Power System Protection & Switchgear

(EE3203)

Module 2: Protection of Synchronous Generator

Dr. Abhinandan De
Department of Electrical Engineering
IIEST, Shibpur
Protection of Synchronous Generators
Synchronous generator or alternator is one of the costliest and most complex equipment in
electrical power system. A synchronous generator comprises of several important
components, such as: prime mover, excitation system, voltage regulator (AVR), cooling
system, etc. As a result, the protection of synchronous generator is very complex and
elaborate. Modern synchronous generators are generally equipped with different types of
protection schemes. The major types of protection which are common for a synchronous
generator are mentioned under.

1. Types of Protection for Modern Synchronous Generators

1.1 Stator protection
(a) Percentage differential protection for stator winding
(b) Protection against stator inter-turn faults
(c) Stator-overheating protection
1.2 Rotor protection
(a) Rotor ground-fault protection
(b) Loss of excitation protection
(c) Protection against rotor overheating due to unbalanced three-phase stator currents
1.3 Miscellaneous protection
(a) Overvoltage protection
(b) Overspeed protection
(c) Protection against motoring
(d) Protection against mechanical vibration
(e) Bearing-overheating protection
(f) Protection against auxiliary failure
(g) Protection against voltage regulator failure

1.1 (a) Percentage differential protection for Stator Winding of Alternators

i. Basics of Differential Protection
“Percentage differential relay” or “biased differential relay” are highly sensitive to even
very minute internal faults taking place within the protected winding section, while
providing high restraint for even severe external faults. Thus, percentage differential relays
are usually more stable compared to ordinary differential relays. The schematic diagram of

2
Power System Protection & Switchgear (EE3203) Lecture Module – 2 by Dr. A. De, EE Dept., IIEST
the percentage (biased) differential relay is shown in Fig. 1. This relay has two coils. One coil
is known as “restraining coil” or “bias coil” which restrains (or inhibits) the operation of
the relay. The other coil is called the “operating coil” which produces the operating torque
for the relay operation.
When the operating coil torque exceeds the restraining coil torque, the relay operates.
The operating coil is connected to the mid-point of the restraining coil as shown in Fig.1. Nr
and No are the total number of turns of the restraining coil and the operating coil
respectively. Since the restraining coil is tapped at the centre, it forms two sections with
equal number of turns: Nr/2.
The restraining coil is connected in the circulating current path in such a way that current I1s
flows through one section of Nr/2 turns and I2s flows through the other section of Nr/2, so
that the complete restraining coil of Nr turns receives a through fault current of (I1s + I2s)/2.
The operating coil, having N0 number of turns, is connected in such a manner that it receives
differential current of (I1s – I2s).

Fig.1 Percentage (or Biased) Differential Relay

ii. The operating condition of the percentage differential relay

The relay operates if the torque produced by the operating coil is more than the restraining
torque produced by the restraining coil. As the torque is proportional to the Ampere-Turns
(AT), the relay will operate when the AT of the operating coil (AT)0, becomes greater than AT
of the restraining coil, (AT)r

3
Power System Protection & Switchgear (EE3203) Lecture Module – 2 by Dr. A. De, EE Dept., IIEST
From the above equation, it can be assumed that the entire Nr turns of the restraining coil
carries a current (I1s + I2s)/2. The current (I1s + I2s)/2 which is the average of the secondary
currents of the two CTs (CT1 and CT2) is known as the through current or restraining
current (Ir).

Thus, Ir = (I1s + I2s)/2

On the other hand, the AT produced by the operating coil: (AT)0 = N0(I1s – I2s)

Neglecting spring restraint, the relay will operate when,

where, Id = (I1s – I2s) is the differential current flowing through the operating coil, and as
explained earlier, Ir = (I1s + I2s)/2 is the restraining current (or through current).

Thus, the relay will be at the verge of operation, when the ratio of the differential current
(Id) to the restraining current (Ir) is equal to a constant K, or in other words, when Id is
a certain fixed percentage of Ir

The relay will operate, only when the differential (operating) current is greater than
this fixed percentage of the restraining (through fault) current. Hence, this relay is
called ‘percentage differential relay’.

4
Power System Protection & Switchgear (EE3203) Lecture Module – 2 by Dr. A. De, EE Dept., IIEST
If the effect of control spring is taken into account, then the above equation takes the form:
Id = KIr + Ks, where Ks represents for the effect of control spring.
The typical values of K (slope or bias) can be 10%, 20%, 30% and 40%. A relay with a
slope of 10% is more sensitive than a relay with a slope of 40%. The slope of the relay
determines its tripping zone and blocking zone as shown in Fig. 2.

Slope of the line (boundary) = K

Id / Ir > K : Trip
Id / Ir < K : Block

Fig.2 Operating characteristic of Percentage (or Biased) Differential Relay

Fig.3 Generator Differential Relay

5
Power System Protection & Switchgear (EE3203) Lecture Module – 2 by Dr. A. De, EE Dept., IIEST
iii. Percentage Differential Protection for Thee-Phase Alternators (with all the 6
winding terminals brought out)
Figure 4(a) shows the schematic diagram of percentage differential protection for a 3-phase
star-connected alternator. It is used for the protection of generators above 1 MW. It protects
against winding faults, i.e. phase to phase and phase to ground faults. This is also called
biased differential protection or longitudinal differential protection. The polarity of the
secondary voltage of CTs at a particular moment for an external fault has been shown in the
figure. In the operating coil, the current sent by the upper CT is cancelled by the current sent
by the lower CT and the relay does not operate. For an internal fault, the polarity of the
secondary voltage of the upper CT is reversed, as shown in Fig. 4(b). Now the operating coil
carries the sum of the currents sent by the upper CT and the lower CT and it operates and
trips the circuit breaker.

Fig.4(a) Percentage differential Fig.4(b) Percentage differential

protection - instantaneous current protection - instantaneous current
directions for external fault directions for internal fault

The percentage differential protection does not respond to external faults and overloads. It
provides complete protection against phase-to-phase faults. But, it provides protection
against ground faults to about 80 - 85 % of the winding length from line terminal. Protection
to 100% of the winding length cannot be provided by percentage differential protection
because it is influenced by the magnitude of the earth fault current which depends upon the
type of neutral grounding. When the neutral is grounded through an impedance, the
differential protection has to be supplemented by sensitive earth fault relays.
Usually, due to the difference in the magnetising currents of the upper and the lower CTs
(which is commonly the case), the current through the operating coil will not be zero even

6
Power System Protection & Switchgear (EE3203) Lecture Module – 2 by Dr. A. De, EE Dept., IIEST
under normal loading conditions or during external fault conditions. Therefore, to provide
stability on external faults, bias coils (restraining coils) are provided. The relay is set to
operate, not at a definite current but at a certain percentage of the through current. To
obtain the required amount of biasing, a suitable ratio of the restraining coil turns to
operating coil turns is provided. High-speed percentage-differential relays having variable
ratio or percentage slope characteristics are commonly used for protection of modern
synchronous generator. The setting of the bias coils varies from 5% to 50% and that of the
relay coil (operating coil) from 10% to 100% of the full load current.
In case of stator faults, the tripping the main circuit breaker to isolate the faulty generator is
not sufficient to prevent further damage, as the generator continues to supply power to the
fault until its field excitation is stopped. Therefore, the percentage differential relay initiate
several auxiliary relays, which in addition to tripping the main circuit breaker, also trip the
field circuit breaker, shuts down the prime mover and operates an alarm.

iv. Percentage Differential Protection for Alternators with Restricted Neutral

Grounding
When the neutral of a Y-connected is solidly grounded, it can be possible to provide
protection to 100% of the winding of the generator against ground faults. However, if the
neutral is grounded through resistance or reactance to limit ground fault currents., it is not
possible to protect the complete winding against ground faults. The percentage of winding
that can be protected depends on the value of the neutral grounding resistor and the relay
setting. The usual practice is to protect 80 - 85% of the generator winding against ground
fault. The remaining 15-20% of the winding length from neutral end remains unprotected.

Fig.5 Percentage Differential Protection of Y-connected Alternators with

restricted neutral grounding
In Fig.5 shows a Y-connected alternator with neutral grounded through resistance. If a
ground fault occurs at the point F of the generator winding, the voltage VFN is available to
circulate the ground-fault current If through neutral grounding resistor Rn. If the fault point
F is nearer to the neutral point N, the forcing voltage VFN will be relatively less. Consequently,
the ground fault current If will also reduce. It is not practicable to keep the relay setting too

7
Power System Protection & Switchgear (EE3203) Lecture Module – 2 by Dr. A. De, EE Dept., IIEST
sensitive to sense the ground fault currents of such small magnitudes, as it may lead to mal-
tripping of the relay due to slight mismatch of the CT currents (due to CT ratio mismatch,
difference in CT characteristics or due to CT saturation) and also during external faults. To
avoid such relay mal-tripping, common practice is to protect only about 80-85% of the
generator winding against phase to ground faults and to leave the 15-20% portion of the
winding from neutral end unprotected. It is however worth mentioning that, near the neutral
end of the winding, the voltage stress on insulation is much less and therefore, phase to
ground faults are less likely to occur near neutral end of the winding.
In Fig. 5, let us assume that p% of the winding from the neutral remains unprotected. Then
(100 – p) % of the winding from line end can be considered as protected. The ground fault
current If for fault occurring at the point F can be expressed as:
𝑉𝐹𝑁 𝑝 1 𝑉×𝑝
𝐼𝑓 = =𝑉× × =
𝑅𝑛 100 𝑅𝑛 100 𝑅𝑛
where V is the line to neutral voltage and Rn is the neutral grounding resistance. For the
operation of the relay, the magnitude of this fault current must be greater than the relay
pick-up current.

v. Percentage Differential Protection for 3-Phase Y-connected Alternators (with only

4 terminals brought out) – Restricted Earth Fault Protection

When the neutral connection is

made internally (within the
generator) and only the neutral
terminal is brought out, the
percentage differential
protection can still be provided,
but in a modified way, as shown
in Fig.6. This scheme can protect
the generator winding only
against ground faults. But, it
cannot protect the windings
against phase-to-phase faults.
This is one form of restricted
earth fault protection because
the relays operate only for
ground faults within the
protected windings.
Fig.6 Percentage Differential Protection for 3-
Phase Y-connected Alternators with only 4-
terminals brought out

8
Power System Protection & Switchgear (EE3203) Lecture Module – 2 by Dr. A. De, EE Dept., IIEST
vi. Problems on Generator Differential Relay
#1. An 11 kV, 100 MVA alternator is grounded through a resistance of 5 W. The CTs have
a ratio 1000/5. The relay is set to operate when there is an out of balance current of 1
A. What percentage of the generator winding will be protected by the percentage
differential scheme of protection?

#2. An 11 kV, 100 MVA alternator is provided with differential protection. The
percentage of winding to be protected against phase to ground fault is 85%. The relay
is set to operate when there is 20% out of balance current. Determine the value of the
resistance to be placed in the neutral to ground connection.

9
Power System Protection & Switchgear (EE3203) Lecture Module – 2 by Dr. A. De, EE Dept., IIEST
1.1 (b) Protection against stator inter-turn faults
Since Longitudinal Percentage Differential Protection cannot detect stator interturn faults,
other protection schemes, as discussed below are generally employed for protection of
generators against stator inter-turn fault.
i. Transverse Percentage Differential protection
This type of protection is used for generators having parallel windings with separately
brought out terminals. The stator coils of modern large steam turbine driven generators
(turbo generators) usually have only one turn per phase per slot and hence they do not
need interturn fault protection. Only hydro generators having parallel windings in each
phase employ such protection. This scheme is also known as split-phase protection. Fig. 7
shows a split-phase protection scheme for inter-turn fault protection in parallel-winding
generators. Since the parallel windings carry identical currents under normal condition,
there is no current through relay operating coil. However, in the event of any inter-turn short-
circuit, out of balance current of that phase will pass through the operating coil of the relay.

Fig. 7 Transverse percentage differential protection for generators having parallel

windings in each phase

10
Power System Protection & Switchgear (EE3203) Lecture Module – 2 by Dr. A. De, EE Dept., IIEST
ii. Inter-turn fault protection based on Zero-Sequence component of current
If generators do not have parallel windings (single winding per phase), or if access to parallel
windings is not available, a method based on zero-sequence voltage measurement can be
employed for the protection against stator interturn faults. Figure 8 shows the schematic
diagram of interturn protection based on zero-sequence voltage measurement. The zero-
sequence voltage does not exist during normal conditions. If one or more turns of a phase
are short circuited, the generated emf contains zero-sequence component due to
unbalancing. A voltage transformer as shown in the Fig. 8 is employed to extract zero-
sequence component from the generator voltage. The secondary winding of the voltage
transformer is in open-delta connection to allow the zero-sequence component of the
current to flow through the operating coil of protective relay. A filter circuit is provided to
extract the third harmonic component from the secondary output of the voltage transformer
output and apply it as the relay bias.

Fig. 8 Interturn protection of generator using zero-sequence voltage

1.1(c) Stator-overheating protection

Stator overheating may be caused by the failure of the cooling system, overloading or core
faults like short-circuited laminations and failure of core bolt insulation. Modern generators
employ two methods to detect overheating both being used in large generators (above 2
MW). In one method, the inlet and outlet temperatures of the cooling medium (hydrogen/
water) are compared for detecting overheating. In the other method, the temperature
sensing elements (Embedded Temperature Detectors or ETDs) are embedded in the stator
slots to sense the temperature. When the temperature exceeds a certain preset maximum
temperature limit, the relay sounds an alarm. The ETD may either be thermistor,
thermocouple or resistance temperature indicators. They are embedded in the stator slots
at different locations. These elements are connected to a multi-way selector switch which
checks each one periodically and operates an alarm relay, if high temperature is detected.

11
Power System Protection & Switchgear (EE3203) Lecture Module – 2 by Dr. A. De, EE Dept., IIEST
1.2 Rotor Protection
1.2 (a) Field Ground Fault protection:
Since the field circuit is operated ungrounded, a single ground fault does not affect the
operation of the generator or does not cause any damage. However, a single rotor fault to
earth can increases the voltage stress on the field winding insulation. Subsequently, if any
stator transient induces extra voltage in the field winding, the probability of the occurrence
of the second ground fault can increase.
If a second ground fault occurs, a part of the field winding is bypassed, thereby increasing
the current through the remaining portion of the field winding. This causes an unbalance in
the air-gap fluxes, thereby creating an unbalance in the magnetic forces on opposite sides of
the rotor. The unbalanced magnetic forces can make the rotor shaft eccentric and may also
cause vibrations. Even if the second ground fault does not bypass enough portion of the field
winding to cause magnetic unbalance, the arcing at the fault may cause local heating, which
slowly causes deformation of the rotor leading to eccentricity and vibration.
Figure 9 shows the schematic diagram of rotor earth-fault protection. A dc voltage is
impressed between the field circuit and earth through a polarised moving iron relay.
Circulating current flow is detected by the sensitive relay coil, whenever a rotor ground-fault
occurs and continuity of current path through ground is created.
It is not necessary to trip the machine when a single field earth fault occurs. Usually, an alarm
is sounded. Then immediate steps are taken to transfer the load from the faulty generator
and to shut it down as quickly as possible to avoid further problems. In case of brushless
machines, the main field circuit is not accessible. If there is a partial field failure due to short-
circuiting of turns in the main field winding, it can be detected by an increase in the field
current magnitude.

Fig. 9 Rotor Earth-Fault Protection

12
Power System Protection & Switchgear (EE3203) Lecture Module – 2 by Dr. A. De, EE Dept., IIEST
1.2 (b) Loss of Excitation Protection
When the excitation of a generator is lost, it speeds up above synchronous speed and
operates as an induction generator. Cylindrical rotor generators do not have damper
windings and hence they are not suitable for such operation. The rotor is overheated quickly
due to heavy induced currents in the rotor core. The rotors of salient pole generators are not
overheated because they have damper windings which can carry the induced currents. The
stators of both salient and non-salient pole generators are overheated due to the
magnetising current drawn by the machine from the system. The stator overheating does not
occur as quickly as rotor overheating. A large synchronous generator, when turn into an
induction generator may upset the system stability because it draws heavy reactive power
from the system instead of supplying reactive power to the system, when it was runs as a
synchronous generator. A machine provided with a quick-acting automatic voltage regulator
and connected to a very large system may however run for several minutes as induction
generator without harm.
A protective scheme employing offset mho relay or directional impedance relay having
characteristic as shown in Fig. 10. is used to detect excitation failure in large modern
generators. When the generator loses its excitation, the locus of the equivalent generator
impedance moves from the first quadrant to the fourth quadrant. This kind of locus shift is
observed only in case of excitation failure. The relay in such condition trips the field breaker
and the generator is disconnected from the system.

Fig. 10 Loss of Excitation protection using impedance relay

13
Power System Protection & Switchgear (EE3203) Lecture Module – 2 by Dr. A. De, EE Dept., IIEST
1.3 Miscellaneous protection
(a) Overvoltage Protection
Overvoltage may be caused by a defective voltage regulator or it may occur due to sudden
loss of electrical load on generators. When a load is lost, there is an increase in speed and
hence the voltage also increases. In case of a steam power station, it is possible to bypass the
steam before the speed reaches a limit above which a dangerous overvoltage can be
produced.
In thermal power stations, the automatic voltage regulator controls the overvoltages which
is associated with overspeed. In hydro-stations it is not possible to stop or divert water flow
so quickly and overspeed may occur. Therefore, overvoltage relays are provided with hydro
and gas-turbine sets. But overvoltage relays are not commonly used with turbo-alternators.

(b) Overspeed Protection

A turbo-generator is provided with a mechanical overspeed sensing device. The speed
governor normally controls its speed. It is designed to prevent any speed rise even with a
100 per cent load rejection. An emergency centrifugal overspeed device is also incorporated
to trip emergency steam valves when the speed exceeds 110 percent. Large turbo alternators
are sometimes provided with overspeed relays.
In USA out of step tripping relays are used which cut off steam when the generator is 180°
out of synchronism and has slipped one pole. In UK a sensitive fast acting power relay is used
to determine whether the power output falls below a certain value or reversed.
Severe electrical faults also cause overspeed and hence main circuit breaker, field circuit
breaker and the steam turbine valves are tripped simultaneously. As water flow cannot be
stopped quickly, hydro generators are provided with overspeed protection. The setting of
overspeed relays for hydro generators is 140%. Overspeed relays are also provided with gas-
turbine sets.

(c) Protection against Motoring

When the steam supply is cut off, the generator runs as a motor. The steam turbine gets
overheated because insufficient steam passes through the turbine to carry away the heat
generated by windage loss. Therefore, a protective relay is required for the protection of the
steam turbine. Generally, the relay operates when power output falls below 3%. Sensitive
reverse power relay is available which has an operating setting of as low as 0.5% of the
generator’s output. Hydro generators also require protection against motoring. Cavitation
problems arise in water turbines due to low water flow. Protection in such case is provided
by reverse power relay having an operating setting of 0.2 to 2% of the rated power. Diesel
sets and gas turbines require 25% and 50% setting respectively.

14
Power System Protection & Switchgear (EE3203) Lecture Module – 2 by Dr. A. De, EE Dept., IIEST