Paper On Protection Coordination
Paper On Protection Coordination
Paper On Protection Coordination
Selectivity can be achieved with devices that are inherently selective. That is, they operate only on faults within
their zone of protection and do not ordinarily sense faults
outside that zone. When a fault occurs inside the zone, the
device typically responds instantaneously and trips breakers
on the edge of the zone that are associated with short circuit
contributors into the zone. If a fault occurs outside the zone,
fault current may flow through the zone but the device will
not operate for this through-fault. Examples of inherently
selective systems are current differential relays (typically applied on buses, motors, generators, and transformers), pilot
wire relays, and transformer sudden pressure relays. While
these devices offer the best in circuit protection, they can
add significant cost to the power system. These protective
devices tend to be much more expensive than overcurrent
devices, but the cost may be worthwhile if it reduces the
amount of damage to a critical piece of equipment or reduces
the amount of lost production resulting from a fault. For
this reason they tend to be used sparingly, usually when the
protected equipment is critical, or if an extended outage to
repair the system would cause significant economic loss.
In most other cases, a simple overcurrent device in the
form of a circuit breaker, fuse, or overcurrent relay is used.
In a properly designed power system, these devices can
provide selective coordination under most circumstances
at a reasonable cost. They are currently the workhorses of
electrical protection.
When evaluating and comparing time-current curves,
you should keep in mind the following. It is assumed that
the fault current is constant up until the time that the fault
is cleared; the graphical evaluation of curves may not necessarily reflect the true characteristics when the fault current
is varying. In the real world, the fault may vary depending
on the contact area of the fault, the degree of arcing that
is occurring, and a number of other factors. In most cases,
the fault current tends to escalate upwards from its initial
value, but this is not always the case.
In the presence of a current of varying magnitude, devices
that use different monitoring or time delay techniques may
not react in the same way, especially if the current falls below
the pickup setting of the device. For example, consider what
may happen when a circuit breaker with a bimetal element
is responding to the same current as an induction disk time
overcurrent relay. If the current drops below their respective pickup settings momentarily, different things happen
within the each device. In the circuit breaker, the bimetal
may cool slightly, but if the current restores itself quickly,
the bimetal will continue to heat and will probably trip out
in much the same time as the curve indicates. In the case of
the induction disk relay, the disk will turn backward toward
its reset position. Whether it gets all the way back to this
position depends on how low and for how long the current
drops. The disk could turn backwards quite a ways and start
up again in a different part of the delay cycle in comparison
to the breaker. The relay then may trip out at a longer time
delay than the breaker as indicated by the curves for this
varying fault condition.
www.netaworld.org
www.netaworld.org
www.netaworld.org
www.netaworld.org
For this example, well assume that the 75 horsepower motor is the largest one in this motor group, and the feeder
breaker settings can be coordinated to the curves we have
already plotted.
The feeder breaker has a fully adjustable trip device. It
has settings for the long time current setting, long-time
delay, short-time and its corresponding delay setting, and
instantaneous. The feeder breaker may also have a ground
fault function; however, it is customary to show a separate
plot for ground fault devices since they seldom can coordinate with phase overcurrent settings. Each of the phase
current settings must coordinate with either the maximum
load characteristic or the maximum protective device characteristic, whichever is greater.
The plot of the inrush current of the motor control
center is a combination of the total load current plus the
inrush of the largest motor, assuming the motors are started
sequentially and not simultaneously. The total load current
should be the full load current of all the motors expected
to be in operation simultaneously. This is plotted from the
top of the graph down to the starting current time of the
largest motor. The total inrush combines the locked rotor
current of the largest motor with the full load current of all
the other motors expected to be in operation. A conservative
calculation would add the full load current of all the other
connected motors to the locked rotor current of the largest
motor. This combination load profile is plotted similarly to
the individual motor load profile.
The sequence in determining circuit breaker settings is
somewhat important. In most cases, the short circuit functions are a function of the current setting. The current setting
needs to be plotted first so that you can then determine the
range of settings available for the short time (if available)
and/or instantaneous functions.
The motor control center bus and the feeder cable should
have approximately the same ampacity. The current setting
of the feeder breaker is usually set to the same value as the
feeder ampacity which is the current rating of the feeder
cable or the motor control center bus, whichever is less. This
is the maximum setting that the NEC permits and allows
for future expansion if the feeder is not already fully loaded.
A long-time delay is then selected which places its band so
that there is daylight between its minimum tolerance limit
and the total clearing time of the overload curve.
The short time setting of the feeder must coordinate with
the maximum inrush current of the motor control center
and the highest set short-circuit device in the motor control
center. In this case, the motor circuit protector setting of
the largest motor has the highest characteristic. The correct short time setting is the one that allows the short time
curve to be completely to the right of the instantaneous
setting of the motor circuit protector. The short time delay
is the lowest possible delay that allows daylight between
its lower tolerance limit and the total clearing time of the
motor circuit protector.
Conclusions
Using time-current curves as a graphical technique to illustrate selective coordination makes it easy to demonstrate
whether or not selectivity has been obtained by the device
settings and whether they adequately protect the distribution equipment. Once you become accustomed to reading
these curves the system evaluation can be done quickly.
References
[1] M. E. Valdes, I. Purkayastha, and T. Papallo, The Single-Processor
Concept For Protection And Control Of Circuit Breakers In LowVoltage Switchgear, IEEE Trans. Ind. Appl., vol. 40, no. 4, pp. 932
940, July-Aug. 2004
Gary H. Fox, PE, received his BSEE from California Polytechnic
State University, San Luis Obispo in 1978. He has been employed
by General Electric Company for 31 years. His current assignment
is a Senior Specification Engineer for GE Consumer & Industrial in Concord, CA, providing application and technical support
for power distribution/control equipment and lighting. Previous assignments included experience in field testing and maintenance of
power equipment, project management, and power system analysis.
Mr. Fox is a Senior Member of the IEEE Industry Applications (IAS) and
IEEE Power Engineering Societies. He has authored several IEEE papers
on protective relay application and surge protection, and was a recipient of
the IEEE Third Millennium Medal. He is actively participating in several
IEEE working groups responsible for developing recommended practices
for the design of industrial and commercial power systems.