Protection of Buses

PROTECTION OF BUSES
Since the buses are the points of connection between lines, transformers, and other
apparatus, their protection is essential. High-speed relays that are well coordinated
with other protective devices are essential in protecting buses at high-voltage
stations. The main problems of bus protection are discussed, along with several
solutions that have been widely adopted in recent years.
1 Overcurrent relays
In some applications, overcurrent relays may be used for bus protection. The
problems of coordination with nearby line and transformer relays, however, make
this application relatively difficult. In an attempt to overcome these coordination
problems, the basic overcurrent differential relay connection of Figure 1 could be
used. This suffers from severe problems with unequal saturation of CT's, and
typically more sophisticated schemes must be used. The next section discusses the
problem of CT saturation, and then in succeeding sections three solutions are
discussed.
ONE PHASE SHOWN, OTHERS SIMILAR

Figure 1. Basic overcurrent differential connection. Simple but subject to incorrect
operation due to unequal CT saturation.
2 Differential relays and saturation of CT's
To illustrate the nature of the basic problem, consider the system shown in Figure
2, having a bus with four connections. Three of the connections are sources (or
lines connected to sources) of fault current, while the fourth is a feeder circuit that
supplies a negligible fault current. For a bus fault current of 5000 A, we suppose
that the first two lines provide 2000 A each and the third supplies 1000 A. This
fault is in the trip zone for the bus differential relay. The fault location may be just
outside the bus zone on the feeder, and the feeder CT will see 5000 A in a direction
opposite to that of the other CT's. Assuming a CT ratio of 1000:5 A, if no CT
saturation occurs, each secondary current will be proportional to the primary
current, and the differential relay will see 25 A for the internal (bus) fault and zero
for the external (feeder) fault. What is likely to happen, though, is that the feeder
CT will saturate on the external fault current, its magnetizing branch reactance will
drop to a small value, and its secondary current will be well below 25 A.
The bus differential relay will see a significant tripping current, and the race is on
to see if the bus differential relay will trip before the feeder relay. This problem is
severe, and a time delay for coordination on the bus differential relay invites
considerable damage to the busbars on a heavy fault. On the other hand, false
tripping of the bus differential relay can easily turn a close-in fault on a minor
feeder circuit into a major operating problem, since it will trip every connection to
the bus. Obviously, the problem must be addressed, especially at important
substations and generating station switchyards. We consider three of the schemes
that have been developed to handle this problem in the following sections:
the multi-restraint percentage differential relay, the linear coupler scheme, and the
high impedance differential relay.
Figure 2. Typical system consisting of a bus with three source connections and one
feeder
3 Multi-restraint percentage differential relay

This scheme uses conventional CT's, attempting to make up their lack of accurate
performance with a more complex relay. Almost invariably, a more complex relay
is more difficult to apply. On the other hand, no special types of instrument
transformers are required.
Figure 3. AC elementary diagram for multi-restraint differential relay Restraint
coils connected in pairs.
Figure 3 shows the multi-restraint relay AC elementary diagram. Note that the
restraint coils are in pairs opposite each other on three induction disks, and the
operating coil is on a fourth disk. All the disks are connected to a common shaft.
The restraint provided by the paired coils is such that currents in the same direction
through the coils restrain on the sum of the currents, while currents through the
coils in opposing directions restrain on the difference between the currents. These
coils always produce restraining torque. The top pair, then, restrains on the
difference between the circuit 1 current and the circuit 2 current (I1 -I2); the bottom
pair restrains on the difference between the circuit 3 current and the circuit 4
current (I3 - I4); the middle pair restrains on the difference between the sum of the
first two circuit currents and the second two (I1 + I2) - (I3 + I4). The operating
coils operate on the sum of all the currents (which is small for external faults,
except for false differential currents due to CT saturation).
An example will help make this clearer. Let the currents be I1 = 1, I2 = 2, I3 = 3, I4
= 4 (on the secondary side of the CT's) for an internal (bus) fault. The top and
bottom units provide restraint of 1 unit each, the middle provides restraint of 4
units, for a total of 6 units of restraint.
Meanwhile, the operating coil sees 10 units, so the relay trips.

For an external (close in) fault on circuit 1, the currents would be I1 = 9, I2 = 2, I3
= 3, I4 = 4, in the absence of CT saturation, and the top unit would restrain with 11
units, the bottom with 1 unit, and the middle with 0 units, for a total of 12 units of
restraint. The operating coil sees zero units. Suppose the circuit 1 CT saturates and
only produces I1 = 4. Then the top unit would restrain with only 6 units, the bottom
with 1, and the middle with 5, for a total of 12 units.
Meanwhile the operating coil sees the false differential current of 5 units, but the
restraint of 12 units overcomes this, so the relay does not trip.
Notice that the operating coil is connected to a saturating autotransformer. This

device presents a relatively high impedance to false differential currents, but not to
heavy fault currents (which drive the core into saturation, reducing the apparent
impedance). This device tends to force a more equal saturation of the CT's under no
fault conditions. It also shunts any DC component.
4 Linear couplers
Another method of preventing current transformer saturation from causing false
differential current on external faults, is to simply get rid of the iron-core current
transformer. A scheme that is applied consists of air-core mutual reactors, which
are wound on non-magnetic toroidal cores. These linear couplers are not current
transformers and cannot be used for backup overcurrent relays or for station
ammeters, etc. They are dedicated to the differential relay,
which is designed to operate with the couplers. Bushing couplers can be obtained
for dead-tank circuit breakers.
The linear coupler produces a secondary voltage of 5 volts per 1000 A of primary
current, and is to be connected in series with the couplers on other circuits, and in
series with the differential relay, as shown in Figure 4. The relay current is
determined by the net voltage developed by the linear couplers divided by the
circuit resistance. The impedance of a coupler is in the range of 2 to 20 ohms, while
the impedance of the relay is 30 to 80 ohms. The circuit resistance will be about 38
to 160 ohms, the internal fault shown will produce 25 V across the relay, so the
relay current will be between 0.16 and 0.66 A. The exact value of the impedance of
the couplers in use should be obtained from the manufacturer.
Figure 4. Linear couplers for bus protection. Connect the couplers in series, since
they produce a voltage on the secondary.
The mutual inductance (nominally 0.005 ohms) may vary by + 1%; an external
fault produces the worst case if the faulted phase coupler is off by -1% and all the
others are off by +1% (or vice versa). This would produce a net error of 2%. If the
relay is set to trip for a minimum internal fault of I, then an external fault of 50 I
could produce a false trip. For a 2:1 safety factor, the relay is usually set for no
more than a 25:1 ratio of maximum external fault to minimum internal fault.
This 25:1 ratio is not usually a problem on effectively grounded systems. (Defined
to be grounded systems that have ratio of zero sequence reactance to positive
sequence reactance of less than 3, and zero sequence resistance to positive
sequence reactance of less than 1; a phase to ground short circuit will then produce
at least 60% of the current of a three-phase short circuit.) On impedance grounded
systems, the ground fault current may be so low that the minimum internal fault
can not be protected by this relay. Then a separate ground relay set very sensitively
may be required. This sensitive ground relay may trip on heavy external faults, so
the manufacturer should be consulted for detailed application information.
5 High impedance relay

The third scheme considered here is the high-impedance relay loading conventional
current transformers, which should be toroidal construction. The toroidal
construction (C type, not T type), has distributed secondary windings that produce
a negligible secondary leakage flux, and so have very little secondary leakage
reactance. The connection is shown in Figure 5, where the CT's are paralleled as
close to each other as practical. The CT ratios must match each other.
The high impedance loading means that a very high voltage, on the order of 1000
V, will be developed across the relay (which is the overvoltage unit, device 59 in
the figure) by an internal fault. The varistor serves to limit the voltage to less than
1500 V, and the instantaneous overcurrent unit, device 50 in the figure, is set high
to catch very heavy internal faults. Note that the overvoltage unit is fed from a
tuned circuit that passes 60 Hz. This branch has an impedance of about 3000 ohms.
An external fault will produce zero voltage if none of the CT's saturate. If one or
more of the CT's saturate, the voltage across the relay will still be relatively low,
since the connection tends to force the false differential currents through the other
CT's instead of the relay.
Setting is based on the worst case of one CT completely saturated, the others not at
all. Calculate the voltage drop produced by the other CT's through the lead
resistance and the secondary winding resistance of the saturated CT. Set the relay at
twice the calculated value.
Figure 5. High-impedance relay for bus protection. CT's must be toroidal cores
with completely distributed windings, all of the same ratio.
6 Partial differential scheme

This scheme is to connect only the source circuits to a differential overcurrent
relay. To obtain coordination with feeders, the overcurrent relay must have a long
time delay. This scheme is really considered a backup relaying scheme today.
One application for the partial differential scheme is when the feeder circuits have
current limiting series reactors. Then the partial differential scheme can be used
with distance relays that are set to see into but not through the reactance of the
reactor.
Figure 6 shows a partial differential scheme on a bus with a tie circuit breaker
between sections. With the tie breaker open, the scheme is straight overcurrent, but
with breaker closed fault currents produce differential current only on faulted bus
segment (the other relay sees an external fault and restrains).
Figure 6. Partial differential scheme applied to bus with tie breaker. Overcurrent
relays can be used for this scheme if set with high tap and long time delay.

Protection of Buses

Uploaded by

Copyright:

Available Formats

Protection of Buses

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Protection of Buses

Uploaded by

Copyright:

Available Formats

PROTECTION OF BUSES

ONE PHASE SHOWN, OTHERS SIMILAR

3 Multi-restraint percentage differential relay

Meanwhile, the operating coil sees 10 units, so the relay trips.

Notice that the operating coil is connected to a saturating autotransformer. This

5 High impedance relay

6 Partial differential scheme

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.