1 CAUSES OF OVERVOLTAGES

Overvoltages (or surges) on power systems are due to various causes. The normal
operating voltages of the system do not stress the insulation severely. But the voltage
stresses due to overvoltages can be so high that they may become dangerous to both
the lines as well as the connected equipment and may cause damage, unless some
protective measures against these overvoltages are taken. Overvoltages arising on a
power system can be generally classified into two main categories as follows.
External Overvoltages These overvoltages originate from atmospheric distur-
bances, mainly due to lightning. These overvoltages take the form of a unidirectional
impulse (or surge) whose maximum possible amplitude has no direct relationship
with the operating voltage of the system. They may be due to any of the following
causes.
(i) Direct lightning strokes
(ii) Electromagnetically induced overvoltages due to lightning discharge taking
place near the line (commonly known as ‘side stroke’)
(iii) Voltages induced due to changing atmospheric conditions along the line
length
(iv) Electrostatically induced overvoltages due to the presence of charge clouds
nearby
(v) Electrostatically induced overvoltages due to the frictional effects of small
particles such as dust or dry snow in the atmosphere or due to change in the
altitude of the line
Internal Overvoltages These overvoltages are caused by changes in the operat-
ing conditions of the network. Internal overvoltages can further be divided into two
groups as follows.
(i) Switching overvoltages (or transient overvoltages of high frequency)
These overvoltages are caused by the transient phenomena which appear when
the state of the network is changed by a switching operation or fault condition.
These overvoltages are generally oscillatory and take the form of a damped
sinusoid. The frequency of these overvoltages may vary from a few hundred
Hz to a few kHz, and it is governed by the inherent capacitances and
inductances of the circuit. For example, switching on and off of equipment,
such as switching of high voltage reactors and switching of a transformer at
 no load causes overvoltages of transient nature. Other examples of this type
of overvoltage are—if a fault occurs on any phase, the voltage with respect to
ground of the other two healthy phases can exceed the normal value until the
fault is cleared; when the contacts of a circuit breaker open in order to clear
the fault, the final extinction of current is followed by the appearance of a
restriking voltage across the contacts whose characteristics are a high
amplitude which may reach twice that of the system voltage, and a relatively
high frequency.
(ii) Temporary overvoltages (or steady-state overvoltages of power frequency)
These overvoltages are the steady-state voltages of power-system frequency
which may result from the disconnection of load, particularly in the case of
long transmission lines.
Transient overvoltages arising on the power system are assessed by an over-
voltage factor which is defined as the ratio of the peak overvoltage to the rated
peak system-frequency phase voltage. This ratio is also referred to as the
amplitude factor.
The examination of overvoltages on the power system includes a study of their
magnitudes, shapes, duration, and frequency of occurrence. This study should
be performed not only at the point where an overvoltage originates but also at
all other points along the transmission network through which the surges may
travel. It is also essential to know the causes and effects of these overvoltages,
so that suitable protective measures can be taken.

2 PHENOMENA
The discharge of the charged cloud to the ground is called lightning phenomenon. A
lightning discharge through air occurs when a cloud is raised to such a high potential
with respect to the ground (or to a nearby cloud) that the air breaks down and the
insulating property of the surrounding air is destroyed. This raising of potential is
caused by frictional effects due to atmospheric disturbances (e.g., thunderstorms)
acting on the particles forming the cloud. The cloud and the ground form two plates
of a gigantic capacitor whose dielectric medium is air. During thunderstorms, posi-
tive and negative charges are separated by the movement of air currents forming ice
crystals in the upper layer of cloud and rain in the lower part. The cloud becomes
negatively charged and has a larger layer of positive charge at its top. As the separa-
tion of charge proceeds in the cloud, the potential difference between concentrations
of charges increases and the vertical electric field along the cloud also increases. The
total potential difference between the two main charge centres may vary from 100 to
1000 MV. Only a part of the total charge is released to the ground by lightning, and
the rest is consumed in inter-cloud discharges.
As the lower part of the cloud is negatively charged, the ground gets positively
charged by induction. Lighting discharge requires breakdown of the dielectric
medium, i.e. air between the negatively charged cloud and the positively charged
ground. Though the electric field (potential gradient) required for the breakdown of
air at STP is 30 kV/ cm peak in a cloud where the moisture content in the air is large
and also because of the high altitude (i.e., lower pressure) the electric field required
for breakdown of air is only about 10 kV/cm. The lightning discharge mechanism is
well explained with the help of Fig. 1.

+ +
+ + + +
+ + + + + +
+ +

Stepped +
Leader + ++
Pilot Return + +
Streamer Streamer +
+++++++++++++++ ++++++++++++++

(a) (b)

+
+ +
+ + +
+

Dart
Leader

Return
Streamer +
+
+
+
+ + + + + + + + + + + + + ++

(c)

Fig. 1 Lightning mechanism

When a potential gradient of about 10 kV/cm is set up in the cloud, the air
immediately surrounding the cloud gets ionised and the first process of the actual
lightning discharge starts. At this instant, a streamer called a ‘pilot streamer’ starts
from the cloud towards the ground which is not visible. The current associated with
this streamer is of the order of 100 amperes, and the most frequent velocity of propa-
gation of the streamer is about 0.15 m/m s (i.e. 0.05 per cent of the velocity of light).
Depending upon the state of ionisation of the air surrounding the pilot streamer, it is
branched into several paths, a stepped leader is formed, as shown in Fig. 1(a). It is
called a stepped leader because of its zig-zag shape. It contains a series of steps about
50 m in length. The velocity of propagation of these steps should be more than 16.6
per cent that of light. A portion of the charge in the centre from which the strike
originated is lowered and distributed over this entire system of temporary conductors.
This process continues until one of the leaders strikes the ground. When
one of the stepped leaders strikes the ground, an extremely bright return streamer, as
shown in Fig. 1(b) propagates upward from the ground to the cloud following the
same path as the main channel of the downward leader. The charge distributed along
the leaders is thus discharged progressively to the ground, giving rise to the very
large currents associated with lightning discharges. The conventional directions of
the current in the stepped leader and return streamer are the same. The current varies
between 1 kA and 200 kA and the velocity of propagation of the streamer is about 10
per cent that of light. It is here that the negative charge of the cloud is being neutral-
ized by the positive induced charge on the ground. It is this instant which gives rise
to the lightning flash which is visible with our naked eye.
After the neutralization of the most of the negative charge on the cloud any further
discharge from the cloud may have to originate from another charge centre within
the cloud near the already neutralized charge centre. Such a discharge from another
charge centre will, however, make use of the already ionised path, and consequently
it will have a single branch and will be associated with a high current. This streamer
of discharge is called a dart leader as shown in Fig. 1(c). The velocity of propaga- tion
of the dart leader is about 3 per cent that of light. The dart leader can cause more severe
damage than the return stroke.
Though the discharge current in the return streamer is relatively large but since it
continues only for a few microseconds, it contains less energy and hence this
streamer is called a cold lightning stroke. The dart leader is called a hot lightning
stroke because even though the current in this leader is relatively smaller, it contains
relatively more energy since it continues for some milliseconds.
It has been observed that every thundercloud may consist of as many as 40
separate charged centres due to which a heavy lightning stroke may occur. This
lightning stroke is called multiple (or repetitive) stroke.
When a lightning stroke takes place to a point adjacent to the line, the current asso-
ciated with it induces a voltage on the line which is in the form of a travelling wave.
A lightning discharge may have currents in the range of 10 kA to about 100 kA, and
the duration of the lightning flash may be of the order of a few microseconds. The
current wave form is generally a unidirectional pulse which rises to a peak value in
about 3 m s and decays to a small values in several tens of microseconds. Therefore,
the overvoltage due to lightning will have travelling waves of very steep wave front
which may be dangerous to the line and equipment of the power system and may
cause damage.

3 WAVE SHAPE OF VOLTAGE DUE TO LIGHTNING

Lightning sets up steep-fronted, unidirectional voltage waves which can be
represented as the difference of two exponentials. Thus
v = V (e–at – e–bt) (1)
where a and b are constants which determine the shape and v is magnitude of the
steep voltage and V is equal to the crest (peak) value of the impulse voltage wave.
The steep is dependent on whether the surge is induced or is the result of a direct
stroke. The wave shape of this type is shown in Fig. 2.
 v
The wave shape is generally
V
defined in terms of the times t1 and
t2 in microseconds. Where t1 is the
time taken by the voltage wave- 0.5 V
front to reach its peak value, and t2
is the time taken for the tail to fall
to 50 per cent of the peak value.
Both times are measured from the 0
virtual zero time point (i.e., from t1 t2 t
the start of the wave). The wave is

Fig. 2 Wave shape of voltage due to lightning

then referred to as1t 2/t wave. The
standard wave chosen for the testing purpose is a 1/50 wave which implies that t1 is
1 ms and t2 is 50 ms.

4 OVERVOLTAGES DUE TO LIGHTNING

Lightning causes two kinds of voltage surges (overvoltage), one by direct stroke to a
line conductor, and the other induced by indirect stroke when bound charges are dis-
sipated following a lightning discharge to an object near the line conductor. A direct
stroke to a phase (line) conductor is the most severe lightning stroke as it produces
the highest overvoltage for a given stroke current. The current in a direct stroke to a
phase conductor is hardly affected by the voltage which it sets up between the
conductor and ground, so that a direct lightning stroke approximates to a constant
current source. If a current I flows in a direct stroke to a phase conductor, forward
and backward travelling currents if and ib, each of magnitude I/2 flow in the conduc-
tor away from the point of strike in both directions of the line as shown in Fig. 3.
Therefore, the voltage surge magnitude at the striking point is given by
I Z0
V= (2)
2
where Z0 is the surge impedance of the line.

Cloud
Direct Lightning Stroke

Phase Conductor I
ib if

I2 I2

V
Z0 Z0
Towers

(a) (b)

Fig. 3 Development of lightning overvoltage

The lightning current magnitude (I) is rarely less than 10 kA. Therefore, for a typi-
cal overhead line having a surge impedance Z0 of 400 W, the lightning surge voltage
will probably have a magnitude in excess of 2000 kV. Any overvoltage surge appear-
ing on a transmission line due to internal or external disturbances propagates in the
form of a travelling wave towards the ends of the line. The lightning surge is rapidly
attenuated, partly due to corona as it travels along the line. In this way, corona acts as
a natural safety valve to some extent.
When a tower of the transmission system is directly struck by lightning, the resis-
tance offered to the lightning current is that of the tower footing and any ground rods
or counterpoise wires in parallel. In calculation of overvoltage developed due to
lightning in this case, certain assumptions must be made concerning several of the
factors such as wave shape and magnitude of lightning current, surge impedance of
lightning stroke channel, shape of potential wave at tower top, effect of surges on
tower and footing impedance, etc. All these assumptions can be thrown into the
lightning current and tower footing resistance, thus giving
V = RI (3)
where, V = voltage across insulation at tower, R = tower footing resistance, and I =
lightning current in tower.
For very high towers, such as the ones at river crossings, which are subjected to
direct strokes of very high rate of change of current, the tower inductance may be
taken into account. The voltage (v) which may be sufficient to cause insulator flash-
over is given by
di
v=L (4)
dt

where, L = tower inductance, and di/dt = rate of change of current due to direct
stroke.
In the case of an indirect stroke, a negative charge in a cloud base causes bound
positive charges on the conductors of a nearby transmission line, and discharge of the
cloud due to a lightning stroke in the neighbourhood causes the conductor potential
to rise to v since high insulation resistance prevents rapid dissipation of the charge.
The peak value of the induced voltage surge is given by
v = Eh (5)
where, E is the mean electric field near the ground (earth) surface under a thunder-
cloud which may vary from 0.5 kV/cm to 2.8 kV/cm, and h is the height of the phase
conductor above ground.
If if be the forward travelling wave or the current and ib be the backward travelling
wave of the current, then in a case where the total conductor current i = 0, we have
if = ib and Vf = Vb = v/2. So that forward and backward travelling waves of voltage
and current set-off in opposite directions along the transmission line from the place
nearest to the cloud discharge.
The amplitudes of voltages induced on lines indirectly by lightning strokes on a
tower, ground wire or nearby ground or object, are however normally much less than
those caused by direct strokes to lines. Such strokes, however, are of real concern on
low voltage lines (33 kV and below) supported on small insulators. They are of little
importance on high-voltage lines whose insulators can withstand hundreds of
kilovolts without a flashover.
 5 KLYDONOGRAPH AND MAGNETIC LINK
5.1 Klydonograph Line

The klydonograph is an instrument for the

measurement of surge voltage on transmis- Electrode
sion lines caused by lightning. It measures Film
voltage by means of Lichtenberg figures, Homogeneous

when suitably coupled to the line whose Insulating Material

Metal Plate
surge voltage is to be measured. The kly-
donograph contains a rounded electrode Ground
connected to the line whose surge voltage

is to be measured. The electrode rests on Fig. 4 Klydonograph

the emulsion side of a photographic film or plate, which in turn rests on the smooth
surface of an insulating plate made of homogeneous insulating material, backed by a
metal plate electrode as shown in Fig. 4.
The photographic plate or the film is turned or moved by a clockwork mechanism
for bringing in the element of time. Three assemblies are generally placed in the same
box, for simultaneously measuring the voltages on the three phases of a trans-
mission line.
With this arrangement, a positive Lichtenberg figure is produced by a positive
surge, and a negative Lichtenberg figure by a negative surge, as illustrated in Fig.5.
Positive Lichtenberg figures are found to be superior to the negative ones for voltage
measurement purpose, since they are much larger than the negative figures for the
same voltage, as shown in Fig. 5. Diameter of positive Lichtenberg figure is a
+ Voltage

E KV
Time

Time
E KV

A Positive Figure A Negative Figure

Produced by a Positive Produced by a Negative
Surge Surge

Fig. 5 Positive and negative Lichtenberg igures produces by positive and

negative surge voltages of same magnitude and wave shape
function of the maximum value of the impressed voltage. The shape and configura-
tion of the figure depend on the wave shape of the impressed voltage.
5.2 Magnetic Link
The magnetic link is an instrument for the measurement of surge currents due to
lightning. It contains a small bundle of laminations made of cobalt-steel inserted in
a cylindrical molded plastic container, with the open end sealed in. It is placed in an
unmagnetised state in the vicinity of the conductor whose surge current is desired to
be measured.
When the current flows through the conductor, a magnetic field is set up around it
which is proportional to the current. The magnetising force of this field magnetizes
the magnetic link which is placed in the vicinity of the conductor. After the current
has passed, the link is left with a residual magnetism which is a function of the mag-
nitude of the current producing it for unidirectional surges. The residual magnetism
may be measured by a suitable instrument, and the magnitude of the current which
produces it may be determined by an experimentally obtained calibration curve of
current versus residual magnetism, or deflection on the instruments.
The principle of operation of the magnetic link is based on the assumption that
the surge current is unidirectional in polarity and that surge currents of any duration,
no matter how short, leave the same residual magnetism as a direct current of the
same magnitude. The magnetic link is extensively used to measure currents in direct
strokes, in transmission-line tower legs, ground wires, phase conductors, in counter-
poise conductors and in ground leads of arresters.
Magnetic links are generally placed in brackets fastened to the conductor so that
their axes coincide with the normal direction of the magnetic lines of force.

6 PROTECTION OF TRANSMISSION LINES

Direct lightning strokes are of concern on lines of all voltage classes, as the voltage
that may be set up is in most instances limited by the flashover of the path to ground.
Increasing the length of insulator strings merely permits a higher voltage before
flashover occurs. Therefore protective methods must be adopted to avoid flashover or
breakdown of line insulators due to overvoltage caused by direct lightning strokes.
The most generally accepted and effective method of protecting lines against
direct strokes is by the use of overhead ground wires. This method of protection is
known as shielding method which does not allow an arc path to form between the
line conductor and ground.
Ground wires are conductors running parallel to the main conductors of the trans-
mission line, supported on the same towers (or supports) and adequately grounded at
every tower or support. They are made of galvanised steel wires or ACSR conduc-
tors. They are provided to shield the lines against direct strokes by attracting the
lightning strokes to themselves rather than allowing them to strike the lines (phase
conductors). When a ground wire is struck by direct lightning stroke, the impedance
through which the current flows is very much reduced and a correspondingly higher
current is required to cause flashover.
 Overhead ground wires protect the lines by intercepting the direct strokes, keeping
them off the phase conductors (lines), and by providing multiple paths for conduct-
ing the strokes current to ground. The multiple paths for the stroke current result in
a reduced voltage drop. When the ground wire is struck in midspan, the current gets
divided and flows towards both towers. At the tower, the current again gets divided
between the tower and the outgoing ground wire. When the tower is struck by a direct
lightning stroke and there is only one overhead ground wire, the current gets divided
into three-path, one through the tower and two through the branches of the ground
wire. In addition to the above functions, ground wires also reduce the volt- ages
induced on the conductors from nearby strokes by increasing the capacitance
between conductors and ground.
In order to provide efficient protection to lines against direct strokes, the ground
wires must satisfy the following requirements.
(i) There should be an adequate clearance between the line conductors and the
ground or the tower structure.
(ii) There should be an adequate clearance between the ground wires and line
conductors, especially at the midspan, in order to prevent flashover to the line
conductors up to the protective voltage level used for the line design.
(iii) The tower footing resistance should be as low as economically justifiable.
The ground wires should be placed at such a height above the conductors and so
located that they should be well out on towers and should not be exactly over the
conductors in order to avoid any possibility of a short circuit occurring in the event
of conductors swinging under ice loading, etc.
A few terms relating to protection of transmission lines are explained below.
Protective Ratio
The protective ratio is defined as the ratio of the induced voltage on a conductor with
ground wire protection to the induced voltage which would exist on the conductor
without ground wire protection. The induced voltage under discussion is the induced
voltage between phase conductor and ground due to nearby stroke.
Protective Angle
The protective angle (a) afforded by a ground wire is defined as the angle between a
vertical line through the ground wire and a slant-
Ground
ing line connecting the ground wire and the phase Wire
conductor to be protected. This angle is shown in
Fig. 6. a
The protective angle (a) is in the region of 20°
to 45°. Experience with lines and also tests have
shown that when the protective angle does not
exceed 30° (on level spans), good shielding of the
line conductors is obtained and the probability of Phase Conductor
direct stroke to the conductor is also reduced.

But a protective angle of 45° is satisfactory when Fig. 6 Protective angle a of

an overhead ground
the tower is on a hill-side. The angle should be
wire
decreased by the angle of the slope of the hill.
Protective Zone
The protective zone is defined as the volume between the base plane cbc and the
slanting planes ac, extending from the ground wire to the plane of the conductors.
Figure 7(a) shows the cross-section of this volume. The plane ac cuts the base plane
at c, at a distance ky from the point b, vertically under the apex a. The ground wire is
at a height ab = y above the base plane. The ratio ky/y = k is called the protec- tive
ratio. The protective zone is sometimes called the protective wedge or protective shed.
a
Ground Wire a

y
Phase Mast y
c conductor

b ky c c b ky c

(a) (b)

Fig. 7 (a) Zone of protection of an overhead ground wire

(b) Cone of protection of a rod or mast

If the line ab in Fig. 7(b) represents a rod or mast with a height y, then it is said to
be protective cone around the mast. The radius on the base line is equal to ky and the
protective ratio is ky/y = k.
Height of Ground Wire
For protection against direct strokes, the ground wire should be located at a height at
least 10 per cent greater than y, calculated from the following equation.

÷( ) ( ) ÷( ) ( )
X y y 2 h h 2
= 2 – – 2 – (6)
H H H H H
where, X = horizontal spacing between the conductor and ground wire, H = height of
cloud, y = height of ground wire, and h = height of conductor.
Though recommendations about the protective angle and protective ratio vary,
experience has shown that double-circuit lines will be completely shielded if two
ground wires arc used, placed one above each outermost conductor, at a height above
the top conductor equal to the vertical spacing between conductors. For horizontally
arranged single-circuit lines, two overhead ground wires will give compete shielding
if placed above the plane of the conductors at a height about two-thirds the spacing
between conductors. The distance between the ground wires should be about equal
to the spacing of conductors.
Coupling Factor
Since a voltage of the same polarity as the ground wire voltage has induced on the
conductors, the entire ground wire or tower top voltage does not appear across the
line insulation. The ratio of this induced voltage on the conductor to the ground wire
voltage is known as the coupling factor. Therefore, the voltage across the line
insulation is the ground wire voltage times the coupling factor, neglecting the normal
frequency voltage.
The electromagnetic and electrostatic coupling factors are equal unless the effec-
tive radius of the ground wire is increased by a corona, in which case, the electro-
static coupling increases and the electromagnetic coupling remains unaffected. The
coupling factor with the effect of corona considered is calculated by the following
equation.
Coupling factor (C) = ÷electrostatic coupling × electromagnetic coupling
Coupling factor is calculated by the equation,
log b/a
C= (7)
log 2h/r

where, C = coupling factor, a = distance from conductor to ground wire, b = distance

from conductor to image of ground wire, h = height of ground wire above ground and
r = radius of ground wire.
The electromagnetic coupling factor is calculated using the actual radius of the
ground wire and the electrostatic coupling factor is calculated by the same equation
using the effective radius of the ground wire due to corona.
Reduction of Tower Footing Resistance
The effectiveness of a ground wire in controlling the lightning overvoltage on a line
to reasonable values depends to a great extent on low tower footing resistance. The
tower-footing resistance should be in the region 10-20 W, preferably closer to 10 W.
For medium voltage lines it may be even less than 10 W. Therefore, sometimes some
special means are employed to reduce the tower footing resistance, since by itself the
tower footing may not be able to give a low resistance. The tower footing resistance
may be reduced by driving rods near the tower and connecting them to the tower base
or by burying conductors (counterpoise wires) in the ground and con- necting them to
the tower base. Driven rods are usually of galvanised iron, copper weld, or copper-
bearing steel. Sometimes driven galvanised iron pipe is also used. Counterpoise wires
are usually of copper, aluminium, galvanised steel, stranded cable, or galvanized
steel strip.
The method to be used for reducing the tower footing resistance either by driving
rods or by counterpoise wires depends upon the soil conditions. Either method is
satisfactory if the proper low resistance is obtained. Deep rods may be driven only in
soil free from rocks. Deep-driven rods can be successfully used in the sandy soil.
As regards counterpoise wires, the hand-trenching method may be used for small
amounts of wire. However, if the amount of wire is considerable, it is usually installed
by feeding into a trench. The trench is shallow, usually 30 to 60 cm deep, and does
not require extensive filling.
Counterpoise wires are used extensively because of the difficulty encountered in
driving rods. They can be arranged either radially (non-continuous) or tower-to-
tower (continuous). Typical arrangements of radial counterpoise wires are shown in
Fig. 8 and tower-to-tower counterpoise wires are shown in Fig. 9.
 Base of Tower

Counterpoise
Wire Direction of Line
(a) (b)

(c) (d)

(e)

Fig. 8 Typical arrangements of radial (non-continuous) counterpoise wires

Continuous counterpoise wires are more effective than non-continuous counter-

poise wires. Current is drawn into the counterpoise from 1.5 to 2.5 km away from the
point of the stroke and fed into the stroke by means of the counterpoise wire and the
overhead ground wires. The concentration of the current at any tower is thus reduced.
The reduced current at any tower causes reduced voltage drop at the tower base, and
thus the tower potential is held to a value below the flashover voltage of the line
insulators.

7 PROTECTION OF STATIONS AND SUB-STATIONS

FROM DIRECT STROKES
The station structure itself is commonly built of steel members on which lines ter-
minate and which support the buses, isolating switches, etc. The power transformers
and circuit breakers usually rest on concrete pads on the ground.
The task of protection of such stations can be resolved into two parts. The first
being the protection of stations (including sub-stations) from direct lightning strokes,
and the second is the protection of electrical equipment of stations from travelling
waves coming in over the lines. The protection of stations from direct strokes will be
discussed in this section and the protection of equipment from travelling waves will
be discussed in the subsequent section.
Wherever it is likely for direct strokes of lightning to strike the line at or near the
station, there is a possibility of exceedingly high rates of surge-voltage rise and
 Base of Tower

(a) Direction of Line (b)

(c) (d)

(e)

Fig. 9 Typical arrangements of tower-to-tower (continuous) counterpoise wires

large magnitude of surge-current discharge. If the stroke is severe enough, the margin
of protection provided by the protective device may be inadequate. The installation
may, therefore, require shielding of the station and the incoming lines enough to limit
the severity of surges particularly in the higher voltage class of 66 kV and above.
This can be done by properly placed overhead ground wires, masts, or rods.
Therefore protection of stations from direct strokes involves the principles already
discussed in Sec. 6. If supports are available for overhead ground wires, these may be
run over the station in such a way that the station and all apparatus will lie in the
protected zone. Sufficient ground wires should be strung so that the entire station is
covered, including any apparatus outside the main structure and any apparatus on top
of the structure as shown in Fig. l0.
For a small station, it is sometimes sufficient to run one or two ground wires across
the station from adjacent line towers as shown in Fig. 11(a). For a more extensive
station, extra ground wire may be used, fanning out from the towers to the station
structure, as also over the station if mechanically feasible as shown in Fig. 11(b). If
there are no isolating switches or other apparatus above the station structure, then the
steel of this structure will act as a grounded object to protect the apparatus and buses
below it.

Zone of
Ground Wires
Protection
Equipment

Fig. 10 Protection of a station by overhead ground wires (elevation view)

 Ground Wires
Station

Station
Tower Tower
Tower Tower

Ground
Wires
(a) (b)

Fig. 11 (a) Overhead ground wires carried over a small station between adjacent
towers (plan view) (b) Overhead ground wires carried over a large station
between adjacent towers (plan view)

If it is not feasible to run ground wires over the station, then it may be possible to
erect masts or rods at the corners or over vertical columns, so that the buses and
apparatus in the station will fall within the cone of protection of the masts or rods as
in Fig. 7(b). Such a cone may be assumed to have a base radius equal to twice the
height of the mast. Figure 12 illustrates this method of protection.

8
Shielding the lines and stations (including sub-stations) by overhead ground wires,
masts or rods undoubtedly provides adequate protection against direct lightning
strokes and also reduces electrostatically or electromagnetically induced over volt-
ages, but such shielding docs not provide protection against travelling waves which
may come in over the lines and reach the terminal equipment. Such travelling waves
can cause the following damage to electrical equipment.
(i) Internal flashover caused by the high peak voltage of the surge may damage
the insulation of the winding.

Cone of Masts
Protection

Fig. 12 Method of protecting station by vertical masts (elevation view)

(ii) Internal flashover between interturns of the transformer may be caused by the
steep-fronted voltage wave.
(iii) External flashover between the terminals of the electrical equipment caused by
the high peak voltage of the surge may result in damage to the insulators.
(iv) Resonance and high voltages resulting from the steep-fronted wave may cause
internal or external flashover of an unpredictable nature causing building up
of oscillations in the electrical equipment.
 Hence it is absolutely necessary to provide some protective devices at the stations
or sub-stations for the protection of equipment against the travelling waves (surges)
caused by lightning. The protective devices used for this purpose are described in the
following sub-sections.
8.1 Rod Gap
A rod gap provides the simplest and cheapest protection to line insulators, equip-
ment insulators and bushings of transformers. It is clear that in case of serious over-
voltages, the over-insulation of any one part of the power system may cause the
breakdown of the insulation of some vital and perhaps inaccessible part. Thus, it
is preferred that it is the line insulators that flashover rather than the bushings of a
transformer breaking down. Again, it is preferred that a bushing breakdown before
the insulation of the transformer of which it forms a part. In the case of transformers,
rod gaps which are also called coordinating gaps are installed to protect the appara-
tus. Rod gaps provide back-up protection to the bushings of transformers in case of
the primary protective devices, i.e. lightning arresters fail.
A rod gap consists of two rods of Bushing
approximately 1.2 cm diameter or Insulator
square, which are bent at right angles as
shown in Fig.13. One rod is con- nected
to the line while the other rod is

Rod Gap
connected to ground. In case of a trans- L >
L
former, they are fixed between bushing L1 L 1 3
insulators. In order to avoid cascading
across the insulator surface under very
steep-fronted waves, the rod gap should
be adjusted to breakdown at about 20
per cent below the impulse flashover

voltage of the insulation of the equip- Fig. 13 Rod gap

ment to be protected (i.e. bushing insu-

lator of a transformer in Fig.13). Further, the distance between the gap and the
L
(
insulator should be more than one-third of the gap length i.e. L1 >
3) in order to
prevent the arc from being blown to the insulator. The accurate breakdown value of
the rod gap cannot easily be predicted because the breakdown of air depends upon
the atmospheric conditions (i.e., humidity, temperature and pressure), as also upon
polarity, steepness and the waveshape of the wave.
The major disadvantage of the rod gap is that it does not interrupt the power fre-
quency follow current after the surge has disappeared. This means that every opera-
tion of the rod gap creates an L-G fault which can only be cleared by the operation
of the circuit breaker. Thus, the operation of the rod gap results in circuit outage and
interruption of power supply.
8.1 Arcing Horn
The damage to line insulators from heavy arcs formed due to overvoltages is a seri-
ous maintenance problem. Several protective devices have been developed to keep
an insulator string free from the arc. Arcing horn
is one of such protective devices. It consists of
small horns attached to the clamp of the line
insulator string. Horns with a large spread, both
at the top of the insulator and at the clamp are
required to be effective, as shown in Fig.14. In
the case of lightning impulse, the arc formed
tends to cascade the string. In order to avoid this
cascading, the gap between horns should be
considerably less than the length of the string.
Protection of line insulators by arcing horns thus FIG. 14 Suspension string with
results in reduced flashover voltage. In any arcing horns
event, flashover persists as a power arc until the

line trips out. The protection of line insulators by arcing horns is especially used in
hilly areas.
The grading ring when used in conjunction with an arcing horn fixed at the top of
the insulator string serves the purpose of an arcing shield. In the event of an arc
forming following a flashover caused by some type of overvoltage, the arc will usu-
ally take the path between the horn and the shield and the insulator string will remain
clear from the arc.
8.1 Lightning Arresters or Surge Diverters
Lightning arresters are also known as surge diverters or surge arresters. They are
connected between the line and ground at the substation and always act in shunt
(parallel) with the equipment to be protected, and perform their protective function
by providing a low-impedance path for the surge currents so that the surge arrester’s
protective level is less than the surge voltage withstanding capacity of the insulation
of equipment being protected. A lightning arrester’s protective level is the voltage
appearing across the terminals of the arrester at sparkover or during the flow of cur-
rent through the arrester after sparkover. The main purpose of lightning arresters is to
divert or discharge the surge to the ground.
The action of the surge diverter can Surge voltage if no
be studied with the help of Fig.15. When protective device
the travelling surge reaches the diverter, is present
Voltage

it sparks over at a certain pre- fixed

voltage as shown by point P and Initial peak
provides a relatively low-impedance Q Residual voltage
path to ground for the surge current. The
P
current flowing to ground through the
surge impedance of the line limits the
amplitude of the overvoltage across the
line and ground known as ‘residual volt- Time
age’ (as shown by point Q in Fig.15) to
such a value which will protect the Fig. 15 Voltage characteristic of surge
insulation of the equipment being diverter
protected.
 It is however, essential that the low-impedance path to ground must not exist
before the overvoltage appears and it must cease to exist immediately after the volt-
age returns back to its normal value.
An ideal lightning arrester or surge diverter should possess the following
characteristics.
(i) It should not draw any current at normal power frequency voltage, i.e. during
the normal operation.
(ii) It should breakdown very quickly when the abnormal transient voltage above
its breakdown value appears, so that a low-impedance path to ground can be
provided.
(iii) The discharge current after breakdown should not be so excessive so as to
damage the surge diverter itself.
(iv) It must be capable of interrupting the power frequency follow-up current after
the surge is discharged to ground.
Impulse Ratio of Lightning Arresters
It is defined as the ratio of the breakdown impulse voltage of a wave of specified
duration to the breakdown voltage of a power-frequency wave. The impulse ratio of
any lightning arrester is therefore a function of time duration of the transient wave.
Types of Lightning Arresters (or Surge Diverters)
The following are the main types of lightning arresters.
(i) Expulsion Type Lightning Arrester
(ii) Non-linear Surge Diverter
(iii) Metaloxide Surge Arrester (MOA)
(i) Expulsion type lightning Line
arrester External
Series Gap
This type of arrester is also known as
Expulsion Gap or Protector Tube. It
Upper Metal Electrode
consists of a fibre tube with an elec-
trode at each end. The lower electrode
is solidly grounded. The upper elec-
trode forms a series gap with the line
Fiber Tube
conductor, as shown in Fig. 16 When Gap
a surge appears on the conduc- tor, the
series gap breaks down, result- ing in
formation of arc in the fibre tube
between the two electrodes. The heat
of the arc vaporises some of the fibre
of the tube walls resulting in the Bottom Metal Electrode
generation of an inert gas. This gas is
expelled violently through the arc so Vent for Gases
that arc is extinguished and the power

Fig.16 Expulsion type lightning arrester

frequency current is prevented from
flowing after the surge discharge.
 Line
The expulsion arrester derives its name from
the fact that gases formed during its operation are Divided
expelled from the tube through a vent. spark gap

This type of arrester is mainly used to prevent flash- Leak tight

porcelain
over of line insulators, isolators and bus insulators. housing
(ii) Non-linear surge diverter
Non-linear
This type of diverter is also called valve type light- resistor
ning arrester, or conventional non-linear type light-
ning arrester. It consists of a divided spark-gap (i.e.
several short gaps in series) in series with non-linear
resistor elements. The divided spark-gap and the non-
linear resistor elements are placed in leaktight porce- Fig. 17 Value-type
lain housing which ensures reliable protection against lightning arrester
atmospheric moisture, condensation and humidity (non-linear diverter)
(see Fig. 17). Figure 17 shows a typical valve type
lightning arrester.

The functions of the diverter’s divided spark-gap are as follows

(a) It prevents the flow of current through the diverter under normal conditions.
(b) It sparks over at a predetermined voltage.
(c) It discharges high-energy surges without any change in sparkover character-
istics.
(d) It interrupts the flow of power-frequency follow current from the power system
after the surge has been dissipated.
The functions of the non-linear resistors are as follows:
(a) They provide a low-impedance path for the flow of surge current after gap
sparks over.
(b) They dissipate surge energy.
(c) They provide a relatively high-resistance path for the flow of power frequency
follow current from the power system, thereby assisting the divided gap to
interrupt the power frequency current (i.e. reseal against system voltage).
The divided spark-gap consists of a number of short-gaps in series. Each of them
has two electrodes across which a grading resistor of high ohmic value is connected.
The grading resistors ensure even grading of voltage between the various gaps. The
system is similar to that of a number of capacitors connected in series and across
each of these capacitors is a high value resistor. The voltage grading by means of
resistors also makes it possible to raise the interrupting capacity of the divided spark-
gap.
In case of relatively slow variations in voltage, there is no sparkover across the
gaps as the influence of parallel grading resistors across these gaps prevails over that
of the spark-gap capacitances with regard to one another and to the ground. But when
there are large rapid variations in voltage, the even grading of the voltage across the
series of gaps no longer remains and the influence of unbalanced capacitances of the
gaps prevails over the grading resistors. The surges are mainly concentrated on the
upper spark gaps which on sparking over result in the sparkover of the complete
arrester.
 Terminal for line connection

Cap casting (Aluminium)

Spring

Thyrite discs

Wet process procelain container

Series gap unit

Steam Cured Cement

Bottom Casting (Aluminium)
Terminal for Ground Connection

Foundation Base

Fig. 18 A typical value-type lightning arrester

The main function of the diverter is the protection of the insulation against dan-
gerously high overvoltages and for this reason the breakdown voltage of the diverter
at system frequency is made greater than 1.8 times the normal value.
The ideal characteristic for the non-linear resistor would be RI = constant, since
such a resistor will maintain constant voltage by changing its resistance in inverse
proportion to the current. The non-linear resistor is in the form of discs of 9 cm diam-
eter and 2.5 cm thickness. The material used for such resistors, called by trade names
of “Thyrite”, and “Metrosil” is a hard ceramic substance in the form of cylindrical
blocks which consist of small crystals of silicon carbide bound together by means of
an inorganic binder, with the whole assembly being subjected to heat treatment. The
non-linear characteristic is attributed to the properties of the electrical contacts
between the grains of silicon carbide. As the material used for non-linear resistors for
the arrester is known by trade name as thyrite, this type of lightning arrester is also
sometimes called Thyrite type lightning arrester.
Figure 19 shows the volt-ampere characteristic of the non-linear resistor, the
dotted curve being the static characteristic, while the solid curve is the dynamic char-
acteristic corresponding to the application of a voltage surge. If a horizontal line,
tangential to the dynamic characteristic is drawn, then its intercept with the voltage
axis gives the residual voltage. The residual voltage is defined as the peak (crest)
 Voltage
value of the voltage which appears between
the terminals of the surge diverter at the
instant when the surge current discharges
through it. This voltage varies between 3

Residual
Voltage
kV and 6 kV, depending on the type i.e. Static
characteristic
station type or line type, the discharge cur-
rent and the rate of change of this current. Dynamic
characteristic
A non-linear resistor type lightning Current
arrester is also known as valve type light-

ning arrester. The valve type arrester Fig.19 Volt-ampere characteristics

derives its name from the fact that the non- of non-linear diverter

linear resistance (similar to valve opening) regulates itself to the flowing current and
limits the voltage when high surge (lightning) currents flow through the arrester.
The valve arrester works as an insulator at normal system voltage, but it becomes
a good conductor of low resistance to limit the lightning (surge) voltages to a value
lower than the basic impulse level of the equipment being protected. It re-establishes
itself as an insulator after the discharge of the surge current.
Comparison of Expulsion and Non-linear Types of Surge Diverters The series
gap of the expulsion-type surge diverter (lightning arrester) acts as an isolator under
normal conditions. When the diverter is required to operate, the impedance of the series
gap is very low and the arc formed in the fibre tube is also very low, with the result that
the follow-up current is almost the same as that due to the line to ground (L-G) fault,
its value therefore being determined by the factors used in the calculation of fault
currents. Thus the current through the expulsion type diverter may become as high as
several kilo-amperes (kA). Secondly, as the follow-up current is determined by the net-
work under fault conditions, it is a reactive current and lags the system voltage by 90
degree, i.e. it passes through a zero value when the system voltage is maximum.
Since the arc voltage at the instant of breakdown falls rapidly to a very low value,
there is no phenomenon which may be analogous to residual voltage. Therefore, the
level of protection is fixed only by the value of the impulse breakdown voltage. It
gets increasingly difficult for the expulsion gap to interrupt currents of low value,
because of the small amount of deionising gas formed by it.
Though the sub-divided series gap of the non-linear surge diverter is also an
isolator during normal conditions, it actually plays a role in the performance of
the diverter during its operation. As the current through the diverter is lim- ited
to a comparatively low value by the non-linear resistors, it is practically in phase with
the system voltage. Because of the current being practically in phase with the system
voltage, an instant of current zero is also an instant of zero for the system voltage.
Thus, the sub-divided gap interrupts the current at the first zero. As the impulse
breakdown voltage is usually less than the residual voltage, it is the latter which
determines the level of protection provided by the non-linear surge diverter.
(iii) Metal-oxide surge arrester (MOA)
The metal oxide surge arrester abbreviated as MOA is a recently developed ideal
surge arrester. It is a revolutionary advanced surge protective device for power sys-
tems. It is constructed by a series connection of zinc oxide (ZnO) elements having
a highly non-linear resistance. The excellent non-linear characteristic of zinc oxide
element has enabled to make surge arresters without series connected spark gaps, i.e.
fully solid-state arresters suitable for system protection up to the highest voltages.
As mentioned earlier, the conventional non-linear surge diverters almost exclu-
sively use Silicon Carbide (SiC) non-linear resistors. As this material is not ideal,
it is not non-linear enough and thus imposes certain design restrictions. Also, its
characteristics call for a large number of spark gaps. A new class of non-linear mate-
rial was recently developed in Japan by Japan’s Matsushita Electric Company. The
new ceramic material is basically formed from zinc oxide together with addition of
other oxides, such as bismuth and cobalt oxides. As the main constituent of this new
ceramic material is zinc oxide, the non-linear resistor made of this material is popu-
larly known as zinc oxide element and surge arresters made of zinc oxide elements
are called metal oxide surge arresters. Such material can be used to make resistors
with a much higher degree of non-linearity over a large current range. With such
resistors, one can design arresters having voltage-current characteristics very close to
ideal. Because of the high degree of non-linearity, this material allows considerable
simplification in arrester (diverter) design.
The metal oxide surge arrester (MOA) which consists of a series connected stack
of discs of zinc oxide elements, operates in a very simple fashion. It is dimensioned
so that the peak value of the phase to ground voltage in normal operation never
exceeds the sum of the rated voltages of the series-connected discs. The resistive
losses in the arrester in normal operation are therefore very small. When an overvolt-
age occurs, the current will rise with the wavefront according to the characteristics
without delay. No breakdown occurs but a rather continuous transition to the con-
ducting state is observed. At the end of the voltage transient, the current is reduced
closely following the I-V curve (i.e. in contrast to the conventional arrester, there is
no follow-up current).
The metal oxide surge arrester has the following marked advantages over conven-
tional arresters.
(i) Series spark-gap is not required.
(ii) It has very simple construction and is a fully solid-state protective device.
(iii) Significant reduction in size.
(iv) Quick response for steep discharge current.
(v) Very small time delay in responding to overvoltages.
(vi) Superior protective performance.
(vii) Outstanding durability for multiple operating duty cycle.
(viii) No abrupt transient such as that occurs at the time of sparkover in a conven-
tional arrester.
(ix) Negligible power follow-up current after a surge operation.
MOA is especially suitable for gas insulated sub-stations (GIS), since it can be
installed directly in SF6.
Figure 20 illustrates the difference in the operations of MOA and conventional
arresters, responding to an on-coming surge v(t). The voltage wave shape of MOA is
smooth and can be expressed as follows
 V (t) V (t)

Z0 Z0
ZnO SiC

Volt without
Volt without
Arrester
Arrester
2 v (t) 2 v (t)
Volt Volt
Current

Current
(a) MOA (b) Conventional arrester

Fig.20 Operation of MOA and conventional arrester responding to an

on-coming surge

V = 2v(t) – Z0i (8)

where Z0 is the surge impedance of the line and i is the instantaneous current of the
arrester.
The voltage wave shape of the conventional arrester has a peak value at the time
of sparkover of the series gap. The voltage after sparkover or the discharge voltage
can be expressed by Eq. (8).
As seen from Fig.20, the protection performance of the conventional arrester is
controlled by both the sparkover voltage and the discharge voltage. However, the
protection performance of MOA is simply related to the discharge voltage, given by
Eq. (8).
The protection level of MOA is decided by the maximum discharge voltage
encountered under normal operating conditions. Though it is difficult to get a clear-
cut expression like the V-t curve in a conventional arrester, the individual voltage
waveform should be considered in precise discussion of insulation coordination.
For rough estimation of the voltage waveform, it is convenient to assume a fixed
functional relation between the voltage and current of MOA, as follows.
V = v(i) (9)
The voltage waveform or the time dependence of V can be calculated by connect-
ing Eqs. (8) and (9). A graphical method of solving the equation is shown in Fig.21
(a). The V and I values corresponding to v(t) are represented by P0 which is the
intersection point of two curves corresponding to Eqs. (8) and (9). In this simple
treatment, the point P0 moves on the v(i) curve shown by a dotted line in Fig. 21(a).
However, the actual V-I point moves along the thick line which cannot be conformed
to a single line. A typical V-I characteristic of conventional SiC arrester is
conceptually shown in Fig.21(b) for comparison. In this case, the change from
sparkover voltage (SOV) to a lower discharge voltage is abrupt.
 V V
2v (t)
Discharge volt SOV
Discharge volt
P

P0 v (t)

Slope: – Z0

l
0 0
(a) MOA (b) Conventional

Fig. 21 Transient V-I characteristics of MOA and conventional arrester

The main drawback of MOA is that the absence of spark-gaps results in a continu-
ous flow of current through the device so that there is theoretically speaking the dan-
ger of thermal runway. However, as very stable resistors have been developed, these
apprehensions are almost eliminated. Furthermore, the absence of spark-gaps makes
the voltage grading system unnecessary. The MOA is inherently self-regulating in
that the current flow of 0.5 to 1 mA at normal supply voltage leads to reliable opera-
tion even in polluted conditions.

8.2 Surge Absorber

A surge absorber is also known as a surge modifier. It is a device which absorbs the
energy contained in a travelling wave and reduces the amplitude of the surge and the
steepness of its wave front. The energy absorption in the form of corona loss due to
corona formation acts as a kind of a natural safety-valve in the case of travelling
waves which raise the line voltage above the corona level. There may be some cases
in which a length of line adjacent to the terminal station may be working at a lower
critical corona voltage than the main length of the line. Another method of lessen-
ing the risk of travelling waves is the installation of a length of steel conductor at the
terminal end, say 1.5 to 3 kms away. The skin effect of the wire to steep fronted
waves being equivalent to an appreciable increase in the high frequency resistance.
These methods are compromises which do not provide adequate protection against
disturbances originating near the station.
A better method of reducing the risk of travelling waves is to terminate the line at
a distance of 0.8 to 1.5 km from the station (or sub-station) and to connect it to the
station with an underground cable. This method has the following three advantages.
(a) The cable gets shielded from all electrostatic phenomena, and from accidental
grounds (earths) through various causes.
(b) When the travelling waves enter the cable from the overhead line, they are
reduced in magnitude to about 20 per cent of their incident value because of
the ratio of the natural impedances.
(c) When a travelling wave is transmitted into the cable, it gets attenuated rapidly
due to the dielectric losses taking place in the cable at high frequencies.
 Dissipator
The most modern surge absorber is the Air-cored Inductor
Ferranti surge absorber (Fig.22). It consists of
an aircored inductor connected in series with
each line and surrounded by a grounded metal-
lic sheet called a dissipator. The inductor is
magnetically coupled but electrically isolated
from the dissipator. The inductor is insulated Fig.22 Ferranti surge absorber
from the dissipator by the air.

This surge absorber acts as an air cored transformer whose primary is the induc-
tor and the dissipator is the short-circuited secondary of a single turn. Whenever the
travelling wave is incident on the surge absorber, the energy contained in the wave is
dissipated in the form of heat generated in the dissipator; firstly, due to the current set
up in it by ordinary transformer action, the secondly, by eddy currents. The steepness
of the wavefront is also reduced because of the series inductance.

9 PETERSON COIL
The Peterson coil is also known as a ground-fault neutralizer or arc suppression coil.
It is an iron cored reactor connected between the neutral point and ground and capa-
ble of being tuned with the capacitance of the healthy phases to produce resonance
when a line to ground (L-G) fault occurs Fig.23(a).1t is mainly used to prevent arcing
grounds which lead to overvoltages on systems with an ungrounded (isolated) neutral.
The Peterson coil makes arcing ground faults self-extinguishing and in case of
sustained ground fault on one of the lines, it reduces the fault current to a very low
value so that the healthy phases can be kept in operation even with one line grounded.
In order to select the proper value of inductive reactance, depending upon the length
of the transmission line and hence, the capacitance to be neutralised. Peterson coil is
provided with tappings.

3l V
3I R
R
3l
N Y 3l
F
B
IL O N
3l
L VY
IL VB
IF
Peterson
coil
IL
(a) Connection of Peterson coil (b) Phasor diagram

Fig.23 Resonant grounded 3-phase system

In an undergrounded system, when a ground fault occurs on any one line, the
voltage of the healthy phases is increased ÷3 times (i.e. ÷3 Vp, where Vp is the phase
voltage). Hence the charging currents become ÷3 I per phase, where I is the charging
current of line to ground of one phase. The phasor sum of the charging currents of
the healthy phases becomes 3 times the normal line to neutral charging current of one
phase, as shown in phasor diagram of Fig.23(b).

Hence, I = 3I = 3VP = 3V w C (10)

c P
1/wC
where Ic is the resultant charging current and I is the charging current of line to
ground of one phase.
If L is the inductance of the Peterson coil connected between the neutral and the
ground, then
VP
IL = (11)
wL

In order to obtain satisfactory cancellation (neutralisation) of arcing grounds, the

fault current IL flowing through the Peterson coil should be equal to the resultant
charging current IC.
Therefore, for balance condition
IL = IC

or VP = 3V w C (From Eqs. (16.10) and (16.11))

p
wL
1
or L= (12)
3w2C
The inductance L of Peterson coil is calculated from Eq. (12).

Example 1 A 33 kV, 3 phase 50 Hz, overhead line 60 km long has a capaci-

tance to ground of each line equal to 0.015 mF per km. Determine the inductance
and kVA rating of the Peterson coil.
1 1
Solution: L = = = 3.75 H
3w2C 3 × (314)2 × 0.015 × 60 × 10–6
XL = 2p fL = 314 × 3.75 = 1177.5 W
For ground fault, the current in the neutral is given by
VP 33 × 1000
IN = IL = = = 16.18 A
wL ÷3 × 1177.5

The voltage across the neutral is the phase voltage.

Hence, kVA rating = Vp × lN
33
= × 16.18 = 308.3 kVA
÷3

Example 2 A 50 Hz overhead line has the line to ground capacitance of 1.2

mF. It is decided to use a ground-fault neutralizer. Determine the reactance to
neutralise the capacitance of (i) 100% of the length of the line, (ii) 95% of the
length of the line, and (iii) 80% of the length of the line.
Solution:
(i) The inductive reactance of the coil for 100% neutralisation will be
1 = 1 106
wL = = = 884.6 W
3wC 3 × 314 × 1.2 × 10–6 3 × 314 × 1.2

(ii) The inductive reactance for neutralizing 95% of the capacitance will be
1 = 106
wL = = 931.2 W
3wC 3 × 314 × 1.2 × 0.95

(iii) The inductive reactance for 80% neutralization will be

1 = 106
wL = = 1105.8 W
3wC 3 × 314 × 1.2 × 0.8

10
Insulation coordination is the correlation of the insulation of electrical equipment and
lines with the characteristics of protective devices such that the insulation of the
whole power system is protected from excessive overvoltages.
The main aim of insulation coordination is the selection of suitable values for the
insulation level of the different components in any power system and their arrange-
ment in a reasonable manner so that the whole power system is protected from over-
voltages of excessive magnitude. Thus, the insulation strength of various equipment,
like transformers, circuit breakers, etc. should be higher than that of the lightning
arresters and other surge protective devices. The insulation coordination is thus the
matching of the volt-time flashover and breakdown characteristics of equipment and
protective devices, in order to obtain maximum protective margin at a reasonable
cost. The volt-time curves of equipment to be protected and the protective device are
shown in Fig.24. Curve A is the
volt-time curve of the protective device
and curve B is the volt-time curve of the
kV Peak

equipment to be protected. From volt-time

curves A and B of Fig.24, it is clear that
any insulation having a voltage with-
standing strength in excess of the insula- Time

tion strength of curve B will be protected Fig.24 Volt-time curves of protective

by the protective device of curve A. device and the equipment to
be protected
10. 1 Volt-time Curve

The breakdown voltage of any insulation or the flashover voltage of a gap depends
upon both the magnitude of the voltage and the time of application of the voltage.
The volt-time curve is a graph of the crest flashover voltages plotted against time to
flashover for a series of impulse applications of a given wave shape. The construction
of the volt-time curve and the terminology associated with impulse testing are shown
in Fig.25. The construction of the volt-time curve is based on the application of
impulse voltages of the same wave shape but of different peak values to the insula-
tion whose volt-time curve is required. If an impulse voltage of a given wave shape
and polarity is adjusted so that the test specimen (i.e. a particular insulation) flashes
over on the front of the wave, the value of voltage corresponding to the point on front
of the wave at which flashover occurs is called front flashover. If an impulse voltage
of the same wave shape is adjusted so that the test specimen flashes over on the tail
of the wave at 50 per cent of the applications and fails to flashover on the other 50 per
cent of the applications, the crest value of this voltage is called the critical flashover
voltage. If an impulse voltage causes flashover of the test specimen exactly at the
crest value, then it is called crest flashover. If flashover does not take place, the wave
is called a full wave and if flashover does take place, the wave is called a chopped

Wave front flashover

voltage range
Front flashover Wave tail flashover
Voltage

voltage range
Crest flashover
Volt-time curve

Critical Critical flashover

Tail flashover 50% of spplication
flashover
flashover
Critical 50% of application
withstand donot
flashover
Rated Time of
withstand crest
flashover Time of critical flashover

Time range wave tail Time

Time Range flashover range
wave-front no impulse
flashover flashover

Time in Microsecond

Fig.25 Construction of volt-time curve and the terminology associated with

impluse testing

wave. The applied impulse voltage reduced to just below the flashover voltage of the
test specimen is called the critical withstand voltage. The rated withstand voltage
is the crest value of the impulse wave that the test specimen will withstand without
disruptive discharge.
Figure 25 illustrates for an assumed impulse, its wave front flashover, crest
flashover, critical flashover voltage, critical withstand voltage, rated flashover volt-
age, wavetail flashover and a volt-time curve.

11 BASIC IMPULSE INSULATION LEVEL (BIL)

The insulation strength of equipment like transformers, circuit breakers, etc. should
be higher than that of the lightning arresters and other surge protective devices. In
order to protect the equipment of the power system from overvoltages of excessive
magnitude, it is necessary to fix an insulation level for the system to see that any insu-
lation in the system does not breakdown or flashover below this level and to apply
protective devices that will give the apparatus effective protection.
 Several methods of providing coordination between insulation levels in
the station and on the line leading to the station have been offered. The best method
is to establish a definite common level for all the insulation in the sta- tion and
bring all insulation to or above this level. This limits the problem to three
fundamental requirements, namely, the selection of a suitable insulation level, the
assurance that the breakdown or flashover strength of all insulation in the station
will equal or exceed the selected level, and the application of pro- tective devices
that will give the apparatus as good protection as can be justified economically.
The common insulation level for all the insulation in the station is known as Basic
Impulse Insulation Level (BIL) which have been established in terms of withstanding
voltages of apparatus and lines.
Basic impulse insulation level can be defined as reference level expressed in
impulse crest voltage with a standard wave not longer than a 1.2/50 microsecond
wave, according to Indian standards (1.5/40 ms in USA and 1/50 ms in UK). Apparatus
insulation as demonstrated by suitable tests should be equal to or greater than the
BIL.
The basic impulse insulation level for a system is selected such that the system
could be protected with a suitable lightning protective device, e.g. a lightning arrester.
The margin between the BIL and the lightning arrester should be fixed such that it is
economical and it also ensures protection to the system. The BIL chosen must be
higher than the maximum expected surge voltage across the selected arrester.

EXERCISES
1. What are the causes of overvoltages arising on a power system? Why is it
necessary to protect the lines and other equipment of the power system against
overvoltages?
2. Describe the phenomenon of lightning and explain the terms pilot streamer,
stepped leader, return streamer, dart leader, cold lightning stroke and hot
lighting stroke.
3. How can the magnitude of overvoltages due to direct and indirect lightning strokes on
overhead lines be calculated?
4. What protective measures are taken against lightning overvoltages?
5. What is a ground wire? What are the requirements to be satisfied by ground
wires to provide efficient protection to lines against direct lightning strokes?
How do ground wires protect the overhead lines against direct lightning
strokes.
6. What is tower-footing resistance? Why is it required to have this resistance as
low as economically possible? What are the methods to reduce this resis-
tance?
7. Explain the terms overvoltage factor, protective ratio, protective angle, protec-
tive zone and coupling factor.
 8. Differentiate between surge diverter and surge absorber. What are the char-
acteristics of an ideal surge diverter?
9. What is the necessity of protecting electrical equipment against travelling
waves? Describe in brief the protective devices used for protection of equip-
ment against such waves.
10. Describe the protection of stations and sub-stations against direct lightning
strokes.
11. Describe the construction and principle of operation of (i) Expulsion type
lightning arrester, and (ii) Valve type lightning arrester.
12. Compare the protection performance of expulsion-type surge diverter with
that of non-linear surge diverter.
13. Describe the construction and the operation of metal oxide surge arrester?
What are its advantages over conventional arresters? Discuss the difference in
protection performance of MOA and conventional non-linear type arrester
with the help of suitable figures.
14. What is a Peterson coil? What protective function are performed by this
device?
15. A 132 kV, 3-phase, 50 Hz transmission line 200 km long contains three con-
ductors of effective diameter 2.2 cm, arranged in a vertical plane with 4.5 m
spacing and regularly transposed. Find the inductance and kVA rating of the
Peterson coil in the system.
16. A 50 Hz overhead line has line to ground capacitance of 1.5 mF. It is decided
to use ground-fault neutralizer. Determine the reactance to neutralize the
capacitance of (i) 85% of the length of the line (ii) 80% of the length of the
line, and (iii) 75% of the length of the line.
17. Explain the term insulation coordination. Describe the construction of volt-
time curve and the terminology associated with impulse testing.
18. Write short notes on the following.
(i) Klydonograph and magnetic link
(ii) Rod gap
(iii) Arcing horns
(iv) Ferranti surge absorber
(v) Basic impulse insulation level