Directional OC and EF Protection (2)
Directional OC and EF Protection (2)
Directional Over Current & Earth Fault Protection
- Dr. G. Pradeep Kumar, Ph.D, MIET, MIEEE
1. Introduction
Directional protection is set to operate for faults in one direction only. There are
many types of directional protection elements, such as distance protection,
directional power protection, directional over current, directional earth fault, etc.
Understanding how a relay is made to operate for faults in a specific direction is
vital to its proper application.
In this document the application of directional over current and earth fault
protection, their operating principle and factors that influence their direction of
operation is explained.
Fig. 2.1: Simple radial system with protection relays and their primary zones marked
This is referred to as a “Radial System” because the source is located only on one
side of the protection relay. In such systems fault current can only flow in one
direction. For example, for a fault at ”F1” in the feeder between the “Location A”
& “Location B”, only the relay R1 will measure the fault current. However for a
fault “F2”, which is in the feeder beyond the “Location C”, the relays R1, R2 & R3
will measure the fault current.
The fault “F2” lies in the “Primary Zone” of the relay “R3”. This means that for
this fault, the relay “R3” should be the first (and only) relay to operate and trip
the circuit breaker “CB3”. Once the circuit breaker “CB3” is open, the fault is
isolated from the system. However for some reason if the relay “R3” fails to
operate or the circuit breaker “CB3” fails to open after receiving the trip
command from the relay “R3”, then the protection system at the “Location C” is
considered to have failed to operate. This is usually referred to as loss of
“Reliability” of the protection system at that location. Under such a circumstance,
the relay “R2” will have to operate and trip the circuit breaker “CB2”. The relay
“R2” is now said to have operated as a back-up protection for the protection
system at “Location C”. The relay “R2” should only operate after confirming that
the protection system at the “Location C” has failed. One way of ensuring this is
by current/time coordinating the settings of the protection relays.
Thus for a fault “F2”, with the relay settings properly current/time coordinated
the operating time of the relays that see the fault will be as follows,
Operating time of R1 (tR1) > Operating time of R2 (tR2) > Operating time of R3 (tR3)
Consequently the relay R1 is referred to as “Upstream” to the relay “R2” and the
relay “R2” is referred to as “Upstream” to the relay “R3”. Similarly the relay R2 is
referred to as “Downstream” to the relay “R1” and the relay “R3” is referred to as
“Downstream” to the relay “R2”. Since the system is “Radial”, it is possible to
clearly designate a particular relay as being “Upstream” or “Downstream” to
another relay in the system. This is because the source is present only on one
side of the protection relay.
An electrical power system ceases to be radial for a protection relay when the
relay will see currents for faults on either side of its location. There are various
system configurations that can be classified as a non-radial power system. Figure
2.2 shows some of the common non-radial system configurations.
In all the systems shown in Figure 2.2, the relays at the locations marked by the
symbol “ ” will see fault currents for faults on either side of its location. It is not
possible to designate these relays as upstream or downstream. On the other
hand the same relay can be upstream or downstream to another relay depending
on the fault location. However the relays at the location marked by the symbol “ ”
will only measure currents for faults on one side of the location.
Let us take the example of a single source ring feeder system to understand this.
Figure 2.3 shows such a system with the protection location marked. For a fault
in one of the feeders forming the ring system, say on “FEEDER 2”, there will be
two fault current paths. One part of the fault current, “I1” will flow from the bus
at the “MAIN SUB-STATION” through the circuit breakers “CB2”, “CB3” and “CB5”.
The second part of the fault current, “I2” will flow from the same bus at “MAIN
SUB_STATION” through the circuit breakers “CB15”, “CB14”, “CB12”, “CB11”,
“CB9”, “CB8” and “CB6”. The relative magnitude of the two currents will depend
on the relative impedances of the two paths.
Fig. 2.2: Various non-radial system configurations
Fig. 2.3: A ring feeder system with current flow shown for a fault in FEEDER 2
For this fault, we would want only the circuit breakers “CB5” and “CB6” only to
trip and isolate the “FEEDER 2” from the system. To achieve this, the operating
time of the relays in the two paths should be as follows,
In the path of current I1: tR2 > tR3 > tR5 (1)
In the path of current I2: tR15 > tR14 > tR12 > tR11 > tR9 > tR8 > tR6 (2)
On the other hand if the fault is on “FEEDER 3”, the fault current flowing through
the circuit breakers “CB6” and “CB8” will be opposite to the direction of the
current flow through them when the fault was in “FEEDER 2”. The fault current
path for this situation is shown in Figure 2.4. For the fault in “FEEDER 3”, the
circuit breakers “CB8” and CB9” only should trip. To achieve this, the operating
time of the relays in the two paths should be as follows,
In the path of current I1: tR2 > tR3 > tR5 > tR6 > tR8 (3)
In the path of current I2: tR15 > tR14 > tR12 > tR11 > tR9 (4)
Fig. 2.4: A ring feeder system with current flow shown for a fault in FEEDER 3
On comparing the conditions (2) and (3), we can see that there is a contradicting
requirement for the operating times of the relays “R6” and ‘R8”. As per condition
(2), relay “R6” should be set with its operating time lower than that of the relay
“R8”, whereas as per condition (3), relay “R6” should be set with its operating
time greater than that of the relay “R8”. It is impossible to meet both the
conditions at the same time.
On the same lines if we compare the operating time requirement of relays for
fault in “FEEDER 1” and “FEEDER2”, we will have similar contradicting time
setting requirement for the relays “R3” & “R5”. Similar situations will arise for
the relays “R9” & “R11” and relays “R12” & “R14” also. In short, for such a
system, it will not be possible to clearly define the upstream and downstream
relays, which is mandatory to decide which relay has to be coordinated with
which relay.
Using directional relays will solve this problem. Directional relays are set to
respond to fault current flow in one direction only. When the fault current flows in
a direction opposite to its set operating direction, this relay will not respond. In
the present example of the ring system, the relays “R3”, “R5”, “R6”, “R8”, “R9”,
“R11”, “R12” and“R14” are to be made directional. They should be set to operate
only when the current flows from the bus to the feeder at their respective
locations. The operating direction of the relays is marked by the direction of the
arrow beside these relays shown in Figure 2.5.
Fig. 2.5: A ring feeder system with directional relays
With this arrangement we can clearly define the upstream and downstream relays
and their operating time will have to be set as follows,
tR2
>
tR5
>
tR8
>
tR11
>
tR14
(5)
tR15 > tR12 > tR9 > tR6 > tR3 (6)
Operating time of the relay “R1” (tR1), shall be set more than the operating times
of the relays “R2” (tR2) and “R15” (tR15). The relays “R1”, “R4”, “R7”, “R10” and
“R13” can be non-directional relays (relays the will operate for fault current
flowing in either direction). However at these locations, there is only one possible
direction for fault current to flow. That is the reason why a non-directional relay
can be used at these locations.
With the use of directional relays, the primary zone of the protection relays will
be in the direction in which they are set to operate (how a relay is made to
operate only for fault current flowing in one direction is explained in the later part
of this document). Figure 2.6 shows the primary zones of the relays applied for
the ring system under discussion. For easy understanding the primary zones of
the relays R1, R2, R4, R5, R7, R8, R10, R11, R13 & R14 and R1, R3, R4, R6, R7,
R9, R10, R12, R13 & R15 are shown separately (a) and (b) respectively.
From this figure it can be understood that for a fault in a particular feeder, the
relays at either end of the feeder would operate first and trip their respective
breakers. For e.g., a fault in “FEEDER 2” the relays “R1”, “R2”, “R5”, “R6”, “R9”,
“R12” and “R15” will see the fault current flowing in their operate direction. Of
these relays, “R2” & “R5” will see the fault current “I1” and relays “R6”, “R9”,
“R12” & “R15” will see the fault current “I2”.
Among the relays “R2” & “R5” which see the fault current “I1”, the relay “R5” will
operate first as its operating time is set lower than the operating time of “R2”.
Similarly among the relays “R6”, “R9”, “R12” & “R15”, the relay “R6” will operate
first as it has the lowest time setting of all the relays seeing the fault current “I2”.
The relay “R1” will see the total fault current “I1+I2”. However its operating time
is set to highest of all the relays. Thus we can see that selective tripping can be
achieved by providing directional relays and setting them as explained above.
For the relay “R3” the current “I1” flowing through its location will be in the
direction opposite to its set direction of operation. Hence this relay will not pick-
up at all. Similarly for the relays “R8”, “R11” & “R14”, the current “I2” flowing
through their location will be in the opposite direction to their set operating
direction. These relays will also not respond to this fault current.
Fig. 2.6: Primary zones of the directional relays for the ring system
3. Giving a Relay its Direction
The direction for which the relay will operate is primarily governed by,
Once we understand the need for directional relays, the next step is to make the
relay to only operate for current flowing in one direction. This cannot be achieved
by measuring the current alone. The directional relays need to have a reference
signal to determine the direction of current flow. This reference signal is referred
to as the “Polarizing signal”. The direction of current flow is determined by
measuring the phase angle of the current with respect to the polarizing signal.
For most applications of directional over current and earth fault protection, the
polarizing signal is voltage. Hence in addition to the current input, voltage input
also has to be wired to these relays. The appropriate choice of the voltage for
over current and earth fault relays is described at the later part of this document.
The “Relay Characteristic Angle” setting (RCA) decides the forward and reverse
zones for the relay. This setting is also referred to as “Maximum Torque Angle”
(MTA), which is more relevant in the context of electromechanical relays.
A typical V-I plane is shown in Figure 3.1. In this figure, the polarizing and
operating signals marked. The RCA is marked as “θ”. The RCA line is at this
angle “θ “ from the polarizing signal line. The angle between the directional line
(which divides the V-I plane into “Forward” and “Reverse” areas) and the RCA
line is marked as λ. In Figure 3.1,
λ
is at
90°, which is normally the case for most
applications. However in some relays, it is possible to set this angle λ to a value
less than 90°. The phase angle between the polarizing signal line and the
directional line will be equal to (λ + θ)°. Kindly note that angle measurements in
counter-clockwise direction is usually taken as positive. Correct signs will have to
be used for the phase angles following a common convention for sign.
Fig.
3.1:
Voltage-‐Current
(V-‐I)
plane
showing
the
polarizing
signal
&
RCA
line
with
the
forward
and
reverse
direction
area
highlighted
If we define the angle between the fault current and the polarizing signal as Φ,
then the condition for defining forward and reverse faults will be,
For e.g., if the, RCA (θ) = -45°
(RCA line lagging Polarizing Signal, hence the
negative sign) and the directional line angle (λ) = 90°, then the relay will
consider the fault as FORWARD if the angle of the fault current with respect to
the polarizing voltage (Φ) is less than 45°
and also more than -135°. On the other
hand if Φ
is more than 45°
and also less than -135°, then the relay will conclude
the fault is in the reverse direction.
Normally, the forward direction is set as the operating direction for the relay and
the reverse direction will be the restraining direction.
Kindly note that in the above discussions we have not defined what signal is used
as the “Polarizing Signal”. This will be discussed in the later part of this document.
Once the polarizing signal and relay characteristic angle settings are chosen, the
“FORWARD” and “REVERSE” areas are clearly defined with respect to the
polarizing signal based on the RCA (θ) and directional line angle (λ) settings.
The next important factor that affects the operating direction of the relay is the
polarity of the CT connections. Every relay will have its forward current flow
direction defined at the design stage itself. This will be clearly explained in the
relay literature (usually by showing the typical connection diagrams for the relay).
The current input wiring to the relay has to be done in such a way to ensure that
for the faults in the direction in which the relay is desired to operate, the current
flow in the relay should be as per the relay’s designed forward direction.
For e.g., let us say the current input terminals of a relay are “C1” & “C2”. And as
per the relay design, current flowing into the relay at terminal “C1” and out of the
relay at terminal “C2” is measured by the relay as positive (forward).
Then
the
terminal “C1” can be termed as the “+” terminal or the “IN” terminal and the
terminal “C2” can be termed as the “-“ terminal or the “OUT” terminal.
Figure 3.2a shows the CT secondary and relay current input connections. With
this connection, when the fault current flows in the primary system from “Bus Q”
to “Bus S”, the current will flow into the relay at terminal “C1” and come out of
the relay from terminal “C2”. The direction of current flow in the system, in the
CT secondary and in the relay is shown for a fault in the feeder between Bus Q
and Bus S.
Figure 3.2b shows the corresponding phasor diagram on the V-I plane. In this
diagram, the faulted phase voltage is used as the Polarizing Signal. The phase
angle of the fault current with respect to the corresponding phase voltage (ϕ) will
depend on the impedance angle of the fault loop. It can be seen from this
diagram that the fault current falls in the “OPERATE AREA” of the V-I plane and
hence the relay will operate for this fault. Thus we can conclude that the operate
direction of the relay is from Bus Q to Bus S.
Fig. 3.2: Directional relay connection & phasor diagram for relay operating direction from
bus Q to S
On the other hand, if the CT input to the relay is reversed as shown in Figure
3.3a, then for a fault between the buses Q & S, the fault current falls in the
“RESTRAINT AREA” of the V-I plane. This is shown in the corresponding phasor
diagram in Figure 3.3b. Hence now the relay will not operate for this fault.
However for a fault in the feeder between Bus P and Bus Q, the fault current flow
in the relay will be as per its normal direction. This fault current will fall in the
“OPERATE AREA” of the V-I, causing it to operate. Thus for the CT connection
shown in Figure 3.3a, the operating direction of the relay will be from Bus Q to
Bus P (shown by the arrow beside the relay in Figure 3.3a).
Fig. 3.3: Directional relay connection & phasor diagram for relay operating direction from
bus Q to P
In some of the modern numerical relays, a setting is provided for reversing the
CT input polarity. If such a setting is available, then this can also be used to
change the directional of the relay.
In the examples illustrated in Figures 3.2 & 3.3, the “Forward Area” in the V-I
plane is defined as the operating zone for the relay. In the earlier generation
electromechanical and static electronic relays, forward direction is always defined
as the operating zone. This could not be changed. In these relays, to change the
operating direction, the current input to the relay had to be reversed.
However in most of the modern numerical relays, settings are provided to define
the operating direction of the relay. In these relays it is possible to set the relay
to operate for “Forward” or “Reverse” faults.
For phase over current protection, usually the operating quantity is the phase
current. Typically a three-phase relay will have overcurrent protection element for
each of the three phases.
Normally one would expect the phase directional over current protection to use
its own phase voltage as the polarizing signal. However this practice is not
followed as the faulted phase voltage can drop during faults. For very close-up
faults (both in forward and reverse directions), the faulted phase voltage will be
almost zero. Thus the faulted phase voltage will not be a reliable polarizing signal.
Most relays use quadrature polarizing (also called as cross-polarization). The
Phase “A” directional over current protection element of a relay using this
polarization technique will use “BC” voltage as the polarizing signal. This is also
called as 90° connection because the polarizing signal will be at 90° to the pre-
fault voltage of the faulted phase.
The operating and polarizing signals of the three-phase over current elements of
a directional over current protection is shown in Table 4.1.
Table 4.1: Operating and polarising signals for a quadrature polarised directional OC relay
Operating Polarizing
Element
Current Voltage
Phase A IA VBC
Phase B IB VCA
Phase C IC VAB
There are few other types of polarization options available for directional over
current protection polarization (such as 30° connection, 60° connection, etc.).
However the quadrature polarization is the most commonly used one due to its
suitability for most applications.
Some numerical relays also complement the quadrature polarization with memory
polarization for three phase faults (when all the three phase voltages can be low).
In this case, the relay would use the pre-fault voltage for polarization.
Information on which signal is used as the polarizing signal by the relay is very
important. The RCA angle setting in the relay depends on the polarizing signal
used.
Let us take the example of a commonly used quadrature polarized over current
relay. If the phase current is expected to lag the faulted phase voltage by an
angle ϕ=-60°, then we would want the RCA line to be at this angle (-60°) with
respect to the faulted phase. The polarizing signal will be at -90° to the faulted
phase voltage. Hence the RCA setting in the relay will have to be set to,
For a directional line angle λ
= 90° (either fixed by design or a setting in the
relay), the forward and reverse area in the V-I plane will be as shown in Figure
4.1.
Fig. 4.1: Phasor diagram for quadrature polarized relay
From the above discussion, it is clear that the RCA setting would depend on the
“Polarizing signal” used in the relay and the expected angle of the fault current
for a forward fault. For a quadrature-polarized directional over current relay, a
general guideline for setting the RCA is,
• RCA = 30° for plain feeder with system grounding behind the relay location
• RCA = 45° for transformer feeders or feeders with system grounding in front
of it
voltage may not be an ideal choice for directional earth fault protection when
applied for protection of double circuit lines. In this application the residual
voltage measured be affected by the zero sequence mutual induction from the
parallel lines. In such applications it is common to negative sequence voltage as
the polarizing signal.
However for a majority of the applications, residual voltage is the most widely
used polarizing signal for directional earth fault protection. This residual voltage is
either derived by connecting the PT secondary as “Broken Delta” or internally
calculated by the numerical relay from the three phase to ground voltages. Which
ever the case may be, if three-phase PT is being used then it should be of 5-Limb
construction. Alternatively three single phase PTs can be used. Care should be
taken to also ensure that the star point of the PT primary winding is grounded. In
case interposing PTs are used to derive the broken delta voltage, then star point
of the primary of the interposing PT should be directly connected to the star point
of the main PT’s secondary winding.
Under normal operation the residual voltage will be negligible. During any phase
to ground fault, the residual voltage will appear and will be at an angle of 180°
with respect to the faulted phase voltage. Hence some relays use residual voltage
shifted by 180° (–VN) as the polarizing voltage so that it becomes in-phase with
the faulted phase voltage. Since the RCA setting is the angle between the
polarizing signal and the operating current (which will be the residual current in
this case), clear understanding on what signal is used for polarizing is vital to
setting the correct RCA.
On the other hand if the relay uses VN (which will be at 180° to the faulted phase
voltage) as the polarizing signal, the RCA setting in the relay will have to be set
to,
Figure 4.2 shows the characteristics of the directional earth fault protection in the
V-I plane. This figure shows the RCA setting for both cases, when VN and –VN is
used as polarizing signal.
Fig. 4.2: Phasor diagram for residual voltage polarized directional EF relay
For a directional earth fault protection also, the RCA setting would depend on the
signal used as the “Polarizing signal” and the expected angle of the fault current
for a forward fault.
For a directional earth fault protection using –VN as the polarizing signal, following
is a typical setting for the RCA,
6. Conclusion
Directional relays are required to ensure selectivity in non-radial systems. The
operating direction of the relays depend on,