distances by short
circuiting at their respective
distances are given below.
P!%cedu!e &%! calculati%n %& t!ansients at
di&&e!ent distances"
Case 4i5" -%istance of /mts
From the above circuit, the ma(imum voltage across the circuit
brea$er can be found.
The transients that are obtained in this case is shown in Fig
:.88. From this graph, the
ma(imum voltage is obtained at /.51 p.u at a rise time of //3
ns.
Case 4ii5" -%istance of :mts
From the above circuit, the ma(imum voltage across the circuit
brea$er can be found.
The transients that are obtained in this case is shown in Fig
:.83. From this graph, the
ma(imum voltage is obtained at /.8 p.u at a rise time of /85
ns.
Case 4iii5" -%istance of /1mts
From the above circuit, the ma(imum voltage across the circuit
brea$er can be found.
The transients that are obtained in this case is shown in Fig
:.85. From this graph, the
ma(imum voltage is obtained at /.58 p.u at a rise time of /01
ns.
6
-
5/19/2018 VFTO documentation.doc
98/119
i /.## T!ansient $%ltae wa$e&%!, at a distance %& 1,t
&!%, l%ad side &%! 1=,ts GIS' with 2%ad
0
-
5/19/2018 VFTO documentation.doc
99/119
i /.#* T!ansient $%ltae wa$e&%!, at a distance %& /,ts
&!%, l%ad side &%! 1=,ts GIS' with 2%ad
7
-
5/19/2018 VFTO documentation.doc
100/119
i /.#+ T!ansient $%ltae wa$e&%!, at a distance %& 1=,ts
&!%, l%ad side &%! 1=,ts GIS' with 2%ad
-
5/19/2018 VFTO documentation.doc
101/119
The magnitudes and rise times of /1mts length )IS due to faults
with load are
tabulated in the Table :.6.
S. N%
Distance in
4,ts5
>anitude %& -%ltae
4(.u5
Rise Ti,e
4nan% sec5
/ 1.1 1 1
8 /.1 /.51 //3
3 :.1 /.8 /85
5 /1.1 /.58 /01
Ta)le /.8 T!ansients due t% &aults &%! 1=,ts lenth GIS
with $a!ia)le a!c !esistance'
with 2%ad
/.+ C%nclusi%ns"
The variable arc resistance is calculated by Toepler=s formulae.
Transients are
calculated due to switching operations and faults with variable
arc resistance along with load.
For any length of )IS it was found that transients due to
variable arc resistance give lower
value of pea$ voltages than that obtained with fi(ed arc
resistance. @hen load is connected at
the open end of )IS, the pea$ voltages that are obtained due to
faults do not follow a definite
pattern.
/11
-
5/19/2018 VFTO documentation.doc
102/119
CHAPTER8
SUPPRESSION O O-ER-O2TAGESAND CO>PARISIONS
8.1 Int!%ducti%n"
The fast transient over voltages during switching operation and
faults can cause
damage to the system e*uipment. ence it is advisable to suppress
these over voltages for
protection of e*uipments. 2ne of the methods of suppressing
these over voltages is by
insertion of resistance during switching. )enerally a
>esistor of :11 is used for this
purpose E/1.
In this analysis, a resistor of :11 is connected in parallel
with the circuit brea$er and
a switch is connected in series with the resistor. The transient
over voltages are suppressed
only if the current during contact operation flows through the
resistor. The switch connected
in series with the resistor is closed at the time ma(imum