PRM DesignGuideComplete v1

Section 1:
Introduction and Basic Principles
Bill Brown, P.E., Square D Engineering Services
Introduction
With the increasing sophistication of modern power systems, it is easy to overlook the fact that the basic function
of a power distribution system has been the same for over 100 years the safe, reliable distribution of power from
a source to the connected loads. This basic function has not changed, although the complexity of the loads
themselves, along with today's reliability and efficiency requirements, do make its realization more complex.
This guide discusses the basic considerations which must be taken into account in order to obtain an optimal
system design. Because the characteristics of each load, process, etc., served are unique, so too will each design
be unique in order to match the requirements imposed.
The purpose of this guide

This guide is intended to present the fundamentals of power system design for commercial and industrial power
systems. It is not designed as a substitute for educational background and experience in this area, nor is it
designed to replace the multitude of detailed literature available about this subject. It does, however, bring into
one volume much material which has previously been available only by referencing a number of different sources
with different formats and terminologies.
This guide is also intended to present the state of the art with regard to power system design for commercial and
industrial facilities, in a consistent format along with traditionally-available material.
For the new college graduate from a four-year electrical engineering curriculum working in the field of commercial
and industrial power systems, this guide can serve as a starting point for learning the different aspects of the
profession. For the licensed design professional, this guide does present a number guidelines in a handy and
convenient reference.
This guide is not intended to substitute for the services of a licensed design professional, but can be of aid when
working with such professionals on commercial and industrial power system design.
Applications of electric power in industrial and commercial facilities

In both industrial and commercial environments, electric power is used for a wide number of applications. The
following is a brief list of the most common uses for electric power. This list is taken in part from[1], which provides
an expanded treatment of this subject.
I
Illumination Whether for providing light for an office environment or a manufacturing shop floor, illumination is
one of the most important applications of electric power, and the oldest.
Environmental systems Electric heating, ventilation, and air-conditioning are a large application for electric
power, and also an area in which electric power receives direct competition from other energy sources such as
natural gas.
Industrial processes Industrial processes account for a large percentage of the global use of electric power.
Typical process applications are listed as follows. These are not all-inclusive but do cover the majority of
process applications:
N Pumping
[1]
Chemical Processes
Semiconductor Preparation Processes
Furnaces
Standard Handbook for Electrical Engineers, New York: McGraw-Hill, 2001, pp. 21-1 - 21-99.
Smelting
Rolling Mills
Pulp-and-Paper Preparation Processes
Welding
Refrigeration
Drying
Well Drilling
Materials Handling
Water Treatment Processes
Computers and Data Centers With the advent of large computer networks the need also arisen for reliable
power for these.
Health Care Reliable power has always been a requirement of the health care industry, but added to this is
the need for power quality due to the nature of the equipment used.
Safety Systems Systems such as fire alarm and smoke detection systems, sprinkler systems and fire pumps
are vital to any commercial or industrial facility.
Communication Systems Systems such as telephone and intrusion detection and monitoring are
critically important.
Basic design philosophy

The following basic considerations are fundamental to any power system design:
I
Basic Safety: The power system must be able to perform all of its basic functions, and withstand basic
abnormal conditions, without damage to the system or to personnel.
B a s i c F u n c t i o n a l i t y : The power system must be able to distribute power from the source to the connected
loads in a reliable manner under normal conditions.
Reasonable Cost: The power system cost to obtain basic safety and functionality should be reasonable.
Code Compliance: All applicable codes must be complied with.
Above and beyond the basics are a multitude of considerations, some of which will apply to each particular
system design:
Enhanced Safety: The ability to withstand extremely abnormal conditions with a minimum of risk to personnel.
E n h a n c e d R e l i a b i l i t y : The ability to maintain service continuity during abnormal system conditions.
E n h a n c e d M a i n t a i n a b i l i t y : The system can be maintained with minimum interruption to service and with
minimum personnel protective equipment.
E n h a n c e d F l e x i b i l i t y : The ability to add future loads to the system, and with loads of a different nature than
currently exist on the system.
Enhanced Space Economy: The power system takes up the smallest possible physical space.
E n h a n c e d S i m p l i c i t y : The power system is easy to understand and operate.
Reduced Cost: The power system costs, both first cost and operating cost, are low.
Enhanced Power Quality: The power system currents and voltages are sinusoidal, without large amounts of
harmonics present. System voltage magnitudes do not change appreciably.
Enhanced Tr a n s p a r e n c y : The power system data at all levels is easily acquired and interpreted, and the power
system is easily interfaced with other building systems. Enhanced control of the system is also possible.
While it should be the goal of every power system design to meet the above basic considerations, no system
design can yield all of the enhanced characteristics listed. The relationship between the considerations listed is
shown in figure 1-1.
As can be seen, some of the enhanced characteristics mentioned are mutually exclusive, and to obtain a
combination of several enhanced characteristics requires a significant increase in cost. The design engineer,
therefore, must take into account the balance between the performance requirements of the system and the cost,
while not compromising the basic safety elements, functionality, and code compliance.
Figure 1-1: Power System Design Consideration Heuristics
Section 2:
Electric Power Fundamentals
Introduction
An understanding of the fundamentals of electric power is vital to successful power system design. It is assumed
that the reader has a degree in electrical engineering or electrical engineering technology, however the following
discussion is presented as review and reference material for completeness.
Basic Concepts
Commercial electric power in the United States is generated and delivered as alternating current, abbreviated as
AC. AC power consists of sinusoidal voltages and currents. Mathematically, an ac voltage or current can be
expressed as follows:
(2-1)
(2-2)
where
v(t)
i(t)
Vmax
Imax
f
v
i
t
is an AC voltage
is an AC current
is the voltage amplitude
is the current amplitude
is the system frequency
is the voltage phase shift in degrees
is the current phase shift in degrees
is the time in ms
All angles are measured in degrees

AC currents and voltages are economical to generate and, further, the magnitudes of the currents or voltages can
be stepped up or down using transformers.
Three-phase AC power is the standard in the United States due to its convenience of generation. Three-phase
(abbreviated 3) power is characterized by three different phases, each with a phase shift 120 degrees from the
other two phases. The three phases are typically referred to as A, B, and C. Further, the standard frequency
for the United States is 60Hz. Therefore, three-phase voltages in the United States can be mathematically
described as follows:
(2-3)
(2-4)
(2-5)
where
va(t)
vb(t)
vc(t)
is the A-phase voltage

is the B-phase voltage
is the C-phase voltage
The voltages from (2-3) - (2-5) are shown graphically in figure 2-1:
Vmax
12
24
t (ms)
Va
Vb
Vc
Figure 2-1: Graphical representation of 3 voltages
The peaks of the voltage waveforms are 120 (5.5 ms at 60 Hz) apart. Note that the peak of phase A occurs
before the peak of phase B, which in turn occurs before the peak of phase C. This is referred to as an ABC phase
sequence or ABC phase rotation. If any two phase labels are swapped, the result will be CBA phase rotation.
Both are encountered in practice. Also note that the definition of time = 0 is arbitrary due to the periodic nature of
the waveforms.
Because the full mathematical representation of AC voltages and currents is not practical, a shorthand notation is
usually used. This shorthand notation treats the sinusoids as complex quantities based upon the following
mathematical relationship:
(2-6)
The voltage quantities from (2-3) - (2-5) can therefore be rewritten as follows:
(2-7)
(2-8)
(2-9)
To further develop this shorthand notation, it must be recognized that the use of the RMS (root-mean-square)
quantity, rather than the amplitude, is advantageous in power calculations (discussed below). The RMS quantity
for a periodic function f(t) is defined as follows:
(2-10)
where
Frms
T
is the RMS value of the periodic function f(t)

is the period of f(t)
Using (2-10), the RMS value of each of the sinusoidal voltages from (2-3) - (2-5) are calculated as:
(2-11)
Because the RMS value is so useful in the calculation of power-related quantities, any time an AC voltage or
current value is given it is assumed to be an RMS value unless otherwise stated.
Assuming that only the real part of eje is kept, the voltages from (2-7) - (2-9) can be written as complex quantities
known as phasors:
V a = Vrms e j 21600t
(2-12)
(2-13)
(2-14)
Assuming a frequency of 60 Hz, the commonly-used shorthand notation for (2-12) - (2-14) is:
(2-15)
(2-16)
(2-17)
The phasor quantities in (2-15) - (2-17) can be treated as complex quantities for the purposes of manipulation and
calculation, but with the understanding that, if required, the basic time-domain voltage relationships (2-3) - (2-5)
can easily be obtained. The phasors can be plotted, as shown in figure 2-2:
Figure 2-2: Plot for phasors per (2-15) - (2-17)
In most instances the Re and Im axes are omitted since the definition of time zero (and thus angle zero) is
arbitrary; the important information conveyed is the angular relationships between the phasors themselves.
Note that the real part of a phasor is its projection on the Re axis; if the phasors are imagined to rotate in a
counter-clockwise direction about the 0,0 point it can be seen that the peak of va(t), represented by the tip of
phasor Va crossing the Re axis, occurs first, followed by the peak of vb(t), followed in turn by the peak of vc(t).
Thus for angles defined as positive in the counter-clockwise direction the ABC phase sequence is indicated by a
counter-clockwise phasor rotation. If angles are defined as positive in the clockwise direction a clockwise phasor
rotation would indicate an ABC phase sequence. Both are encountered in practice. In this guide all angles in
phasor diagrams will be assumed to be positive in the counter-clockwise direction.
A very general representation of a 3 system is shown in figure 2-3:
Figure 2-3: General 3 system representation
In figure 2-3 the three phases A, B, and C have been labeled, along with the neutral (N) and ground (G).
The neutral is optional, however the ground always exists. The AC voltages Va, Vb and Vc per the discussion
above could represent phase-to-phase voltages (Vab, Vbc,Vca), phase-to-neutral voltages (Van,Vbn, Vcn) or
phase-to-ground voltages (Vag, Vbg, Vcg). The existence of the neutral, and the relationship between the phases
and ground, is dependent upon the system grounding and is discussed in section 6 of this guide. Note
that a ground current is not defined; this is because the ground is not intended to carry load current, only ground
fault current as discussed in subsequent sections of this guide. In practice, when 3 voltages are discussed, they
are assumed to be phase-to-phase voltages unless otherwise noted.
AC power
With the basic concepts per above, AC electrical power can be described.
Consider the following DC circuit element:
Figure 2-4: DC Circuit element for power calculation
For the circuit element of figure 2-4 the following is true:

(2-18)
where
Vdc
Idc
P
is the DC voltage across the circuit element under consideration, with polarity as shown
is the DC current through the circuit element under consideration, considered positive for
the direction shown
is the power generated by, or dissipated through, the circuit element under consideration
The sign of P in (2-18) is dependent upon the direction of current flow with respect to the DC voltage. A positive
value for P indicates power dissipated, while a negative value for P indicates power generated. DC power is
measured in Watts, where one Watt is 1V x 1A.
With AC voltages and currents the expression for power is more complex. Assume that one phase is taken under
consideration, with and AC current and voltage as defined by (1-1) and (1-2) respectively. The expression for the
instantaneous power, after some manipulation, is:
(2-19)
Thus, the instantaneous power consists of two parts: A DC component and an AC component with a frequency
twice that of the system frequency. The quantity (v - i) is defined as the power angle or power factor angle and
is the angle by which the current peak lags behind the voltage peak on their respective waveforms. The quantity
P= cos(v-i) is known as the power factor of the circuit.
The average value of p(t) is of concern in AC circuits. The average value of p(t) is:
(2-20)
Recall that Vmax can be expressed in terms of Vrms per (2-11); substituting Vrms per (2-11) into (2-20) yields:
(2-21)
However, the absolute value of the product VrmsIrms cos(v-j) will always be less than VrmsIrms unless (v-j) = 0.
Further, if (v-j) = 90 , as is the case with a purely inductive or capacitive load, VrmsIrms cos(v-j) = 0.
Because energy is required to force current to flow, and energy is always conserved, AC power must have
another component. This component is most easily defined if AC power is treated as a complex quantity. To do
this, Complex Power S is defined as follows:
(2-22)
The quantities V and I are the AC current and voltage in their complex forms per (2-15) above, with the
*operator denoting the complex conjugate, or angle negation, of the current. This conjugation of the current is
done to yield the correct value for the power angle as described below. Real Power P and Reactive Power Q
are defined as follows:
(2-23)
(2-24)
(2-25)
(2-26)
P is expressed in Watts. Q has the same units but to differentiate it from P it is expressed in Voltamperes. rather
than Watts. S is the Apparent Power and is also expressed in Voltamperes.
The relationship between P, Q, S, S and (v-j) can be shown graphically:
Figure 2-5: Graphical depiction of AC power
The depiction in figure 2-5 is referred to as the power triangle since P, Q and S form a right triangle. It is also
important to note that the power factor angle is the same as the load impedance angle of the circuit. The power
factor is referred to as a lagging power factor if the current lags the voltage (i.e., (v-l) is positive up to 90)
and as a leading power factor if the current leads the voltage (i.e., (v-l) is negative down to -90). For a lagging
power factor, the real and reactive power flow in the same direction; for a lagging power factor they flow in
opposite directions. Of the passive circuit elements, resistors exhibit a unity power factor, inductors exhibit a zero
power factor lagging, and capacitors exhibit a zero power factor leading.
The foregoing discussion considers only single-phase circuits. For 3 circuits the power quantities for all three
phases must be added together, i.e.,
(2-27)
(2-28)
(2-29)
(2-30)
If the voltage magnitudes and power factor angles for each phase are equal, the power quantities per phase can
be represented as S1, S1, P1, and Q1; equations (2-27) - (2-30) can then be simplified as:
(2-31)
(2-32)
(2-33)
(2-34)
Transformers
Transformers are vital components for AC power systems. They are used to change the voltage and current
magnitudes to suit the application.
A.) The Ideal Transformer

Transformers are relatively simple devices that utilize Faradays law of electromagnetic induction. In its simplest
form, this law can be written:
= N
d
dt
(2-35)
where is the voltage induced in a coil of N turns that is linked by a magnetic flux .
In turn, the magnetic flux for a coil of N turns which through which a current I passes and linked by a magnetic
path with reluctance can be expressed as:
(2-36)
Consider the simple transformer shown in the following figure:
Figure 2-6: Basic transformer model
From (2-35) and (2-36),

(2-37)
(2-38)
(2-39)
(2-40)
(2-41)
Dividing (2-40) by (2-41),
(2-42)
Equations (2-38) and (2-42) are the basic equations for a single-phase transformer. The voltage ratio (V1/V2) is
equal to the turns ratio (N1/N2), and the current ratio is equal to the inverse of the turns ratio. By re-writing (2-38)
in terms of the turns ratio (N1/N2) an substituting into (2-42), the following is obtained:
(2-43)
This is to be expected, since the apparent flowing into the transformer should ideally equal the apparent power
flowing out of the transformer.
The usefulness of the transformer lies in the fact that it can adjust the voltage and current to the application. For
example, on a transmission line it is advantageous to keep the voltage high in order to be able to transmit the
power with as small a current as possible, in order to minimize line losses and voltage drop. At utilization
equipment, it is advantageous to work with low voltages that are more conducive to equipment design and
personnel safety.
Another important aspect of the transformer is that it changes the impedance of the circuit. For example, if an
impedance Z2 is connected to winding 2 of the ideal transformer in figure 2-6 it can be stated by definition that
(2-44)
Using (2-38) and (2-42), (2-44) can be written in terms of V1 and I1:
(2-45)
By definition,
(2-46)
Therefore, (2-45) can be re-written as
(2-47)
As can be seen, the impedance as seen through the transformer is the load impedance at the transformer output
winding multiplied by the square of the turns ratio.
B.) A Practical Transformer Model

The idealized transformer model just presented is not sufficient for practical electric power applications due to the
fact that the core is not lossless and not all of the magnetic flux links both sets of windings. To take this into
account, a more realistic model is used:
Figure 2-7: Practical transformer model
The resistance Rc represents the core losses due to hysteresis, and inductance Lc represents the magnetizing
inductance. Resistances R1 and R2 represent the winding resistances of winding 1 and winding 2, respectively.
Inductances L1 and L2 represent the leakage inductances of windings 1 and 2, respectively. For quick
calculations, the core losses and magnetizing inductance are often ignored, and the model is treated as an
impedance in series with an ideal transformer.
To insure the proper polarity, the circuit representation for a transformer includes polarity marks as shown in figure
2-8. If the current for one winding flows into its terminal with the polarity mark, the current for the other winding
flow out of its terminal with the polarity mark. In addition, the ANSI polarity markings per [1] are shown; H
denotes the higher voltage winding, and X denotes the lower voltage winding.
Figure 2-8: Standard transformer symbolic representation
C.) 3 Transformer Connections

To be useful in 3 systems transformers must be connected for use with 3 voltage. This is accomplished by the
use of 3 transformer connections.
The wye-wye connection is shown in figure 3-9. This could be a bank of three single-phase transformers or one
3 transformer which consists of all three sets of windings on a common ferromagnetic core. Polarity markings
for three single-phase transformer connections are shown at the individual transformers, and polarity markings for
a 3 transformer are shown next to the A, B, C, and N terminals.
For both the primary and secondary windings the magnitude of the line-to-line voltage is equal to the magnitude of
the line-to-neutral voltage multiplied by 3.
For convenience the transformer turns ratio is taken as 1:1 on the phasor diagram.
Figure 2-9: Wye-Wye transformer connection
If a three-phase transformer is used, the wye-wye connection has the disadvantage of requiring a four-legged
core to allow for a magnetic flux imbalance. Further, the solidly-grounded neutrals allow for ground currents to flow
that can create interference in communications circuits [2]. Both the primary and secondary neutrals terminals
must be solidly-grounded to allow for triplen-harmonic currents to flow; if the neutrals are allowed to float harmonic
overvoltages will be developed from phase to neutral on each winding. These overvoltages can damage the
transformer insulation. Wye-wye transformers are often used on systems above 25 kV to minimize a problem
known as ferroresonance. Ferroresonance is a condition which results from the transformer magnetizing
impedance resonating with the upstream cable charging capacitance, resulting in destructive overvoltages as the
transformer core moves into and out of saturation in a non-linear manner. Single-phase switching is usually the
cause of ferroresonance.
The delta-delta connection is shown in figure 2-10. Note that there is no neutral on the delta-delta connection.
A unique feature of this connection is that if one transformer is taken out of service, the two remaining
transformers can still provide three-phase service at a reduced capacity (57.7% of the capacity with all three
transformers in service).
Figure 2-10: Delta-Delta transformer connection
The delta-wye connection is shown in figure 2-11. Note that for the given turns ratios of 1:1 that the magnitude
of the phase-to-phase output voltage is equal to the magnitude of the phase-to-phase input voltage multiplied
by 3 . The input and output voltages of 3 transformers and 3 banks of single-phase transformers are always
referenced as the phase-to-phase magnitude. Therefore, for a delta-wye transformer the winding turns ratios for
each set of windings must be compensated by (1/3 ) to produce the desired input-to-output voltage ratio.
Note also that the phase-to-phase voltages on the lower voltage side of the transformer lag the phase-to-phase
voltages on the high voltage side by 30. This is dictated by [1].
The delta-wye transformer connection is by far the most popular choice for commercial and industrial applications.
3 transformers do not require a four-legged core like the wye-wye connection, but the advantages of a wye
secondary winding (elaborated on in section 6 of this guide) are obtained. Further, the secondary neutral can be
left unconnected in this arrangement, unlike the wye-wye arrangement.
Figure 2-11: Delta-Wye transformer connection
10
The wye-delta connection is shown in figure 2-12. This connection is seldom used in commercial and industrial
applications. Note that the delta is arranged differently from the delta-wye connection, in order to satisfy the
requirement from [1] to have the phase-to-phase voltages on the low-voltage side of the transformer lag the
corresponding voltages on the primary side by 30.
Figure 2-12: Wye-Delta transformer connection
Basic electrical formulae

The following formulae are given as a convenient reference for the reader. These formulae include both formulae
derived in this section and those basic formulae which are derived from basic circuit theory.
A.) DC Circuits
(2-48)
(2-49)
(2-50)
where
Vdc
Idc
Rdc
P
is the DC voltage across the circuit element under consideration

is the current through the circuit element under consideration. Idc is considered positive if it flows from
the circuit element terminal at the higher voltage to the terminal at the lower voltage
is the DC resistance of the circuit element under consideration, measured in ohms
is the power dissipated or generated by the circuit element. A positive power from (2-49) indicates
power dissipated by the circuit element, and a negative value indicates power generated by the circuit
element. The sign of P in (2-50) is lost due to the squaring of the current or voltage
11
B.) Passive Energy Storage Elements

Capacitors store energy in the form of voltage, with a governing equations:
(2-51)
(2-52)
where
vc
ic
C
E
is the voltage across the capacitor

is the current through the capacitor, considered positive if it flows toward the terminal
from which vc is referenced
is the capacitance value of the capacitor, measured in Farads
is the energy stored in the capacitor
Inductors store energy in the form of current, with governing equations:

(2-53)
(2-54)
where
vi
ii
L
E
is the voltage across the inductor

is the current through the inductor, considered positive if it flows toward the terminal
from which vI is referenced
is the inductance value of the inductor, measured in Henries
is the energy stored in the inductor
C.) AC Voltages and Currents, Time-Domain Form

Single-phase AC voltage and current can be expressed as follows
(2-55)
(2-56)
where
v(t)
i(t)
Vmax
Imax
f
v
i
t
is an AC voltage
is an AC current
is the voltage amplitude
is the current amplitude
is the voltage phase shift in degrees
is the current phase shift in degrees
is the time in ms
All angles are measured in degrees
12
If the frequency is considered to be 60 Hz, 3 voltages can be written as:

(2-57)
(2-58)
(2-59)
where
va(t)
vb(t)
vc(t)
is the A-phase voltage

is the B-phase voltage
is the C-phase voltage
The RMS value of a perfectly sinusoidal ac voltage or current is

(2-60)
(2-61)
C.) AC Power, Time-Domain Form

The instantaneous single-phase AC power resulting from a current per (2-56) flowing through a circuit element
with voltage (2-55) across it is
(2-62)
where all terms are as defined for (2-54) and (2-55).
The average power in this case is
(2-63)
(2-64)
where P is the average power.
E.) AC Currents, Voltages and Circuit Elements, Frequency-Domain Form

Assuming only the real part is kept when converting to time-domain and that the same frequency applies
throughout, AC currents and voltages can be written in the frequency domain as
(2-65)
(2-66)
where
V
I
v
I
is the AC voltage in frequency-domain form

is the AC current in frequency-domain form
is the voltage phase shift
is the current phase shift
13
Capacitor and Inductor impedances in frequency domain form are

(2-67)
(2-68)
where
Zc
Zi
is the impedance of the capacitor

is the impedance of the inductor
= -1
f
C
L

is the capacitance of the capacitor
is the inductance of the inductor
AC Voltage and current for an impedance Z are related as follows:

(2-69)
Average single-phase power in AC form can be expressed as:
(2-70)
where S is the complex power.
Complex power S can be separated into real power P and reactive power Q:
(2-71)
(2-72)
(2-73)
(2-74)
For 3 circuits the total power is the sum of the power in each phase, i.e.,
(2-75)
(2-76)
(2-77)
(2-78)
If the current and voltage magnitudes and angles are equal for each phase (2-76) - (2-78) can be simplified as
follows by considering the power quantities per phase to be S1, S1, P1, and Q1:
(2-79)
(2-80)
(2-81)
(2-82)
14
F.) Basic Transformers

The input voltage and current V1 and I1 and the output voltage and current V2 and I2 for an ideal single-phase
transformer are related as follows:
(2-83)
(2-84)
The impedance Z1 at the input terminals of a transformer is related to the load impedance Z2 connected to the
transformer output terminals by the following equation:
(2-85)
References
Because the subject matter for this section is basic and general to the subject of electrical engineering, it is
included in most undergraduate textbooks on basic circuit analysis and electric machines. Where material is
considered so basic as to be axiomatic no attempt has been made to cite a particular source for it.
For material not covered per the above, references specifically cited in this section are:
[1]
IEEE Standard Terminal Markings and Connections for Distribution and Power Transformers,
IEEE Std. C57.12.70-2000.
[2]
Turan Gonen, Electric Power Distribution System Design, New York: McGraw-Hill, 1986, p.137.
15
Section 3:
Load Planning
Basic Principles
The most vital, but often the last to be acquired, pieces of information for power system design are the load
details. An important concept in load planning is that due to non-coincident timing, some equipment operating
at less than rated load, and some equipment operating intermittently rather than continuously, the total demand
upon the power source is always less than the total connected load [1]. This concept is known as load diversity.
The following standard definitions are given in [1] and [2] and are tools to quantify it:
Demand: The electric load at the receiving terminals averaged over a specified demand interval. of time, usually
15 min., 30 min., or 1 hour based upon the particular utilitys demand interval. Demand may be expressed in
amperes, kiloamperes, kilowatts, kilovars, or kilovoltamperes.
Demand Interval: The period over which the load is averaged, usually 15 min., 30 min., or 1 hour.
Peak Load: The maximum load consumed or produced by a group of units in a stated period of time. It may be
the maximum instantaneous load or the maximum average load over a designated period of time.
Maximum Demand: The greatest of all demands that have occurred during a specified period of time such as
one-quarter, one-half, or one hour. For utility billing purposes the period of time is generally one month.
Coincident Demand: Any demand that occurs simultaneously with any other demand.
Demand Factor: The ratio of the maximum coincident demand of a system, or part of a system, to the total
connected load of the system, or part of the system, under consideration, i.e.,
(3-1)
Diversity Factor: The ratio of the sum of the individual maximum demands of the various subdivisions of a
system to the maximum demand of the whole system, i.e.,
(3-2)
where
Di
DG
= maximum demand of load i, regardless of time of occurrence.

= coincident maximum demand of the group of n loads.
Using (1), the relationship between the diversity factor and the demand factor is
(3-3)
where
TCLi
DFi
= total connected load of load group i

= the demand factor of load group i
Load Factor: The ratio of the average load over a designated period of time to the peak load occurring
in that period, i.e.,
(3-4)
If T is the designated period of time, an alternate formula for the load factor may be obtained by manipulating
(3-4) as follows:
(3-5)
These quantities must be used with each type of load to develop a realistic picture of the actual load requirements
if the economical sizing of equipment is to be achieved. Further, they are important to the utility rate structure
(and thus the utility bill).
As stated in [2], the following must be taken into account in this process:
I
Load Development/Build-Up Schedule Peak load requirements, temporary/construction power

requirements, and timing
Load Profile Load magnitude and power factor variations expected during low-load, average load,
and peak load conditions
Expected Daily and Annual Load Factor
Large motor starting requirements
Special or unusual loads such as resistance welding, arc welding, induction melting, induction heating, etc.
Harmonic-generating loads such as variable-frequency drives, arc discharge lighting, etc.
Forecasted load growth over time
Reference [4] and individual engineering experience on previous projects are both useful in determining demand
factors for different types of loads. In addition, the National Electrical Code [3] gives minimum requirements for
the computation of branch circuit, feeder, and service loads.
NEC Basic branch circuit requirements

NEC [3] Article 220 gives the basic requirements for load calculations for branch circuits, feeders, and services.
In order to understand these requirements, the basic NEC definitions of branch circuit, feeder, and service must
be understood, along with several other key terms:
Branch Circuit: The circuit conductors between the final overcurrent device protecting the circuit
and the outlet(s).
Feeder: All circuit conductors between the service equipment, the source of a separately derived system, or other
power supply source and the final branch-circuit overcurrent device.
Service: The conductors and equipment for delivering electric energy from the serving utility to the wiring system
of the premises served.
Outlet: The point on the wiring system at which current is taken to supply utilization equipment.
Receptacle: A receptacle is a contact device installed at the outlet for the connection of an attachment plug.
A single receptacle is a single contact device with no other contact device on the same yoke. A multiple receptacle
is two or more contact devices on the same yoke.
Continuous Load: A load where the maximum current is expected to continue for three hours or more.
The NEC definition of Demand Factor is essentially the same as given above.
I
Minimum lighting load (Article 220.12): Minimum lighting load must not be less than as specified in table 3-1
(NEC Table 220.12):
Table 3-1: General lighting loads by occupancy (NEC [3] table 220.12)
Type of Occupancy
Unit Load
Volt-Amperes
Per quare Meter
Unit Load
Volt-Amperes
per Square Foot
Armories and auditoriums
11
Banks
39
3.5b
Barber shops and beauty parlors
33
Churches
11
Clubs
22
22
Dwelling Units
33
Garages commercial (storage)
0.5
Hospitals
22
Hotels and motels, including apartment houses without provision

for cooking by tenantsa
22
Industrial commercial (loft) buildings
22
Lodge rooms
17
1.5
Office buildings
39
3.5b
Restaurants
22
Schools
33
Stores
33
Warehouses (storage)
0.25
Assembly halls and auditoriums
11
Halls, corridors, closets, stairways
0.5
Storage Spaces
0.25
Court Rooms
a
In any of the preceding occupancies except one- family dwellings and

individual dwelling units of two-family and multi-family dwellings:
a
b
See NEC Article 220.14(J)

See NEC Article 220.14(K)
Motor Loads (Article 220.14(C)): Motor loads must be calculated in accordance with Articles 430.22,
430.24, and 440.6, summarized as follows:):
N The full load current rating for a single motor used in a continuous duty application is 125% of the motors
full-load current rating as determined by Article 430.6, which refers to horsepower/ampacity tables 430.247,
430.248, 430.249, or 430.250 as appropriate (Article 430.22).
N
The load calculation for several motors, or a motor(s) and other loads, is 125% of the full load current rating
of the highest rated motor per a.) above plus the sum of the full-load current ratings of all the other motors in
the group, plus the ampacity required for the other loads (Article 430.24).
Luminaires (lighting fixtures) (Article 220.14(D)): An outlet supplying luminaire(s) shall be calculated based upon
the maximum volt-ampere rating of the equipment and lamps for which the luminaire(s) is rated.
Heavy-Duty Lampholders (Article 220.14(E)): Loads f for heavy-duty lampholders must be calculated at a
minimum of 600 volt-amperes.
Sign and outline lighting (Article 220.14(F)): Sign and outline lighting loads shall be calculated at a minimum of
1200 volt-amperes for each required branch circuit specified in article 600.5(A).
Show windows (Article 220.14(G)): Show windows can be calculated in accordance with either:
N The unit load per outlet as required in other provisions of article 220.14.
N
200 volt-amperes per 300mm (1ft.) of show window.
Loads for fixed multioutlet assemblies in other than dwelling units or the guest rooms and guest suites of hotels
or motels must be calculated as follows (Article 220.14(H)):
N Where appliances are unlikely to be used simultaneously, each 1.5m (5 ft.) or fraction thereof of each
separate and continuous length must be considered as one outlet of 180 volt-amperes.
N
For hermetic refrigerant motor compressors or multi-motor equipment employed as part of air conditioning or
refrigerating equipment, the equipment nameplate rated load current should be used instead of the motor
horsepower rating (Article 440.6).
Where appliances are likely to be used simultaneously, each 300mm (1 ft.) or fraction thereof must be
considered as an outlet of 180 volt-amperes.
Receptacle outlets (Articles 220.14(I), 220.14(J), 220.14(K), 220.14(L)): Loads for these are calculated
as follows:
N Dwelling occupancies (Article 220.14(J)): In one-family, two-family, and multifamily dwellings and in guest
rooms or guest suites of hotels and motels, general-use receptacle outlets of 20A rating or less are included
in the general lighting load per above. No additional load calculations are required for these.
N
Banks and office buildings (Article 220.14(K)): Receptacle outlets must be calculated to be the larger of
either the calculated value per c.) below or 11 volt-amperes/square meter (1 volt-ampere per square ft.).
All other receptacle outlets (Article 220.14(I)): Each receptacle on one yoke must be calculated as
180 volt-amperes. A multiple receptacle consisting of four or more receptacles must be calculated at
90 volt-amperes per receptacle.
Sufficient branch circuits must be incorporated into the system design to serve the loads per Article 220.10
(summarized 1.) 8.) above), along with branch circuits for any specific loads not covered in Article 220.10.
The total number of branch circuits must be determined from the calculated load and the size or rating of the
branch circuits used. The load must be evenly proportioned among the branch circuits (Article 210.11(C)).
In addition, Article 210.11(C) requires several dedicated branch circuits as follows for dwelling units:
N Two or more 20A small-appliance branch circuits (Article 210.11(C)(1)).
N
One or more 20A laundry branch circuits (Article 210.11(C)(2)).
One or more bathroom branch circuits (Article 210.11(C)(3)).
Continuous Loads (Article 210.20): The rating of the overcurrent protection for a branch circuit must be at least
the sum of the non-continuous load +125% of the continuous load unless the overcurrent device is 100%-rated.
Because the rating of the overcurrent protection determines the rating of the branch circuit (Article 210.3), the
branch circuit must be sized for the non-continuous load +125% of the continuous load. In load calculations,
continuous loads should therefore be multiplied by 1.25 unless the circuit overcurrent device is 100% rated.
Note that motor loads are not included in this calculation as the 125% factor is already included in the applicable
sizing per above.
NEC Basic Feeder Circuit Sizing Requirements

Once the branch circuit loads are calculated, the feeder circuit loads may be calculated by applying demand
factors to the branch circuit loads.
I
General Lighting Loads (Article 220.42): The feeder general lighting load can be calculated by multiplying the
branch circuit general lighting load calculated per B.) 1.) above, for those branch circuits supplied by the feeder,
by a demand factor per table 3-2 (NEC table 220.42).
Table 3-2: Lighting load feeder demand factors (NEC [3] table 220.42)
Type of Occupancy
Portion of Lighting Load to

Which Demand Factor Applies
(Volt-Amperes)
Demand Factor
(Percent)
Dwelling units
First 3,000 or less at

From 3,001 to 120,000 at
Remainder over 120,000 at
100
35
25
Hospitals*

40
20
Hotels and motels, including apartment houses without

provision for cooking by tenants*

From 20,001 to 100,000 at
50
40
30
Warehouses (storage)

100
50
All others
Total volt-amperes
100
* The demand factors of this table shall not apply to the calculated load of feeders or services supplying areas in hospitals, hotels,
and motels where the entire lighting is likely to be used at one time, as in operating rooms, ballrooms, or dining rooms.
Show window or track lighting (Article 220.43): Show windows must use a calculated value of 660 voltamperes per linear meter (200 volt-amperes per linear foot), measured horizontally along its base. Track
lighting in other than dwelling units must be calculated at an 150 volt-amperes per 660mm (2 ft.) of lighting
track or fraction thereof.
Receptacles in other than dwelling units (Article 220.44): Demand factors for non-dwelling receptacle loads
are given in table 3-3 (NEC table 220.44).
Table 3-3: Demand factors for non-dwelling receptacle loads (NEC [3] table 220.44)
Portion of Receptacle Load to Which Demand Factor Applies (Volt-Amperes)
Demand Factor (Percent)
First 10 kVA or less at
100
Remainder over 10 kVA
50
Motors (Article 220.50): The feeder demands for these are calculated as follows:
N The load calculation for several motors, or a motor(s) and other loads, is 125% of the full load current rating
of the highest rated motor per II.) B.) ii.) above plus the sum of the full-load current ratings of all the other
motors in the group, plus the ampacity required for the other loads (Article 430.24).
N
The load calculation for factory-wired multimotor and combination-load equipment should be based upon the
minimum circuit ampacity marked on the equipment (Article 430.25) instead of the motor horsepower rating.
Where allowed by the Authority Having Jurisdiction, feeder demand factors may be applied based upon the
duty cycles of the motors. No demand factors are given in the NEC for this situation.
Fixed Electric Space Heating (Article 220.51): The feeder loads for these must be calculated at 100% of
the connected load.
Noncoincident Loads (Article 220.60): Where it is unlikely that two or more noncoincident loads will be in use
simultaneously, it is permissible to use only the largest loads that will be used at one time to be used in
calculating the feeder demand.
Feeder neutral load (Article 220.61): The feeder neutral load is defined as the maximum load imbalance on the
feeder. The maximum load imbalance for three-phase four-wire systems is the maximum net calculated load
between the neutral and any one ungrounded conductor. A demand factor of 70% may be applied to this
calculated load imbalance. Refer to NEC article 220.61 for neutral reductions in systems other than
three-phase, four-wire systems. This demand factor does not apply to non-linear loads; in fact, it may be
necessary to oversize the neutral due to current flow from non-linear load triplen harmonics.
Continuous Loads (Article 215.3): The rating of the overcurrent protection for a feeder circuit must be at least
the sum of the non-continuous load +125% of the continuous load, unless the overcurrent device is 100%-rated.
Because the rating of the overcurrent protection determines the rating of the branch circuit (Article 210.3),
the branch circuit must be sized for the non-continuous load +125% of the continuous load. In the final feeder
circuit load calculation, the continuous portion of the load should therefore be multiplied by 1.25 unless the
overcurrent device for the circuit is 100%-rated. Note that motor loads are not included in this calculation as the
125% factor is already included in the applicable sizing per above.
Additional calculation data is given in NEC Article 220 for dwelling units, restaurants, schools, and farms. This
data is not repeated here. Refer to NEC Article 220 for details.
As this guide only presents the basic NEC requirements for load calculations, it is imperative to refer to the NEC
itself when in doubt about a specific load sizing application. Computer programs are commercially available to
automate the calculation of feeder and branch circuit loads per the NEC methodology described above.
References
Because the subject matter for this section is basic and general to the subject of electrical engineering, it is
included in most undergraduate textbooks on basic circuit analysis and electric machines. Where material is
considered so basic as to be axiomatic no attempt has been made to cite a particular source for it.
For material not covered per the above, references specifically cited in this section are:
[1]
IEEE Recommended Practice for Electric Power Distribution for Industrial Plants,
IEEE Standard 141-1993, December 1993.
[2]
Turan Gonen, Electric Power Distribution System Design, New York: McGraw-Hill, 1986, pp. 37-51.
[3]
The National Electrical Code, NFPA 70, The National Fire Protection Association, Inc., 2005 Edition.
[4]
IEEE Recommended Practice for Electric Power Systems in Commercial Buildings,

IEEE Standard 241-1990, December 1990.
Section 4:
System Voltage Considerations
Basic Principles
The selection of system voltages is crucial to successful power system design. Reference [1] lists the standard
voltages for the United States and their ranges. The nominal voltages from [1] are given in table 4-1.
As can be seen, ANSI C84.1-1989 divides system voltages into voltage classes. Voltages 600 V and below are
referred to as low voltage, voltages from 600 V-69 kV are referred to as medium voltage, voltages from
69 kV-230 kV are referred to as high voltage and voltages 230 kV-1,100 kV are referred to as extra high
voltage, with 1,100 kV also referred to as ultra high voltage. The emphasis of this guide is on low and
medium voltage distribution systems.
Table 4-1: Standard nominal three-phase system voltages per ANSI C84.1-1989
Voltage Class
Three-wire
Low Voltage
240
480
600
Medium Voltage
2,400
4,160
4,800
6,900
13,800
Four-wire
208 Y/120
240/120
480 Y/277
4,160 Y/2400
8,320 Y/4800
12,000 Y/6,930
12,470 Y/7,200
13,200 Y/7,620
13,800 Y/7,970
20,780 Y/12,000
22,860 Y/13,200
23,000
34,500
46,000
69,000
High Voltage
115,000
138,000
161,000
230,000
Extra-High Voltage
345,000
500,000
765,000
Ultra-High Voltage
1,100,000
24,940 Y/14,400
34,500 Y/19,920
The choice of service voltage is limited to those voltages which the serving utility provides. In most cases only one
choice of electrical utility is available, and thus only one choice of service voltage. As the power requirements
increase, so too does the likelihood that the utility will require a higher service voltage for a given installation.
In some cases a choice may be given by the utility as to the service voltage desired, in which case an analysis of
the various options would be required to arrive at the correct choice. In general, the higher the service voltage the
more expensive the equipment required to accommodate it will be. Maintenance and installation costs also
increase with increasing service voltage. However, equipment such as large motors may require a service voltage
of 4160 V or higher, and, further, service reliability tends to increase at higher service voltages.
Another factor to consider regarding service voltage is the voltage regulation of the utility system. Voltages defined
by the utility as distribution should, in most cases, have adequate voltage regulation for the loads served.
Voltages defined as subtransmission or transmission, however, often require the use of voltage regulators or
load-tap changing transformers at the service equipment to give adequate voltage regulation. This situation
typically only occurs for service voltages above 34.5 kV, however it can occur on voltages between 20 kV and
34.5 kV. When in doubt the serving utility should be consulted.
The utilization voltage is determined by the requirements of the served loads. For most industrial and commercial
facilities this will be 480 Y/277 V, although 208 Y/120 V is also required for convenience receptacles and small
machinery. Large motors may require 4160 V or higher. Distribution within a facility may be 480 Y/277 V or, for
large distribution systems, medium voltage distribution may be required. Medium voltage distribution implies a
medium voltage (or higher) service voltage, and will result in higher costs of equipment, installation, and
maintenance than low voltage distribution. However, this must be considered along with the fact that medium
voltage distribution will generally result in smaller conductor sizes and will make control of voltage drop easier.
Power equipment ampacity limitations impose practical limits upon the available service voltage to serve a given
load requirement for a single service, as shown in table 4-2.
Voltage drop considerations

Because all conductors exhibit an impedance to the flow of electric current, the voltage will not be constant
throughout the system, but will tend to drop as one moves closer to the load. Ohms Law, expressed in phasor
form for AC circuits, gives the basic relationship for voltage drop vs. the load current:
(4-1)
where
Vdrop
Il
Zcond
is the voltage drop along a length of conductor or across a piece of equipment in volts
is the load current in amperes
is the conductor or equipment impedance, in ohms
Thus, the larger the load current and larger the conductor impedance, the larger the voltage drop.
Unbalanced loads will, of course, give an unbalanced voltage drop, which will lead to an unbalanced voltage
at the utilization equipment.
A voltage drop of 5% or less from the utility service to the most remotely-located load is recommended by NEC
article 210.19(A)(1), FPN No. 4. Because this is a note only, it is not a requirement per se but is the commonly
accepted guideline.
Table 4-2: Power equipment design limits to service voltage vs. load requirements,
for a single service
Voltage (V)
Equipment Type
Maximum Equipment
Ampacity (A)
Maximum Load (kVA)
5000
1,800
4,157
5,196
208
480
600
Switchboard or Low Voltage

Power Switchgear
2,400
4,160
4,800
Metal-Enclosed Switchgear,
w/Fuses 69,000
6,900
8,320
12,000
12,470
13,200
13,800
w/Fuses
720
8,605
10,376
14,965
15,551
16,461
17,210
20,780
22,860
23,000
24,940
w/Fuses
175
6,299
6,929
6,972
7,560
34,500
w/Fuses
115
6,872
2,400
4,160
4,800
6,900
8,320
12,000
12,470
13,200
13,800
Metal-Clad Switchgear
3000
12,471
21,616
24,942
38,853
43,232
62,354
64,796
68,589
71,707
20,780
22,860
23,000
24,940
Metal-Clad Switchgear
2000
71,984
79,189
79,674
86,395
1080
4,489
7,782
8,979
Because conductor impedance increases with the length of the conductor, it can be seen that unless the power
source is close to the center of the load the voltage will vary across the system, and, further, it can be more costly
to maintain the maximum voltage drop across the system to within 5% of the service voltage since larger
conductors must be used to offset longer conductor lengths.
Also from equation (4-1) it can be seen that as load changes, so does the voltage drop. For a given maximum
load, a measure of this change at a given point is the voltage regulation, defined as
(4-2)
where
Vno load
Vload
is the voltage, at a given point in the system, with no load current flowing from that point to the load.
is the voltage, at the same point in the system, with full load current flowing from that point to the load.
Another source of concern when planning for voltage drop is the use of power-factor correction capacitors.
Because these serve to reduce the reactive component of the load current they will also reduce the voltage drop
per equation (4-1).
Both low and high voltage conditions, and voltage imbalance, have an adverse effect on utilization equipment (see
[2] for additional information). Voltage drop must therefore be taken into account during power system design to
avoid future problems.
References
[1]
American National Standard Preferred Voltage Ratings for Electric Power Systems and Equipment (60 Hz),
ANSI C84.1-1989.
[2]
IEEE Recommended Practice for Electric Power Distribution for Industrial Plants, IEEE Standard 141-1993,
December 1993.
Section 5:
System Arrangements
Introduction
The selection of system arrangement has a profound impact upon the reliability and maintainability of the system.
Several commonly-used system topologies are presented here, along with the pros and cons of each. The figures
for each of these assume that the distribution and utilization voltage are the same, and that the service voltage
differs from the distribution/utilization voltage. The symbology (low voltage circuit breaker, low voltage drawout
circuit breaker, medium voltage switch, medium voltage breaker) reflects the most commonly-used equipment for
each arrangement. The symbology used throughout this section is shown in figure 5-1:
Figure 5-1: Symbology
Radial system
The radial system is the simplest system topology, and is shown in figure 5-2. It is the least expensive in terms of
equipment first-cost. However, it is also the least reliable since it incorporates only one utility source and the loss
of the utility source, transformer, or the service or distribution equipment will result in a loss of service. Further, the
loads must be shut down in order to perform maintenance on the system. This arrangement is most commonly
used where the need for low first-cost, simplicity, and space economy outweigh the need for enhanced reliability.
Typical equipment for this system arrangement is a single unit substation consisting of a fused primary switch,
a transformer of sufficient size to supply the loads, and a low voltage switchboard.
Radial system with primary selectivity

This arrangement is shown in figure 5-3. If two utility sources are available, it provides almost the same economic
advantages of the radial system in figure 2 but also gives greater reliability since the failure of one utility source
will not result in a loss of service (note that an outage will occur between the loss of the primary utility source and
switching to the alternate source unless the utility allows paralleling of the two sources). The loss of the
transformer or of the service or distribution equipment would still result in a loss of service. Maintenance on the
system requires all loads to be shut down.
Figure 5-2: Radial System
Figure 5-3: Radial System with Primary Selectivity
An automatic transfer scheme may optionally be provided between the two primary switches to
automatically switch from a failed utility source to an available source. Most often metal-clad circuit
breakers are used, rather than metal-enclosed switches, if this is the case. More about typical
equipment application guidelines follows in a subsequent section of this guide.
Expanded radial system

The radial systems shown in figures 5-2 and 5-3 can be expanded by the inclusion of additional transformers.
Further, these transformers can be located close to the center of each group of loads to minimize voltage drop.
Reliability increases with a larger number of substations since the loss of one transformer will not result in a loss
of service for all of the loads.
Figure 5-4 shows an expanded radial system utilizing multiple substations, but still with only one utility source and
only one primary feeder:
Figure 5-4: Expanded Radial System with one utility source and a single primary feeder
A more reliable and maintainable arrangement utilizing multiple primary feeders is shown in figure 5-5. In the
system of figure 5-5, each unit substation is supplied by a dedicated feeder from the service entrance switchgear.
Each substation is also equipped with a primary disconnect switch to allow isolation of each feeder on both ends
for maintenance purposes.
Typical service entrance equipment consists of a metal-clad switchgear main circuit breaker and metal-enclosed
fused feeder switches. Metal-Clad circuit breakers may be used instead of metal-enclosed feeder switches
if required.
Figure 5-5: Expanded Radial System with one utility source and multiple primary feeders
Figure 5-6 shows an expanded radial system utilizing multiple substations and two utility sources, again with
metal-clad primary switchgear but with a duplex metal-enclosed switchgear for utility source selection:
Figure 5-6: Expanded Radial System with two utility sources and multiple primary feeders
Of the arrangements discussed this far, the arrangement of figure 5-6 is the most reliable it does not depend
upon a single utility source for system availability, nor does the failure of one transformer or feeder cause a loss of
service to the entire facility. However, the loss of a transformer or feeder will result in the loss of service to a part
of the facility. More reliable system arrangements are required if this is to be avoided.
Loop system
The loop system arrangement is one of several arrangements that can allow one system component, such as a
transformer or feeder cable, to fail without causing a loss of service to a part of the facility.
Figure 5-7 shows a primary loop arrangement. The advantages of this arrangement over previously-mentioned
arrangements are that a failure of one feeder cable will not cause one part of the facility to experience a loss of
service and that one feeder cable can be maintained without causing a loss of service (note that an outage to part
of the system will be experienced after the failure of a feeder cable until the loop is switched to accommodate the
loss of the cable).
In figure 5-7 metal-clad circuit breakers are used as the feeder protective devices. Fused metal-enclosed-feeder
switches could be utilized for this, but caution must be used if this is considered since the feeder fuses would
have to be able to serve both transformers and the feeder and transformer fuses would have to coordinate for
maximum selectivity.
It must be noted that the system arrangement of figure 5-7 is designed to be operated with the loop open, i.e., one
of the four loop switches shown would be normally-open. If closed-loop operation were required, metal-clad circuit
breakers should be used instead to provide maximum selectivity (this arrangement is discussed further below).
Momentary paralleling to allow maintenance of one section of the loop without causing an outage to one part of
the facility can be accomplished with metal-enclosed loop switches, however, if caution is used in the system
design and maintenance.
Figure 5-7: Primary Loop System
Secondary-Selective system
Another method of allowing the system to remain in service after the failure of one component is the secondaryselective system. Figure 5-8 shows such an arrangement.
The system arrangement of figure 5-8 has the advantage of allowing one transformer to fail without causing a loss
of service to one part of the plant. This is a characteristic none of the previously-mentioned system arrangements
exhibit. The system can be run with the secondary bus tie breaker normally-open or normally-closed. If the bus tie
breaker is normally-closed the failure of one transformer, if directional overcurrent relays are supplied on the
transformer secondary main circuit breakers, will not cause an outage, however care must be taken in the system
design as the available fault current at the secondary switchgear can be doubled in this case.
Typical equipment for this arrangement is low voltage power circuit-breaker switchgear with drawout circuit
breakers, both for reasons of coordination and maintenance. However, a low voltage switchboard may be utilized
also if care is taken in the system design and the system coordination is achievable. For a normally-closed bus tie
breaker, low voltage power switchgear is essential since the breakers lend themselves more readily external
protective relaying.
Note that if one transformer fails the other transformer and its associated secondary main circuit must carry the
entire load. This must be taken into account in sizing the transformer and secondary switchgear for this type of
system to be effective.
Figure 5-8: Secondary-Selective System
A larger-scale version of the secondary selective system is the transformer sparing scheme, as shown in figure 9.
This type of system allows good flexibility in switching. The system is usually operated with all of the secondary tie
breakers except one (the sparing transformer secondary main/tie breaker) normally-open. The sparing transformer
5
secondary main/tie breaker) normally-open. The sparing transformer supplies one load bus if a transformer fails or
is taken off-line for maintenance. A transformer is switched out of the circuit by opening its secondary main
breaker and closing the tie breaker to allow the sparing transformer to feed its loads. The sparing transformer may
be allowed to feed multiple load busses if it is sized properly. Care must be used when allowing multiple
transformers to be paralleled as the fault current is increased with each transformer that is paralleled, and
directional relaying is required on the secondary main circuit breakers to selectively isolate a faulted transformer.
An electrical or key interlock scheme is required to enforce the proper operating modes of this type of system,
especially in light of the fact that the switching is carried out over several pieces of equipment that can be in
different locations from one another. A properly-designed interlocking system will allow for the addition of future
substations without modification of the existing interlocking.
With both types of secondary-selective system, an automatic transfer scheme may be utilized to switch between a
failed transformer and an available transformer.
Figure 5-9: Transformer Sparing Scheme
Primary-Selective system
A selective system arrangement may also utilize the primary system equipment. Such an arrangement is shown in
figure 5-10.
As with the secondary selective system, an automatic transfer scheme may be used to automatically perform
the required transfer operations, should a utility source become unavailable. The bus tie circuit breaker may be
normally-closed or normally-open, depending upon utility allowances. If the bus tie circuit breaker is
normally-closed care must be taken in the protective relaying to insure that a fault on one utility line does not
cause the entire system to be taken off-line. The available fault current with the tie breaker normally closed
increases with each utility service added to the system.
Metal-Clad switchgear is most commonly used with this type of arrangement, due to the limitations of
metal-enclosed load interrupter switches.
Figure 5-10: Primary-Selective System
Secondary Spot-Network system

In large municipal areas where large loads, such as high-rise buildings, must be served and a high degree of
reliability is required, secondary network systems are often used. In a secondary network system several utility
services are paralleled at the low voltage level, creating a highly reliable system.
Network protectors are used at the transformer secondaries to isolate transformer faults which are backfed
through the low voltage system. These devices are designed to automatically isolate a faulted transformer which
is backfed from the rest of the system. The transformers typically have higher-than-standard impedances to
reduce the available fault current on the low voltage network. The common secondary bus is often referred to as
the collector bus. An example of a secondary spot-network system is shown in figure 5-11.
Figure 5-11: Secondary Spot Network
Ring Bus system

Essentially a loop system in which the loop is normally closed, the ring bus is a highly reliable system
arrangement. A typical ring-bus system is depicted in figure 5-12.
A fault at any bus causes only the loads served by that bus to lose service. Bus differential relaying is
recommended for optimum reliability with this scheme. The bus differential relaying will open both breakers
feeding a bus for a fault on that bus. Metal-clad switchgear is usually used for the primary ring bus.
Although figure 5-12 shows two utility sources, this system arrangement can be easily expanded to incorporate
additional utility sources. As with the primary-selective system with a normally-closed bus tie breaker, the available
fault current is increased with each utility source added to the system.
Figure 5-12: Primary Ring Bus System
Composite systems
The above system arrangements are the basic building blocks of power distribution system topologies, but are
rarely used alone for a given system. To increase system reliability it is usually necessary to combine two or more
of these arrangements. For example, one commonly-used arrangement is shown in figure 5-13.
As can be seen, a fault on a primary loop cable or the failure of one transformer can be accommodated without
loss of service to either load bus (but with an outage to part of the system until the system is switched to
accommodate the failure). In addition, a single section of the primary loop or one transformer can be taken out of
service while maintaining service to the loads.
The system of figure 5-13 can be expanded by the addition of an additional utility source and a primary bus tie
breaker to form an even more reliable system, as shown in figure 5-14. With this arrangement, the failure of a
single utility source, a single primary circuit breaker, a single loop feeder cable, or a single transformer can be
accommodated without loss of service. And, any one primary circuit breaker, any one section of the primary
distribution loop, or any one transformer can be taken out of service without loss of service to the loads.
However, the cost of a second utility service and two additional metal-clad breakers must be taken into account.
Figure 5-13: Composite System Primary Loop/Secondary-Selective
A logical expansion of this system, resulting in a further increase in system reliability, can be had by replacing the
primary distribution loop with dedicated feeder circuit breakers from each primary bus, as shown in figure 5-15. In
this system arrangement multiple primary feeder cable failures can be accommodated without jeopardizing service
to the loads (an outage will be taken until the system is switched to accommodate the failures, however).
An example of an extremely reliable system arrangement is given in figure 16. Note that figure 5-16 is a
re-arrangement of the primary ring-bus configuration shown in figure 5-12, along with the primary source-selective
configuration shown in figure 5-3 and a variant of the transformer sparing scheme given in figure 5-9. This system
arrangement gives good flexibility in switching for maintenance purposes, and also allows any one utility, primary
switchgear bus, or transformer fail without loss of service to any of the loads (again, an outage may be taken until
the system is switched to accommodate the failure, depending upon the failure under consideration). It also allows
any three primary feeders to be faulted without loss of service to any of the loads. Other composite arrangements
are possible.
Figure 5-14: Composite System Primary Selective/Primary Loop/Secondary Selective
Figure 5-15: Composite System Primary Double-Selective,/Secondary Selective
Summary
Various system arrangements have been presented in this section, starting with the least complex and
progressing to a very complex, robust system arrangement. In general, as reliability increases so does complexity
and cost. It must be remembered that economic considerations will usually dictate how complex a system
arrangement can be used, and thus will have a great deal of impact on how reliable the system is. Tables 5-6 and
5-7 show the features of each system arrangement given in this section.
Figure 5-15: Composite System Primary Double-Selective,/Secondary Selective
Please note that the formulas given in these tables are for the systems as shown in the figures above. They will
hold true for expanded versions of these system arrangements where the expansion is made symmetrically
with respect to the configuration shown. They will not hold true when modifications are made to the system
arrangements with respect to symmetry, with altered numbers of switching/protective devices, or for concurrent
failures of different types of system components. When in doubt regarding a system which is derived from,
but not identical, to the systems shown in the figures above, double-check these numbers.
From a maintenance perspective, the number of system elements that can be taken down for maintenance is the
same as the number that can fail while maintaining service to the loads.
These tables do not attempt to address concurrent failures of different types of system components, nor are they
a guarantee of loss of service to a particular load after a component failure while the system is being switched to
an alternate configuration. However, they are a guide to the relative strengths and weaknesses of each of the
system arrangements presented.
Table 5-6: Power system arrangement summary for the basic arrangements as shown
in this section
U
PB
SF
T
SB
$
= Number of Utility Sources

= Number of Primary Circuit Breakers
= Number of Primary Feeders
= Number of Transformers
= Number of Secondary Main and Tie Circuit Breakers
= Relative Cost, with $=Least Expensive
Arrangement
Utility
Failures
Allowed
Pri. Bkr
Failures
Allowed
Pri. Feeder
Failures
Allowed
Transformer
Failures
Allowed
Sec.
Main/Tie
Bkr
Failures
Allowed
U-1 O
$+
Expanded Radial,
Single Primary Feeder
$$
Expanded Radial,
Multiple Primary Feeders
$$
Expanded Radial,
Multiple Utility Sources,
Multiple Primary Feeders
U-1 O
$$+
Primary Loop System
$$$
Secondary-Selective System
$$$
Transformer Sparing Scheme
Varies;
Maximum of T-1
TL
$$$$
U-1 O,N
PB-F-U O,N,
$$$$$
Secondary Spot Network
U-1 O,N,$,
PB-1 O,N,$,
F-1 O,N,$,
T-1 O,N,$,
SB-1 O,N,$,
$$$$$
Primary Ring Bus
U-1 O,N,M
U O,N,,M
$$$$$$
Radial
Radial w/ Primary Selectivity
Primary Selective
Assumes that each utility source has sufficient capacity to supply the entire system.
Assumes that all secondary circuit breakers, including feeder breakers, are interchangeable.
N Assumes that each primary main and bus tie (if applicable) circuit breakers has sufficient capacity to
supply the entire system.
Assumes that all primary circuit breakers, including feeder breakers, are interchangeable.
$ Assumes that each primary feeder has sufficient capacity to supply the entire system.
Assumes that each transformer, secondary main and bus tie (if applicable) circuit breaker have sufficient
capacity to supply the entire system.
M Assumes that the ring bus has sufficient capacity to supply the entire system.
L
10
Cost
Table 7:
Power system arrangement summary for the composite arrangements as shown

in this section
U
PB
SF
T
SB
$
= Number of Utility Sources

= Number of Primary Circuit Breakers
= Number of Primary Feeders
= Number of Transformers
= Number of Secondary Main and Tie Circuit Breakers
= Relative Cost, with $=Least Expensive
Arrangement
Utility
Failures
Allowed
Pri. Bkr
Failures
Allowed
Pri. Feeder
Failures
Allowed
Transformer
Failures
Allowed
Sec.
Main/Tie
Bkr
Failures
Allowed
Primary Double-Selective /
Secondary-Selective
U-1 O,N
PB-F/2-U
F/2
T-1
T-1 ,L
$$$$$$$$
F/2
T-1
T ,L
$$$$$$$$+
Primary Ring Bus /

Primary-Selective/
Secondary-Selective
Cost
O,N,
U-1 O,N,M
PB-F/2-U+1
O,N,,M
Assumes that each utility source has sufficient capacity to supply the entire system.
Assumes that all secondary circuit breakers, including feeder breakers, are interchangeable.
N Assumes that each primary main and bus tie (if applicable) circuit breakers has sufficient capacity to
supply the entire system.
Assumes that all primary circuit breakers, including feeder breakers, are interchangeable.
$ Assumes that each primary feeder has sufficient capacity to supply the entire system.
Assumes that each transformer, secondary main and bus tie (if applicable) circuit breaker have sufficient
capacity to supply the entire system.
M Assumes that the ring bus has sufficient capacity to supply the entire system.
L
11
Section 6:
System Grounding
Introduction
The topic of system grounding is extremely important, as it affects the susceptibility of the system to voltage
transients, determines the types of loads the system can accommodate, and helps to determine the system
protection requirements.
The system grounding arrangement is determined by the grounding of the power source. For commercial and
industrial systems, the types of power sources generally fall into four broad categories:
A Utility Service The system grounding is usually determined by the secondary winding configuration of the
upstream utility substation transformer.
B Generator The system grounding is determined by the stator winding configuration.
C Transformer The system grounding on the system fed by the transformer is determined by the transformer
secondary winding configuration.
D Static Power Converter For devices such as rectifiers and inverters, the system grounding is determined by
the grounding of the output stage of the converter.
Categories A to D fall under the NEC definition for a separately-derived system. The recognition of a separatelyderived system is important when applying NEC requirements to system grounding, as discussed below.
All of the power sources mentioned above except D are magnetically-operated devices with windings.
To understand the system voltage relationships with respect to system grounding, it must be recognized that there
are two common ways of connecting device windings: wye and delta. These two arrangements, with their system
voltage relationships, are shown in figure 6-1. As can be seen from the figure, in the wye-connected arrangement
there are four terminals, with the phase-to-neutral voltage for each phase set by the winding voltage and the
resulting phase-to-phase voltage set by the vector relationships between the voltages. The delta configuration
has only three terminals, with the phase-to-phase voltage set by the winding voltages and the neutral terminal
not defined.
Neither of these arrangements is inherently associated with any particular system grounding arrangement,
although some arrangements more commonly use one arrangement vs. the other for reasons that will be
explained further below.
Figure 6-1: Wye and delta winding configurations and system voltage relationships
Solidly-grounded systems
The solidly-grounded system is the most common system arrangement, and one of the most versatile. The
most commonly-used configuration is the solidly-grounded wye, because it will support single-phase phase-toneutral loads.
The solidly-grounded wye system arrangement can be shown by considering the neutral terminal from the
wye system arrangement in figure 6-1 to be grounded. This is shown in figure 6-2:
Figure 6-2: Solidly-Grounded Wye System arrangement and voltage relationships
Several points regarding figure 6-2 can be noted.

First, the system voltage with respect to ground is fixed by the phase-to-neutral winding voltage. Because parts of
the power system, such as equipment frames, are grounded, and the rest of the environment essentially is at
ground potential also, this has big implications for the system. It means that the line-to-ground insulation level of
equipment need only be as large as the phase-to-neutral voltage, which is 57.7% of the phase-to-phase voltage.
It also means that the system is less susceptible to phase-to-ground voltage transients.
Second, the system is suitable for supplying line-to-neutral loads. The operation of a single-phase load connected
between one phase and neutral will be the same on any phase since the phase voltage magnitudes are equal.
This system arrangement is very common, both at the utilization level as 480 Y/277 V and 208 Y/120 V, and also
on most utility distribution systems.
While the solidly-grounded wye system is by far the most common solidly-grounded system, the wye arrangement
is not the only arrangement that can be configured as a solidly grounded system. The delta system can also be
grounded, as shown in figure 6-3. Compared with the solidly-grounded wye system of figure 6-2 this system
grounding arrangement has a number of disadvantages. The phase-to-ground voltages are not equal, and
therefore the system is not suitable for single-phase loads. And, without proper identification of the phases there is
the risk of shock since one conductor, the B-phase, is grounded and could be mis-identified. This arrangement is
no longer in common use, although a few facilities where this arrangement is used still exist.
Figure 6-3: Corner-Grounded Delta System arrangement and voltage relationships
The delta arrangement can be configured in another manner, however, that does have merits as a solidlygrounded system. This arrangement is shown in figure 6-4. While the arrangement of figure 6-4 may not appear at
first glance to have merit, it can be seen that this system is suitable both for three-phase and single-phase loads,
so long as the single-phase and three-phase load cables are kept separate from each other. This is commonly
2
used for small services which require both 240 VAC three-phase and 120/240 VAC single-phase. Note that the
phase A voltage to ground is 173% of the phase B and C voltages to ground. This arrangement requires the BC
winding to have a center tap.
Figure 6-4: Center-Tap-Grounded Delta System arrangement and voltage relationships
A common characteristic of all three solidly-grounded system shown here, and of solidly-grounded systems in
general, is that a short-circuit to ground will cause a large amount of short-circuit current to flow. This condition is
known as a ground fault and is illustrated in figure 6-5. As can be seen from figure 6-5, the voltage on the faulted
phase is depressed, and a large current flows in the faulted phase since the phase and fault impedance are small.
The voltage and current on the other two phases are not affected. The fact that a solidly-grounded system will
support a large ground fault current is an important characteristic of this type of system grounding and does affect
the system design. Statistically, 90-95% of all system short-circuits are ground faults so this is an important topic.
The practices used in ground-fault protection are described in a later section of this guide.
Figure 6-5: Solidly-Grounded System with a ground fault on phase A
The occurrence of a ground fault on a solidly-grounded system necessitates the removal of the fault as quickly
as possible. This is the major disadvantage of the solidly-grounded system as compared to other types of
system grounding.
A solidly-grounded system is very effective at reducing the possibility of line-to-ground voltage transients.
However, to do this the system must be effectively grounded. One measure of the effectiveness of the
system grounding is the ratio of the available ground-fault current to the available three-phase fault current.
For effectively-grounded systems this ratio is usually at least 60% [2].
Most utility systems which supply service for commercial and industrial systems are solidly grounded. Typical
utility practice is to ground the neutral at many points, usually at every line pole, creating a multi-grounded neutral
system. Because a separate grounding conductor is not run with the utility line, the resistance of the earth limits
the circulating ground currents that can be caused by this type of grounding. Because separate grounding
conductors are used inside a commercial or industrial facility, multi-grounded neutrals not preferred for power
systems in these facilities due to the possibility of circulating ground currents. As will be explained later in this
section, multi-grounded neutrals in NEC jurisdictions, such as commercial or industrial facilities, are actually
prohibited in most cases by the NEC [1]. Instead, a single point of grounding is preferred for this type of system,
creating a uni-grounded or single-point grounded system.
In general, the solidly-grounded system is the most popular, is required where single-phase phase-to-neutral loads
must be supplied, and has the most stable phase-to-ground voltage characteristics. However, the large ground
fault currents this type of system can support, and the equipment that this necessitates, are a disadvantage and
can be hindrance to system reliability.
Ungrounded systems
This system grounding arrangement is at the other end of the spectrum from solidly-grounded systems.
An ungrounded system is a system where there is no intentional connection of the system to ground.
The term ungrounded system is actually a misnomer, since every system is grounded through its inherent
charging capacitance to ground. To illustrate this point and its effect on the system voltages to ground, the delta
winding configuration introduced in figure 6-3 is re-drawn in figure 6-6 to show these system capacitances.
If all of the system voltages in figure 6-6 are multiplied by 3 and all of the phase angles are shifted by 30 (both
are reasonable operations since the voltage magnitudes and phase angles for the phase-to-phase voltage were
arbitrarily chosen), the results are the same voltage relationships as shown in figure 6-4 for the solidly-grounded
wye system. The differences between the ungrounded delta system and the solidly-grounded wye system, then,
are that there is no intentional connection to ground, and that there is no phase-to-neutral driving voltage on the
ungrounded delta system. This becomes important when the effects of a ground fault are considered. The lack of
a grounded system neutral also makes this type of system unsuitable for single-phase phase-to-neutral loads.
Figure 6-6: Ungrounded Delta System winding arrangement and voltage relationships
In figure 6-7, the effects of a single phase to ground fault are shown. The equations in figure 6-7 are not
immediately practical for use, however if the fault impedance is assumed to be zero and the system capacitive
charging impedance is assumed to be much larger than the phase impedances, these equations reduce into a
workable form. Figure 6-8 shows the resulting equations, and shows the current and voltage phase relationships.
As can be seen from figure 6-8, the net result of a ground fault on one phase of an ungrounded delta system is a
change in the system phase-to-ground voltages. The phase-to-ground voltage on the faulted phase is zero, and
the phase-to-ground voltage on the unfaulted phases are 173% of their nominal values. This has implications for
power equipment the phase-to-ground voltage rating for equipment on an ungrounded system must be at least
equal the phase-to-phase voltage rating. This also has implications for the methods used for ground detection, as
explained later in this guide.
4
Figure 6-7: Ungrounded Delta System with a ground-fault on one phase
Figure 6-8: Ungrounded Delta System simplified ground fault voltage and current relationships
The ground currents with one phase is faulted to ground are essentially negligible. Because of this fact, from an
operational standpoint ungrounded systems have the advantage of being able to remain in service if one phase is
faulted to ground. However, suitable ground detection must be provided to alarm this condition (and is required in
most cases by the NEC [1] as described below). In some older facilities, it has been reported that this type of
system has remained in place for 40 years or more with one phase grounded! This condition is not dangerous in
and of itself (other than due to the increased phase-to-ground voltage on the unfaulted phases), however if a
ground fault occurs on one of the ungrounded phases the result is a phase-to-phase fault with its characteristic
large fault current magnitude.
Another important consideration for an ungrounded system is its susceptibility to large transient overvoltages.
These can result from a resonant or near-resonant condition during ground faults, or from arcing [2]. A resonant
ground fault condition occurs when the inductive reactance of the ground-fault path approximately equals the
system capacitive reactance to ground. Arcing introduces the phenomenon of current-chopping, which can cause
excessive overvoltages due to the system capacitance to ground.
The ground detection mentioned above can be accomplished through the use of voltage transformers connected
in wye-broken delta, as illustrated in figure 6-9.
A
VT
LT A
VT
LT B
LT M
VT
LT C
GROUND
FAULT
LOCATION
LTA
LTB
LTC
NONE
DIM
DIM
DIM
OFF
PHASE A
OFF
BRIGHT
BRIGHT
BRIGHT
PHASE B
BRIGHT
OFF
BRIGHT
BRIGHT
PHASE C
BRIGHT
BRIGHT
DIM
BRIGHT
LTM
Figure 6-9: A Ground Detection method for ungrounded systems
In figure 6-9, three ground detection lights LTA, LTB and LTC are connected so that they indicate the A, B and
C phase-to-ground voltages, respectively. A master ground detection light LTM indicates a ground fault on any
phase. With no ground fault on the system LTA, LTB and LTB will glow dimly. If a ground fault occurs on one
phase, the light for that phase will be extinguished and LTM will glow brightly along with the lights for the other
two phases. Control relays may be substituted for the lights if necessary. Resistor R is connected across the
broken-delta voltage transformer secondaries to minimize the possibility of ferroresonance. Most ground detection
schemes for ungrounded systems use this system or a variant thereof.
Note that the ground detection per figure 6-10 indicates on which phase the ground fault occurs, but not
where in the system the ground fault occurs. This, along with the disadvantages of ungrounded systems
due to susceptibility to voltage transients, was the main impetus for the development of other ground
system arrangements.
Modern power systems are rarely ungrounded due to the advent of high-resistance grounded systems as
discussed below. However, older ungrounded systems are occasionally encountered.
High-resistance grounded systems

One ground arrangement that has gained in popularity in recent years is the high-resistance grounding
arrangement. For low voltage systems, this arrangement typically consists of a wye winding arrangement with the
neutral connected to ground through a resistor. The resistor is sized to allow 1-10 A to flow continuously if a
ground fault occurs. This arrangement is illustrated in figure 6-10.
Figure 6-10: High-Resistance Grounded System with no ground fault present
The resistor is sized to be less than or equal to the magnitude of the system charging capacitance to ground. If
the resistor is thus sized, the high-resistance grounded system is usually not susceptible to the large transient
overvoltages that an ungrounded system can experience. The ground resistor is usually provided with taps to
allow field adjustment of the resistance during commissioning.
If no ground fault current is present, the phasor diagram for the system is the same as for a solidly-grounded wye
system, as shown in figure 6-10. However, if a ground fault occurs on one phase the system response is as
shown in figure 6-11. As can be seen from figure 6-11, the ground fault current is limited by the grounding resistor.
If the approximation is made that ZA and ZF are very small compared to the ground resistor resistance value R,
which is a good approximation if the fault is a bolted ground fault, then the ground fault current is approximately
equal to the phase-to-neutral voltage of the faulted phase divided by R. The faulted phase voltage to ground in
that case would be zero and the unfaulted phase voltages to ground would be 173% of their values without a
ground fault present. This is the same phenomenon exhibited by the ungrounded system arrangement, except
that the ground fault current is larger and approximately in-phase with the phase-to-neutral voltage on the faulted
phase. The limitation of the ground fault current to such a low level, along with the absence of a solidly-grounded
system neutral, has the effect of making this system ground arrangement unsuitable for single-phase line-toneutral loads.
Figure 6-11: High-Resistance Grounded System with a ground fault on one phase
The ground fault current is not large enough to force its removal by taking the system off-line. Therefore, the
high-resistance grounded system has the same operational advantage in this respect as the ungrounded system.
However, in addition to the improved voltage transient response as discussed above, the high-resistance
grounded system has the advantage of allowing the location of a ground fault to be tracked.
A typical ground detection system for a high-resistance grounded system is illustrated in figure 6-12. The ground
resistor is shown with a tap between two resistor sections R1 and R2. When a ground fault occurs, relay 59 (the
ANSI standard for an overvoltage relay, as discussed later in this guide) detects the increased voltage across the
resistor. It sends a signal to the control circuitry to initiate a ground fault alarm by energizing the alarm indicator.
When the operator turns the pulse control selector to the ON position, the control circuit causes pulsing contact
P to close and re-open approximately once per second. When P closes R2 is shorted and the pulse indicator is
energized. R1 and R2 are sized so that approximately 5-7 times the resistor continuous ground fault current flows
when R2 is shorted. The result is a pulsing ground fault current that can be detected using a clamp-on ammeter
(an analog ammeter is most convenient). By tracing the circuit with the ammeter, the ground fault location can be
determined. Once the ground fault has been removed from the system pressing the alarm reset button will
de-energize the alarm indicator.
This type of system is known as a pulsing ground detection system and is very effective in locating ground
faults, but is generally more expensive than the ungrounded system ground fault indicator in figure 6-10.
Figure 6-12: Pulsing Ground Detection System
For medium voltage systems, high-resistance grounding is usually implemented using a low voltage resistor and
a neutral transformer, as shown in figure 6-13.
Figure 6-13: Medium Voltage implementation for high-resistance grounding
Reactance grounding
In industrial and commercial facilities, reactance grounding is commonly used in the neutrals of generators. In
most generators, solid grounding may permit the level of ground-fault current available from the generator to
exceed the three-phase value for which its windings are braced [2]. For these cases, grounding of the generator
neutral through an air-core reactance is the standard solution for lowering the ground fault level. This reactance
ideally limits the ground-fault current to the three-phase available fault current and will allow the system to operate
with phase-to-neutral loads.
Low-resistance grounded systems

By sizing the resistor in figure in 6-11 such that a higher ground fault current, typically 200-800 A, flows during a
ground fault a low-resistance grounded system is created. The ground fault current is limited, but is of high
enough magnitude to require its removal from the system as quickly as possible. The low-resistance grounding
arrangement is typically used in medium voltage systems which have only 3-wire loads, such as motors, where
limiting damage to the equipment during a ground fault is important enough to include the resistor but it is
acceptable to take the system offline for a ground fault. The low-resistance grounding arrangement is generally
less expensive than the high-resistance grounding arrangement but more expensive than a solidly grounded
system arrangement.
Creating an artificial neutral in an ungrounded system

In some cases it is required to create a neutral reference for an ungrounded system. Most instances involve
existing ungrounded systems which are being upgraded to high-resistance grounding. The existence of
multiple transformers and/or delta-wound generators may make the replacement of this equipment
economically unfeasible.
The solution is a grounding transformer. Although several different configurations exist, by far the most popular in
commercial and industrial system is the zig-zag transformer arrangement. It uses transformers connected as
shown in figure 6-14:
Figure 6-14: Zig-Zag grounding transformer arrangement
The zig-zag transformer will only pass ground current. Its typical implementation on an ungrounded system, in
order to convert the system to a high-resistance grounded system, is shown in figure 6-15. The zig-zag
transformer distributes the ground current IG equally between the three phases. For all practical purposes the
system, from a grounding standpoint, behaves as a high-resistance grounded system.
Figure 6-15: Zig-Zag grounding transformer implementation
The solidly-grounded and low-resistance grounded systems can also be implemented by using a grounding
transformer, depending upon the amount of impedance connected in the neutral.
NEC system grounding requirements

The National Electrical Code [1] does place constraints on system grounding. While this guide is not intended to
be a definitive guide to all NEC requirements, several points from the NEC must be mentioned and are based
upon the basic principles stated above. As a starting point, several key terms from the NEC need to be defined:
Ground: A conducting connection, whether intentional or accidental, between an electrical circuit or equipment
and the earth or to some body that serves in place of the earth.
Grounded: Connected to earth or to some body that serves in place of the earth.
Effectively Grounded: Intentionally connected to earth through a ground connection or connections of

sufficiently low impedance and having sufficient current-carrying capacity to prevent the buildup of voltages that
may result in undue hazards to connected equipment or to persons.
Grounded Conductor: A system or circuit conductor that is intentionally grounded.
Solidly Grounded: Connected to ground without inserting any resistor or impedance device.
Grounding Conductor: A conductor used to connect equipment or the grounded circuit of a wiring system to a
grounding electrode or electrodes.
Equipment Grounding Conductor: The conductor used to connect the non-current-carrying metal parts of
equipment, raceways and other enclosures to the system grounded conductor, grounding electrode conductor, or
both, at the service equipment or at the source of a separately-derived system.
Main Bonding Jumper: The connection between the grounded circuit conductor and the equipment grounding
conductor at the service.
System Bonding Jumper: The connection between the grounded circuit conductor and the equipment grounding
conductor at a separately-derived system.
Grounding Electrode: The conductor used to connect the grounding electrode(s) to the equipment grounding
conductor, to the grounded conductor, or to both, at the service, at each building or structure where supplied by
a feeder(s) or branch circuit(s), or at the source of a separately-derived system.
Grounding Electrode Conductor: The conductor used to connect the grounding electrode(s) to the equipment
grounding conductor, to the grounded conductor, or to both, at the service, at each building or structure where
supplied by a feeder(s) or branch circuit(s), or at the source of a separately-derived system.
Ground Fault: An unintentional, electrically conducting connection between an ungrounded conductor of
an electrical circuit and the normally noncurrent-carrying conductors, metallic enclosures, metallic raceways,
metallic equipment, or earth.
Ground Fault Current Path: An electrically conductive path from the point of a ground fault on a wiring system
through normally noncurrent-carrying conductors, equipment, or the earth to the electrical supply source.
Effective Ground-Fault Current Path: An intentionally constructed, permanent, low-impedance electrically
conductive path designed and intended to carry current under ground-fault conditions from the point of a ground
fault on a wiring system to the electrical supply source and that facilitates the operation of the overcurrent
protective device or ground fault detectors on high-impedance grounded systems.
Ground-Fault Circuit Interrupter: A device intended for the protection of personnel that functions to
de-energize a circuit or portion thereof within an established period of time when a current to ground exceeds
the values established for a Class A device.
FPN: Class A ground-fault circuit interrupters trip when the current to ground has a value in the range of 4 mA to
6 mA. For further information, see UL 943, Standard for Ground-Fault Circuit Interrupters.
Ground Fault Protection of Equipment: A system intended to provide protection of equipment from damaging
line-to-ground fault currents by operating to cause a disconnecting means to open all ungrounded conductors of
the faulted circuit. This protection is provided at current levels less than those required to protect conductors from
damage through the operation of a supply circuit overcurrent device.
Qualified Person: One who has the skills and knowledge related to the construction and operation of the
electrical equipment and installations and has received safety training on the hazards involved.
10
With these terms defined, several of the major components of the grounding system can be illustrated by
redrawing the system of figure 6-2 and labeling the components:
Figure 6-16: NEC [1] system grounding terms illustration
Several key design constraints for grounding systems from the NEC [1] are as follows. These are paraphrased
from the code text (Note: This guide is not intended as a substitute for familiarity with the NEC, nor is it intended
as an authoritative interpretation of every aspect of the NEC articles mentioned.):
I
Electrical systems that are grounded must be grounded in such a manner as to limit the voltage imposed by
lightning, line surges, or unintentional contact with higher voltage lines and that will stabilize the voltage to earth
during normal operation [Article 250.4(A)(1)]. In other words, if a system is considered solidly grounded the
ground impedance must be low.
If the system can be solidly grounded at 150 V to ground or less, it must be solidly grounded [Article 250.20(B)].
There is therefore no such system as a 120 V Ungrounded Delta in use, even though such a system is
physically possible.
If the system neutral carries current it must be solidly grounded [Article 250.20(B)]. This is indicative of
single-phase loading and is typical for a 4-wire wye (such as figure 6-2) or center-tapped 4-wire delta
(such as figure 6-4) system.
Certain systems are permitted, but not required, to be solidly grounded. They are listed as electric systems used
exclusively to supply industrial electric furnaces for melting, refining, tempering, and the like, separately derived
systems used exclusively for rectifiers that supply only adjustable-speed industrial drives, and separately
derived systems supplied by transformers that have a primary voltage rating less than 1000 volts provided that
certain conditions are met [Article 250.21].
If a system 50-1000 VAC is not solidly-grounded, ground detectors must be installed on the system unless the
voltage to ground is less than 120 V [Article 250.21].
Certain systems cannot be grounded. They are listed as circuits for electric cranes operating over combustible
fibers in Class III locations as provided in Article 503.155, circuits within hazardous (classified) anesthetizing
locations and other isolated power systems in health care facilities as provided in 517.61 and 517.160, circuits
for equipment within electrolytic cell working zone as provided in Article 668, and secondary circuits of lighting
systems as provided in 411.5(A) [Article 250.22]. Some of the requirements for hazardous locations and health
care facilities are covered in section XVI.
For solidly-grounded systems, an unspliced main bonding jumper must be used to connect the equipment
grounding conductor(s) and the service disconnect enclosure to the grounded conductor within the enclosure
for each utility service disconnect [Article 250.24(B)].
For solidly-grounded systems, an unspliced system bonding jumper must be used to connect the equipment
grounding conductor of a separately derived system to the grounded conductor. This connection must be made
at any single point on the separately derived system from the source to the first system disconnecting means or
overcurrent device [250.30(A)(1)]
A grounding connection on the load side of the main bonding or system bonding jumper on a solidly-grounded
system is not permitted [Articles 240.24(A)(5), 250.30(A)]. The reasons for this are explained in below and in
section VIII.
11
Ground fault protection of equipment must be provided for solidly grounded wye electrical services, feeder
disconnects on solidly-grounded wye systems, and building or structure disconnects on solidly-grounded wye
systems under the following conditions:
N The voltage is greater than 150 V to ground, but does not exceed 600 V phase-to-phase.
N
The utility service, feeder, or building or structure disconnect is rated 1000 A or more.
The disconnect in question does not supply a fire pump or continuous industrial process.
[Articles 215.10, 230.95, 240.13].

I
Where ground fault protection is required per Article 215.10 or 230.95 for a health care facility, an additional step
of ground fault protection is required in the next downstream device toward the load, with the exception of
circuits on the load side of an essential electrical system transfer switch and between on-site generating units for
the essential electrical system and the essential electrical system transfer switches [Article 517.17]. Specific
requirements for health-care systems are described in a later section of this guide.
The alternate source for an emergency or legally-required standby system is not required to have ground fault
protection. For an emergency system, ground-fault indication is required [Articles 700.26, 701.17]. A later
section of this guide describes the requirements for Emergency and Standby Power Systems.
All electrical equipment, wiring, and other electrically conductive material must be installed in a manner that
creates a permanent, low-impedance path facilitating the operation of the overcurrent device. This circuit must
be able to safely carry the ground fault current imposed upon it. [Article 250.4(A)(5)]. The intent of this
requirement is to allow ground fault current magnitudes to be sufficient for the ground fault protection/detection
to detect (and for ground fault protection to clear) the fault, and for a ground fault not to cause damage to the
grounding system.
High-impedance grounded systems may utilized on AC systems of 480-1000 V where:

N Conditions of maintenance and supervision ensure that only qualified persons access the installation.
N
Continuity of power is required.
Ground detectors are installed on the system.
Line-to-neutral loads are not served.
[Article 250.36]
I
For systems over 1000 V:

N The system neutral for solidly-grounded systems may be a single point grounded or multigrounded neutral.
Additional requirements for each of these arrangements apply [Article 250.184].
N
The system neutral derived from a grounding transformer may be used for grounding [Article 250.182].
The minimum insulation level for the neutral of a solidly-grounded system is 600 V. A bare neutral is
permissible under certain conditions [Article 250.184 (A) (1)].
Impedance grounded neutral systems may be used where conditions 1, 3, and 4 for the use of highimpedance grounding on systems 480-1000 V above are met [Article 250.186].
The neutral conductor must be identified and fully insulated with the same phase insulation as the phase
conductors [Article 250.186 (B)].
Zig-zag grounding transformers must not be installed on the load side of any system grounding connection
[Article 450.5].
When a grounding transformer is used to provide the grounding for a 3 phase 4 wire system, the grounding
transformer must not be provided with overcurrent protection independent of the main switch and common-trip
overcurrent protection for the 3 phase, 4 wire system [Article 450.5 (A) (1)]. An overcurrent sensing device must
be provided that will cause the main switch or common-trip overcurrent protection to open if the load on the
grounding transformer exceeds 125% of its continuous current rating [Article 450.5 (A) (2)].
Again, these points are not intended to be an all-inclusive reference for NEC grounding requirements. They do,
however, summarize many of the major requirements. When in doubt, consult the NEC.
12
References
[1]
[2]
IEEE Recommended Practice for Grounding of Industrial and Commercial Power Systems, IEEE Std.
142-1991, December 1991.
13
Section 7:
System Protection
Introduction
An important consideration in power system design is system protection. Without system protection, the power
system itself, which is intended to be of benefit to the facility in question, would itself become a hazard.
The major concern for system protection is protection against the effects of destructive, abnormally high currents.
These abnormal currents, if left unchecked, could cause fires or explosions resulting in risk to personnel and
damage to equipment. Other concerns, such as transient overvoltages, are also considered when designing
power system protection although they are generally considered only after protection against abnormal currents
has been designed.
Characterization of power system faults

Any current in excess of the rated current of equipment or the ampacity of a conductor may be considered an
overcurrent. Overcurrents can generally be categorized as overloads or faults. An overload is a condition where
load equipment draws more current that the system can safely supply. The main hazard with overload conditions
is the thermal heating effects of overloaded equipment and conductors. Faults are unintentional connections of the
power system which result in overcurrents much larger in magnitude than overloads.
Faults can be categorized in several different ways. A fault with very little impedance in the unintended connection
is referred to as a short circuit or bolted fault (the latter term is used due to the fact that a short circuit can be
thought of as a bus bar inadvertently bolted across two phase conductors or from phase to ground). A fault to
ground is referred to as a ground fault. A fault between all three phases is referred to as a 3 phase fault. A fault
between two phases is referred to as a phase-to-phase fault. A fault which contains enough impedance in the
unintentional connection to significantly affect the fault current vs. a true short circuit is known as an impedance
fault. An arcing fault has the unintentional connection made via an electrical arc through an ionized gas such as
air. All of these terms are used in practice to characterize the nature of a fault.
In order to quantitatively characterize a fault, it is necessary to calculate how much fault current could be
produced at a given location in the system. In most cases this will be the three-phase short-circuit current, which
is the current produced if all three phases were shorted to each other and/or to ground. The simplest method for
illustrating this is to reduce the power system at the point in question to its Thevenin equivalent. The Thevenin
equivalent is the equivalent single voltage source and impedance that produce the same short-circuit results as
the power system itself. The Thevenin equivalent voltage Vth is the open-circuit voltage at the point in question,
and the Thevenin equivalent impedance Zth is the impedance of the power system at the point in question with
the source voltage equal to zero. If a further simplification is made such that the system can be reduced to its
single-phase equivalent, then a simple 3-phase fault current calculation for the three-phase fault current If3
can be performed as shown in figure 7-1:
Z th
V th = V ln
I f 3 =
V ln
Z th
Figure 7-1: Simplified 3-phase fault calculation
The Thevenin impedance for a power system at a given point in the system is referred to as the short-circuit
impedance. In the great majority of power systems the short-circuit impedance is predominately inductive,
therefore one simplification that is often made is to treat the impedance purely as inductance. This has the
effect of causing the fault current to lag the system line-to-neutral voltage by 90. If the system is an ungrounded
delta system the equivalent line-to-neutral voltage can be obtained by performing a delta-wye conversion of
the source voltage.
The phase-to-phase fault value can be calculated from the three-phase fault value if it is remembered that the
line-to-line voltage magnitude is equal to the line-to-neutral voltage magnitude multiplied by 3, and that there will
be twice the impedance in the circuit since the return path must be considered. These two facts, taken together,
allow computation of the line-to-line fault current magnitude I f l l as:
I
f l l
3 I
f 3
(7-1)
This, however, is as far as this simplified analysis method will take us. In order to further characterize fault
currents, a method for calculating unbalanced faults must be used. The universally-accepted method for this is
a method known as the method of symmetrical components.
In the method of symmetrical components, unbalanced currents and voltages are broken into three distinct
components: positive sequence, negative sequence, and zero sequence.These sequence components can be
thought of as independent sets of rotating balanced phasors. The positive sequence set rotates in the standard
A-B-C phase rotation. The negative sequence set rotates in the negative or C-B-A phase rotation. In the zero
sequence set all three phase components are in phase with one another. The positive, negative and zero
sequence components can be further simplified by referring only to the A-phase phasor of each set; these are
referred to as V1 for the positive sequence set, V2 for the negative sequence set and V0 for the zero-sequence
set. For a given set of phase voltages Va, Vb and Vc, the sequence components are related to the phase voltages
as follows:
V1 =
(7-2)
V2
(7-3)
V0
1
V a + aV b + a 2 V c
3
1
=
V a + a 2 V b + aV c
3
1
=
V a +V b +V c
3
(7-4)
V a =V 1 +V 2 +V 0
(7-5)
V b = a 2 V 1 + aV 2 + V 0
(7-6)
V c = aV 1 + a 2 V 2 + V 0
(7-7)
where
a
= 1<120
The system may be separated into positive, negative, and zero-sequence networks depending upon the fault type
and the resulting sequence quantities then combined per (7-5), (7-6), and (7-7) to yield the phase values.
Modern short-circuit analysis is performed using the computer. Even large systems can be quickly analyzed
via short-circuit analysis software. Even so, some heuristic benefit can be gained by knowing how the method
of symmetrical components works. For example, certain protective relays are often set in terms of negativesequence values and ground currents are often referred to as zero-sequence quantities in the literature.
Another factor that must be taken into account is the existence of DC quantities in fault currents. Because of
the system inductance the current cannot change instantaneously, therefore upon initiation of a fault the system
must go through a transient condition which bridges the gap between the faulted and unfaulted conditions.
This transition involves DC currents. For a generic single-phase AC circuit with an open-circuit voltage
v ( t ) = Vm sin( t + ) and a short-circuit impedance consisting of resistance R and inductance L, the fault
current for a fault initiated at time t=0 can be expressed as [2]:
i( t ) =
t
sin( t + ) sin( )e L
R 2 + ( L ) 2
Vm
(7-5)
where
L
= tan 1
The angle can be recognized to be the angle of the Thevenin impedance. Several key points can be
taken from (7-5):
I
When the fault occurs such that ( - )= 0 no transient will occur. For a purely inductive circuit this would mean
that = 90 and thus the fault is initiated when the voltage is at its peak.
When the fault occurs such that ( - ) = 90 the maximum transient will occur. For a purely inductive circuit
this would mean that = 0 and thus the fault is initiated when the voltage is zero.
The time constant of the circuit is (L/R) and thus the higher the value of L/R the longer the transient will last.
Instead of (L/R) power systems typically are defined in terms of (X/R), where (X/R) is the ratio of the inductive
reactance of the short-circuit impedance to its resistance. Thus the higher (X/R) or the X/R ratio, the longer the
short-circuit transient will last. This has great implications on the rating of equipment.
A typical plot of fault current on a distribution system with a low X/R ratio and closing angle such that a small
transient is produced is shown in figure 7-2. In contrast with this is the plot shown in figure 7-3, which is the fault
current for a system with a high X/R ratio and closing angle of 0 such that there is a large transient.
1.5
1
i(t)
0.5
0
0.0005
-0.5
0.009
0.0175
0.026
0.0345
0.043
-1
-1.5
t(s)
Figure 7-2: Fault current for system with low X/R ratio and small-transient closing angle, normalized
to a steady-state magnitude of 1
2
1.5
i(t)
1
0.5
0
0.0005
-0.5
0.009
0.0175
0.026
0.0345
0.043
-1
t(s)
Figure 7-3: Fault current for system with higher X/R ratio and closing angle of 0, normalized
to a steady-state magnitude of 1
Figure 7-4 shows only the steady-state component of the waveform of figure 7-3, and figure 7-5 shows only the
transient component.
1.5
1
ss i(t)
0.5
0
0.0005
-0.5
0.009
0.0175
0.026
0.0345
0.043
-1
-1.5
t(s)
tran i(t)
Figure 7-4: Steady-state component of waveform in Figure 7-3

1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0.0005
0.009
0.0175
0.026
0.0345
0.043
t(s)
Figure 7-5: Transient component of waveform in figure 7-3
The fault current is often described in terms of its RMS Symmetrical and RMS Asymmetrical values. The RMS
symmetrical value is the RMS value considering the steady-state component only. The RMS asymmetrical value is
the RMS value over the first cycle considering both the steady-state and transient components at the worst-case
closing angle. As a simplification of (7-5) an approximate asymmetry factor can be calculated as [3]
2
Asymmetry factor = 1+ 2e
X
R
(7-9)
For example, this asymmetry factor for an X/R ratio of 25 is 1.6, meaning that the approximate worst-case RMS
asymmetrical value over the first cycle for the fault current at an X/R ratio of 25 will be no greater than the RMS
symmetrical value multiplied by 1.6.
For motors and generators, which have a high X/R ratios, calculations for the transient performance during a fault
are simplified by representing the short-circuit impedances differently for different time periods after the fault
initiation. The reactive component of the short-circuit impedance for the first half-cycle into the fault is the
subtransient reactance (X"d). For the first several cycles into the fault the reactance is larger and is termed the
transient reactance (X'd). For the long-term fault currents (up to 30 cycles or so into the fault) the reactance is
even larger and is termed the synchronous reactance (Xd). The synchronous reactance is much larger than either
the transient or subtransient reactance and models the phenomenon of AC decrement; after the DC component
decays the AC component continues to decay, eventually reaching a value that can be less than the generator
rated load current.
In general, the closer the fault is to a generator or generators the higher the X/R ratio and thus the larger the DC
offset. The AC decrement of the fault from generator sources is pronounced. Faults from most utility services are
sufficiently far removed from generation and have enough resistance in the distribution lines that there is less DC
offset and essentially no AC decrement. The fault current contribution from induction motors has a high DC offset
but also decays rapidly to zero over the first few cycles since there is no applied field excitation. The fault current
contribution from synchronous motors has a large DC component and decays to zero but at a slower rate than for
4
induction motors due to the applied field excitation. For a given point in the system, the fault current is the sum of
the contributions from all of these sources and the DC offset, DC decay, and AC decrement are all dependent
upon the relative location of the fault with respect to these sources.
The existence of the transient is of vital importance to selecting the proper equipment for system protection.
Because standards for equipment short-circuit ratings vary (more will be stated regarding this in subsequent
sections of this guide), all the more speed and efficiency is gained by using the computer for short circuit
calculations; the various equipment rating standards can be taken into account to produce accurate results for
comparison with the equipment ratings.
Low voltage fuses

The simplest of all overcurrent protective devices is the fuse. A fuse is an overcurrent protective device with a
circuit-opening fusible part that is heated and severed by the passage of the overcurrent through it [3].
Several definitions are of interest for low voltage fuses [3]:
Ampere rating: The RMS current that the fuse can carry continuously without deterioration and without
exceeding temperature rise limits. In accordance with NEC [1] article 210.20 [1] a fuse (or any branch-circuit
overcurrent device) should not be loaded continuously to over 80% of its ampere rating unless the assembly,
including the fuse and enclosure, is listed for operation at 100% of its rating.
Current-limiting fuse: A current-limiting fuse interrupts all available currents its threshold current and below its
maximum interrupting rating, limits the clearing time at rated voltage to an interval equal to or less than the first
major or symmetrical loop duration, and limits peak let-through current to a value less than the peak current that
would be possible with the fuse replaced by a solid conductor of the same impedance.
Dual-element fuse: A cartridge fuse having two or more current-responsive elements in series in a single
cartridge. The dual-element design is a construction technique frequently used to obtain a desired time-delay
response characteristic.
I2t: A measure of heat energy developed within a circuit during the fuses melting or arcing. The sum of melting
and arcing I2t is generally stated as total clearing I2t.
Interrupting rating: The rating based upon the highest RMS current that the fuse is tested to interrupt under the
conditions specified.
Melting time: The time required to melt the current-responsive element on a specified overcurrent.
Peak let-through current (Ip): The maximum instantaneous current through a current-limiting fuse during the
total clearing time.
Time delay: For Class H, K, J, and R fuses, a minimum opening time of 10s to an overload current five times the
ampere rating of the fuse, except for Class H, K, and R fuses rated 0-30 A, 250 V, in which case the opening time
can be reduced to 8s. For Class G, Class CC, and plug fuses, a minimum time delay of 12s on an overload of
twice the fuses ampere rating.
Total Clearing time: The total time between the beginning of the specified overcurrent and the final interruption
of the circuit, at rated voltage.
Voltage Rating: The RMS voltage at which the fuse is designed to operate. All low voltage fuses will operate at
any lower voltage (note that this is characterized as AC or DC, or both).
Low voltage fuses are classified according to the standard to which they are designed. The 7-1 table lists the
various fuse classes and pertinent data for each class.
Fuses, like most protective devices, exhibit inverse time-current characteristics. A typical fuse time-current
characteristic is shown in figure 7-6. Logarithmic scales are used for both the time and current axes, in order to
cover a wide range. The characteristic represents a band of operating times for which the lower boundary is the
minimum melting time curve, above which the fuses can be damaged. The upper boundary is the total clearing
time curve, above which the fuse will open. For a given fault current, the actual fuse opening time will be within
this band.
5
Table 7-1: Low Voltage fuse classes [3]

Fuse
Class
C
CA
CB
CC
Plug Fuses
Type C or
Type S
Voltage
Ratings
Ampere
Ratings
Interrupting
Ratings
(RMS)
Current
Limiting?
Standards
Notes
Varies
UL 248-3-2000,
CSA C22.2 NO.
248.2-2000
Plug-style
UL 248-3-2000,
CSA C22.2 NO.
248-3-2000
No mounting holes
UL 248-3-2000,
CSA C22.2 NO.
248-3-2000
Mounting holes in
end blades
600 Vac
0-12000 A
200,000 A
0-600 Vdc
Varies
Varies
600 Vac
0-30A
200,000A
Yes
0-600 Vdc
Varies
Varies
Yes
600 Vac
0-60 A
200,000 A
Yes
0-600 Vdc
Varies
Varies
Yes
600 Vac
0-30 A
200,000 A
Yes
Rejection-style;
0-600 Vdc
Varies
Varies
UL 248-4-2000 ,
CSA C22.2 NO.
248.4-2000
480 Vac
25-60 A
100,000 A
Yes
UL 248-5-2000,
CSA C22.2 NO.
248.5-2000
Non-inter-changeable dimen-sions with

other fuse classes
6000 V
0-20 A
100,000 A
Yes
480 Vdc
Varies
Varies
Yes
250 Vac
0-600 A
10,000 A
No
No
600 Vac
0-600 A
10,000 A
0-600 Vdc
Varies
Varies
600 Vac
0-600 A
200,000 A
Yes
0-600 Vdc
Varies
Varies
Yes
250 Vac
0-600A
50,000 A
Yes*
250 Vac
0-600A
100,000 A
Yes*
250 Vac
0-600A
200,000 A
Yes*
600 Vac
0-600A
50,000 A
Yes*
600 Vac
0-600A
100,000 A
Yes
600 Vac
0-600A
200,000 A
Yes*
0-600 Vdc
Varies
Varies
600 Vac
601-6000 A
200,000 A
0-600 Vdc
Varies
Varies
250 Vac
0-600 A
600 Vac
UL 248-6-2000,
CSA C22.2 NO.
248.6-2000
UL 248-8-2000,
CSA C22.2 NO.
248.8-2000
UL 248-9-2000,
CSA C22.2 NO.
248-9-2000
Divided into low

(K-1), medium (K-5),
and high (K-9) Ip and
I2t sub-classes;
Dimen-sions
interchange-able
with class H fuses
Yes
UL 248-10-2000,
CSA C22.2 NO.
248.10-2000
Bolt-on construction
200,000 A
Yes
0-600 A
200,000 A
Yes
UL 248-12-2000,
CSA C22.2 NO.
248.12-2000
0-600 Vdc
Varies
Varies
Divided into medium

(RK-1) and high
(RK-5) Ip and I2t subclasses; Will fit class
H or Class K fuse
holders, but Class R
fuse holders will not
fit any other type;
300 Vac
0-1200 A
200,000 A
Yes
600 Vdc
0-1200 A
200,000 A
Yes
UL 248-15-2000,
CSA C22.2 NO
248.15-2000
Similar to Class J,
but dimension-ally
smaller
0-600 Vdc
Varies
Varies
125 Vac
0-30 A
10,000 A
125 Vdc
0-30 A
10,000 A
UL 248-11-2000,
CSA NO. 248.112000
Type S has rejection

features
No
* Because of their interchangeability with Class H fuses, class K-1, K-5, and K-9 fuses cannot be marked as current-limiting.
100K
10K
1K
100
C UR R E NT IN AMP E R E S
1000
100
100
10
10
100K
0.01
10K
0.01
1K
0.10
100
0.10
T IME IN S E C ONDS
1000
Figure 7-6: Typical class J fuse time-current characteristic
In some cases the fuse average melting time only is given. This can be treated as the fuse opening time
with a tolerance of 15%. The -15% boundary is the minimum melting time and the +15% boundary is the
total clearing time.
Note that the time-current characteristic does not extend below .01 seconds. This is due to the fact that below
.01 seconds the fuse is operating in its current-limiting region and the fuse I2t is of increasing importance.
The time-current characteristic curves are used to demonstrate the coordination between protective devices in
series. The basic principle of system protection is that for a given fault current ideally only the device nearest the
fault opens, minimizing the effect of the fault on the rest of the system. This principle is known as selective
coordination and can be analyzed with the use of the device time-current characteristic curves.
As an example, consider a 480 V system with two sets of fuses in series, with a system available fault current of
30,000 A. Bus A is protected using 400 A class J fuses which supply, among others, bus B. Bus B is protected
using 100 A class J fuses. Coordination between the 400 A and 100 A fuses can is shown via the time-current
curves of figure 7-7, along with a one-line diagram of the part of the system under consideration. Because the
time bands for the two fuses do not overlap, these are coordinated for all operating times above .01 seconds.
It can also be stated that these two sets of fuses are coordinated through approximately 4200 A, since at 4200 A
Fuse A has the potential to begin operating in its current-limiting region. Fuse B has the potential to begin
operating its current-limiting region at 1100 A. For currents above approximately 4000 A, therefore, both sets of
fuses have the potential to be operating in the current-limiting region. When both sets of fuses are operating the
current-limiting region the time-current curves cannot be used to the determine coordination between them.
Instead, for a given fault current the minimum melting I2t for Fuse A must be greater than the maximum clearing I2t
for Fuse B. In practice, instead of publishing I2t data fuse manufacturers typically publish ratio tables showing the
minimum ratios of fuses of a given type that will coordinate with each other.
100K
10K
1K
100
10
1000
1000
FUSE A
100
100
FUSE B
10
10
BUS A
FUSE A
400 .0 A
T IME IN S E C ONDS
UTILITY SOURCE
480 V
300 00.00A Available Fau lt
BUS B
FUSE B
100 .0 A
0.10
0.10
BUS C
100K
1K
10K
0.01
10
100
0.01
Figure 7-7: Fuse coordination example
Low voltage fuse AC interrupting ratings are based upon a maximum power factor of .2, corresponding to a
maximum X/R ratio of 4.899. In order to evaluate a low voltage fuses interrupting rating on a system with a higher
X/R ratio the system symmetrical fault current must be multiplied by a multiplying factor [3]:
MULT =
1+ e
X

R actual
X

1+ e R
(7-10)
test
where
X

R actual
is the actual system X/R
X

R test
is the test X/R
The available symmetrical fault current multiplied by the multiplying factor per (7-10) can be compared to the fuse
interrupting rating.
The use of fuses requires a holder and a switching device in addition to the fuses themselves. Because they are
single-phase devices, a single blown fuse from a three-phase set will cause a single-phasing condition, which can
lead to motor damage. Replacing fuses typically requires opening equipment doors and/or removing cover panels.
Also, replacement fuses must be stocked to get a circuit with a blown fuse back on-line quickly. For these
reasons, the use of low voltage fuses in modern power systems is generally discouraged. For circuit breakers that
have a short-time rating
Low voltage molded-case circuit breakers

The molded-case circuit breaker is the workhorse for system protection 600V and below. A circuit breaker is a
device designed to open and close by nonautomatic means and to open the circuit automatically on a
predetermined overcurrent without damage to itself when properly applied within its rating [1].
The following terms apply to molded-case circuit breakers [3], [4]:
Voltage: Circuit breakers are designed and marked with the maximum voltage at which they can be applied.
Circuit breaker voltage ratings distinguish between delta-connected, 3-wire systems and wye-connected, 4-wire
systems. As stated in NEC article 240.85 [1], a circuit breaker with a straight voltage rating, such as 240 or 480 V
can be used in a circuit in which the nominal voltage between any two conductors does not exceed the circuit
breakers voltage rating. Breakers with slash ratings, such as 120/240 V or 480 Y/277 V, can be applied in a
solidly-grounded circuit where the nominal voltage of any conductor to ground does not exceed the lower of the
two values of the circuit breakers voltage rating and the nominal voltage between any two conductors does not
exceed the higher value of the circuit breakers voltage rating.
Frequency: Molded-case circuit breakers are normally suitable for 50Hz or 60Hz. Some have DC ratings as well.
Continuous current or Rated current: This is the maximum current a circuit breaker can carry continuously
at a given ambient temperature rating without tripping (typically 40C). In accordance with NEC [1] article 210.20
a circuit breaker (or any branch circuit overcurrent device) should not be loaded to over 80% of its continuous
current unless the assembly, including the circuit breaker and enclosure, is listed for operation at 100% of
its rating.
Poles: The number of poles is the number of ganged circuit breaker elements in a single housing. Circuit
breakers are available with one, two, or three poles, and also four poles for certain applications. Per NEC [1]
article 240.85 a two-pole circuit breaker cannot be used for protecting a 3-phase, corner-grounded delta circuit
unless the circuit breaker is marked 1 - 3 to indicate such suitability.
Control voltage: The control voltage rating is the AC or DC voltage designated to be applied to control
devices intended to open or close a circuit breaker. In most cases this only applies to accessories that are
custom-ordered, such as motor operators.
Interrupting rating: This is the highest current at rated voltage that the circuit breaker is intended to interrupt
under standard test conditions.
Short-time or Withstand Rating: This characterizes the circuit-breakers ability to withstand the effects of
short-circuit current flow for a stated period. Molded-case circuit breakers typically do not have a withstand rating,
although some newer-design breakers do.
Instantaneous override: A function of an electronic trip circuit breaker that causes the instantaneous function to
operate above a given level of current if the instantaneous function characteristic has been disabled.
Current Limiting Circuit Breaker: This is a circuit breaker which does not employ a fusible element and, when
operating in its current-limiting range, limits the let-through I2t to a value less than the I2t of a _-cycle wave of the
symmetrical prospective current.
HID: This is a marking that indicates that a circuit breaker has passed additional endurance and temperature rise
tests to assess its ability to be used as the regular switching device for high intensity discharge lighting. Per NEC
240.80 (D) a circuit breaker which is used as a switch in an HID lighting circuit must be marked as HID. HID
circuit breakers can also be used as switches in fluorescent lighting circuits.
SWD: This is a marking that indicates that a circuit breaker has passed additional endurance and temperature
rise tests to assess its ability to be used as the regular switching device fluorescent lighting. Per NEC 240.80 (D)
a circuit breaker which is used as a switch in an HID lighting circuit must be marked as SWD or HID.
Frame: The term Frame is applied to a group of circuit breakers of similar configuration. Frame size is expressed
in amperes and corresponds to the largest ampere rating available in that group.
Thermal-magnetic circuit breaker: This type of circuit breaker contains a thermal element to trip the circuit
breaker for overloads and a faster magnetic instantaneous element to trip the circuit breaker for short circuits.
On many larger thermal-magnetic circuit breakers the instantaneous element is adjustable.
Electronic trip circuit breaker: An electronic circuit breaker contains a solid-state adjustable trip unit. These
circuit breakers are extremely flexible in coordination with other devices.
Sensor: An electronic-trip circuit breakers sensor is usually an air-core current transformer (CT) designed
specifically to work with that circuit breakers trip unit. The sensor size, in conjunction with the rating plug,
determines the electronic-trip circuit breakers continuous current rating.
Rating plug: An electronic trip circuit breakers rating plug can vary the circuit breakers continuous current rating
as a function of its sensor size.
Typical molded-case circuit breakers are shown in figure 7-8. In figure 7-8 on the left is a thermal-magnetic circuit
breaker, and on the right is an electronic-trip circuit breaker. The thermal-magnetic circuit breaker is designed for
cable connections and the electronic circuit breaker is designed for bus connections, but neither type is inherently
suited for one connection type over another. Note the prominently-mounted operating handle on each circuit
breaker.
Circuit breakers may be mounted in stand-alone enclosures, in switchboards, or in panelboards (more information
on switchboards and panelboards is given in a later section of this guide).
Figure 7-8: Molded-Case circuit breakers
A typical thermal-magnetic circuit breaker time-current characteristic is shown in figure 7-9. Note the two distinct
parts of the characteristic curve: The thermal or long-time characteristic is used for overload protection and the
magnetic or instantaneous characteristic is used for short-circuit protection. Note also that there is a band of
operating times for a given fault current. The lower boundary represents the lowest possible trip time and the
upper boundary represents the highest possible trip time for a given current.
10
100K
10K
1K
100
1000
1000
Thermal (Long-Time) Operating Region
100
100
10
10
0.01
10K
100K
0.01
1K
0.10
100
0.10
T IME IN S E C ONDS
Magnetic (Instantaneous) Operating Region

1
Figure 7-9: Thermal magnetic circuit breaker time-current characteristic
10K
100K
1K
100
The time-current characteristic for an electronic-trip circuit breaker is shown in figure 7-10. The characteristic for
an electronic trip circuit breaker consists of the long time pickup, long-time delay, short-time pickup, short time
delay, and instantaneous pickup parameters, all of which are adjustable over a given range. This adjustability
makes the electronic-trip circuit breaker very flexible when coordinating with other devices. The adjustable
parameters for an electronic trip circuit breaker are features of the trip unit. In many cases the trip unit is also
available without the short-time function. In catalog data the long-time characteristic is listed as L, the short-time is
listed as S, and the instantaneous as I. Therefore an LSI trip unit has long-time, short-time, and instantaneous
characteristics, whereas an LI trip unit has only the long-time and instantaneous characteristics. For circuit
breakers that have a short-time rating, the instantaneous feature may be disabled, enhancing coordination with
downstream devices.
1000
1000
Long-Time Operating Region

Long Time Delay
100
100
Short Time Pickup

10
10
T IME IN S E C ONDS
Short Time Delay
Instantaneous Pickup
100K
0.01
10K
0.01
1K
0.10
100
0.10
Figure 7-10: Electronic-trip circuit breaker time-current characteristic
11
If the instantaneous feature has been disabled one must still be cognizant of any instantaneous override feature
the breaker has, which will engage the instantaneous function above a given level of current even if it has been
disabled in order to protect the circuit breaker from damage.
Another feature available on electronic-trip circuit breakers is ground-fault protection, which is discussed in detail
later in this section.
100K
10K
10
1000
1K
100
Typical coordination between an electronic and a thermal magnetic circuit breaker is shown in figure 7-11.
Because the time bands do not overlap, these two devices are considered to be coordinated.
1000
CB A
100
100
CB B
10
10
BUS A
CB A
250 0.0 A
T IME IN S E C ONDS
UTILITY SOURCE
480 V
300 00.00A Available Fau lt
BUS B
CB B
400 .0 A
0.10
0.10
BUS C
100K
10K
1K
0.01
10
100
0.01
Figure 7-11: Typical molded-case circuit breaker coordination
A further reduction in the let-through energy for a fault in the region between two electronic-trip circuit breakers
can be accomplished through zone-selective interlocking. This consists of wiring the two trip units such that if the
downstream circuit breaker senses the fault (typically this will be based upon the short-time pickup) it sends a
restraining signal to the upstream circuit breaker. The upstream circuit breaker will then continue to time out as
specified on its characteristic curve, tripping if the downstream device does not clear the fault. However, if the
downstream device does not sense the fault and the upstream devices does, the upstream device will not have
the restraining signal from the downstream device and will trip with no intentional delay. For example, if zone
selective interlocking were present in the system of figure 7-11 and fault occurs on bus C circuit breaker B will
sense the fault and send a restraining signal to circuit breaker A. Circuit breaker A is coordinated with circuit
breaker B, so circuit breaker B will trip first. If circuit breaker B fails to clear the fault, circuit breaker A will time out
on its time-current characteristic per figure 7-11 and trip. If the fault occurs at bus B, circuit breaker B will not
detect the fault and thus will not send the restraining signal to circuit breaker A. Circuit breaker A will sense the
fault and will trip with no intentional delay, which is faster than dictated by its time-current characteristic per figure
7-11. Care must be used when applying zone-selective interlocking where there are multiple sources of power and
fault currents can flow in either direction through a circuit breaker.
Table 7-2 shows typical characteristics of molded-case circuit breakers [3]. This table is for reference only; when
specifying circuit breakers manufacturers actual catalog data should be used.
12
Table 7-2: Typical characteristics of molded case circuit breakers for commercial and
industrial applications (Largely same as [3] table 7-1)
Frame Size (A)
Number of
Poles
Interrupting Rating at AC voltage (kA, RMS symmetrical)

120 V
100
100, 150
225, 250
400, 600
800, 1000
1200
1600, 2000
3000, 4000
240 V
277 V
10
14
65
65
480 V
600 V
2,3
18
14
14
2,3
65
25
18
2,3
100
65
25
2,3
25
22
22
2,3
65
25
22
2,3
100
65
25
2,3
42
30
22
2,3
65
65
25
2,3
100
42
30
22
65
50
25
200
100
65
42
30
22
65
50
25
200
100
65
65
50
42
125
100
65
100
100
85
200
150
100
35
Note that the continuous current rating is set by the sensor and rating plug sizes for a given electronic-trip circuit
breaker. This can be smaller than the frame size. As can be seen from table 7-2, more than one interrupting rating
can be available for a given frame size.
Molded-case circuit breakers are tested for interrupting capabilities with test X/R ratios as shown in table 7-3 [4].
As with fuses, when a circuit breaker is applied in a circuit with an X/R ratio larger than its test X/R then the
available RMS symmetrical fault current should be multiplied by the multiplying factor per equation (7-10) in order
to be compared with the circuit breaker interrupting rating.
Table 7-3: AC test circuit characteristics for molded-case circuit breakers [4]
Interrupting rating
(RMS Symmetrical)
Test circuit power factor
(X/R)test
10,000 or LESS
0.45 - 0.50
1.732
10,001-20,000
0.25-0.30
3.180
Over 20,000
0.15 - 0.20
4.899
Current-limiting circuit breakers are also available. Coordination between two current-limiting circuit breakers when
they are both operating in the current limiting range is typically determined by test.
13
By definition, low voltage molded case circuit breakers are not maintainable devices. Failure of a component
generally requires replacement of the entire circuit breaker unless the circuit breaker has been specifically
designed for maintainability.
Magnetic-only circuit breakers which have only magnetic tripping capability are available. These are often used
as short-circuit protection for motor circuits (discussed in more detail in a later section of this guide). For this
reason these are often referred to as motor circuit protectors.
Molded case switches are also available. These do not have a thermal element, however most have a magnetic
element which opens the switch above a specified current to protect the switch from damage due to lack of a
short-time rating.
Molded-case circuit breakers are available with several different options, such as stored-energy mechanisms, key
interlocks, motor operators, etc. Refer to specific manufacturers literature for details.
Because the switching means is included with the device, molded-case circuit breakers give inherent flexibility of
operation. This allows circuits to be reclosed without removing cover panels and exposing the operator to
hazardous voltages. For three-phase circuits three-pole circuit breakers are used, which alleviates single-phasing
concerns. And, circuit breakers are not one-time devices, eliminating need to store spares in the event of a fault.
These characteristics make molded-case circuit breakers very versatile protective devices.
Low voltage power circuit breakers

For larger systems, those devices closest to the source of power often require the ability to coordinate with
multiple levels of coordinating devices. In the case of circuit breakers, this generally requires a short-time
rating as described in Low voltage molded-case circuit breakers section above. In addition, in this type of
application maintainability is desired due to the cost of a single circuit breaker. Low voltage power circuit
breakers fill these needs.
AC low voltage power circuit breakers are designed and manufactured per ANSI/IEEE Standard C37.13-1990
and UL 1066-1997. These are generally electronic-trip circuit breakers, although in existing installations older
dashpot-operated units may be encountered. The tripping characteristics are essentially identical to those for
electronic-trip molded-case circuit breakers, per figure 7-9, except that the instantaneous function may be
disabled in all cases, unlike that of a molded-case circuit breaker. Tables 7-4 and 7-5 give the preferred ratings
for low voltage AC power circuit breakers [3]. In addition, fused power circuit breakers are also available with
higher interrupting ratings, although many modern-design power circuit breakers do not require fuses to obtain
short-circuit ratings up to 200kA RMS symmetrical.
The short-time rating is an important characteristic of the low voltage power circuit breaker. The rated time
duration of the short-time rating is _ s (two periods of _-s current separated by a 15s interval of zero current) [5].
Because of this short-time rating, low voltage power circuit breakers are also suitable for protective relaying
applications, as described below. Therefore, if a low voltage power circuit breaker is not equipped with a directacting trip unit it should not be subjected to more than _ s trip delay time at its short-time rating [5].
Figure 7-12: Low voltage power circuit breaker
14
Table 7-4: Preferred ratings for low voltage AC power circuit breakers with instantaneous
direct-acting phase trip elements (Same as [3] table 7-3)a
System
Nominal
Voltage (V)
Rated
Maximum
Voltage (V)
Insulation
(dielectric)
withstand (V)
Three-Phase ShortCircuit current

rating (A)b
Frame Size (A)
Range of trip
device current
ratings (A)c
600
635
2,200
14,000
225
40-225
600
635
2,200
22,000
600
40-600
600
635
2,200
22,000
800
100-800
600
635
2,200
42,000
1,600
200-1,600
600
635
2,200
42,000
2,000
200-2,000
600
635
2,200
65,000
3,000
2,000-3000
600
635
2,200
65,000
3,200
2,000-3200
600
635
2,200
85,000
4,000
4,000
480
508
2,200
22,000
225
40-225
480
508
2,200
30,000
600
100-600
480
508
2,200
30,000
800
100-800
480
508
2,200
50,000
1,600
400-1,600
480
508
2,200
50,000
2,000
400-2,000
480
508
2,200
65,000
3,000
2,000-3,000
480
508
2,200
65,000
3,200
2,000-3,200
480
508
2,200
85,000
4,000
4,000
240
254
2,200
25,000
225
40-225
240
254
2,200
42,000
600
150-600
240
254
2,200
42,000
800
150-800
240
254
2,200
65,000
1,600
600-1,600
240
254
2,200
65,000
2,000
600-2,000
240
254
2,200
85,000
3,000
2,000-3,000
240
254
2,200
85,000
3,200
2,000-3,200
240
254
2,200
130,000
4,000
4,000
See IEEE Std C37.13-1990 and ANSI C37.16-2000
Ratings in this column are RMS symmetrical values for single-phase (two pole) circuit breakers and three-phase average RMS
symmetrical values of three-phase (three-pole) circuit breakers. When applied on systems where rated maximum voltage may
appear across a single pole, the short-circuit current ratings are 87% of these values. See 5.6 in IEEE Std C37.13-1990.
The continuous-current-carrying capability of some circuit-breaker-trip-device combinations may be higher than the trip-device
current rating. See 10.1.3 in IEEE Std C37.13-1990.
15
Table 7-5: Preferred ratings for low voltage AC power circuit breakers without instantaneous
direct-acting phase trip elements (Largely same as [3] table 7-4)a
Range of trip device current ratings (A)d
Rated Maximum
Voltage (V)
Frame Size (A)
Setting of short-time delay trip element

Minimum time
band
Inter-mediate
Time Band
Maximum
Time Band
635
225
100-225
125-225
150-225
635
600
175-600
200-600
250-600
635
800
175-800
200-800
250-800
635
1,600
360-1,600
400-1,600
500-1,600
635
2,000
250-2,000
400-2,000
500-2,000
635
3,000
2,000-3,000
2,000-3,000
2,000-3,000
635
3,200
2,000-3,200
2,000-3,200
2,000-3,200
635
4,000
4,000
4,000
4,000
508
225
100-225
125-225
150-225
508
600
175-600
200-600
250-600
508
800
175-800
200-800
250-800
508
1,600
350-1600
400-1,600
500-1,600
508
2,000
350-2,000
400-2,000
500-2,000
508
3,000
2,000-3,000
2,000-3,000
2,000-3,000
508
3,200
4,000
4,000
2,000-3,200
508
4,000
4,000
4,000
4,000
254
225
100-225
125-225
150-225
254
600
175-600
200-600
250-600
254
800
175-800
200-800
250-800
254
1,600
350-1,600
400-1,600
500-1,600
254
2,000
350-2,000
400-2,000
500-2,000
254
3,000
2,000-3,000
2,000-3,000
2,000-3,000
254
3,200
2,000-3,200
2,000-3,200
2,000-3,200
254
4,000
4,000
4,000
4,000
See IEEE Std C37.13-1990 and ANSI C37.16-2000.
The continuous-current-carrying capability of some circuit-breaker-trip-device combinations may be higher than the tripdevice current rating. See 10.1.3 in IEEE Std C37.13-1990.
As with molded-case circuit breakers, low voltage power circuit breakers are tested at a given power factor. The
test power factor is 15% for unfused circuit breakers and 20% for fused circuit breakers. Table 7-6 shows the
multiplying factors for both fused and unfused circuit breakers for various short-circuit power factors. The
multiplying factors for unfused circuit breakers are calculated similarly to those for molded-case circuit breakers,
but those for fused circuit breakers are based upon RMS rather than peak current and differ slightly from the
multiplying factors obtained from equation (7-10) [5].
16
Table 7-6: Short-circuit multiplying factors for low voltage power circuit breakers
(Largely same as [5] table 3)
System Short-Circuit
Power Factor
System X/R Ratio
Multiplying Factor x
RMS Symmetrical ShortCircuit Current, for
Unfused Power Circuit
Breakers
Multiplying Factor x
RMS Symmetrical ShortCircuit Current, for
Fused Power Circuit
Breakers
20
4.9
1.00
1.00
15
6.6
1.00
1.07
12
8.27
1.04
1.12
10
9.95
1.07
1.15
8.5
11.72
1.09
1.18
14.25
1.11
1.21
20.0
1.14
1.26
Use of low voltage power circuit breakers allows optimum flexibility in coordination, since the instantaneous
function may be disabled. Further, since these are designed for heavy-duty use in an industrial environment
they are most often configured as drawout circuit breakers with stored-energy mechanisms in ANSI low voltage
metal enclosed switchgear (described in a later section of this guide). This makes them ideal for low voltage
automatic transfer applications. Their inherent operational flexibility serves to make them the ideal device
for circuit protection in industrial applications where the ability to coordinate with downstream devices is a
premium consideration.
Medium voltage fuses

The definition of fuse in Low voltage fuses section above is equally applicable to medium voltage fuses. Recall
from Table 4-1 that the medium voltage level is defined by ANSI C84 as containing standard system voltages
from 2400 through 69,000 V, and that the high voltage level contains standard system voltages from 115 kV
through 230 kV. The medium voltage level, strictly, is defined by ANSI C84 as greater than 1000 V and less
than 100,000 V. Similarly, the high voltage level is defined as greater than 100,000 V through 230,000 V.
Strictly-speaking, high voltage fuse standards are used for both medium and high voltage fuses. However the
focus of this section will be on medium voltage fuses through 38 kV.
The following standards apply to medium voltage fuses [3]:
I
IEEE Std. C37.40-2003
IEEE Std. C37.41-2000
ANSI C37.42-1996
ANSI C37.44-1981
ANSI C37.46-1981
ANSI C37.47-1981
IEEE Std. C37.48-1997
ANSI C37.53.1-1989
Those definitions in Low voltage fuses section above which do not specifically reference low voltage fuses are
also valid for medium voltage fuses. Generally, medium voltage fuses can be divided into two major categories:
Current-limiting and expulsion. Current-limiting fuses were defined in Low voltage fuses section above, and the
same basic definition applies to medium voltage fuses. Expulsion fuses are defined as follows [3]:
17
Expulsion fuse: A vented fuse in which the expulsion effect of the gases produced by internal arcing, either
alone or aided by other mechanisms, results in current interruption.
In addition, medium voltage fuses are further classified as power fuses or distribution fuses as follows [3]:
Power fuse: Defined by ANSI C37.42-1996 as having dielectric withstand (BIL) strengths at power levels, applied
primarily in stations and substations, with mechanical construction basically adapted to station and substation
mountings.
Distribution fuse: Defined by ANSI C37.42-1996 as having dielectric withstand (BIL) strengths at distribution
levels, applied primarily on distribution feeders and circuits, and with operating voltage limits corresponding to
distribution voltages. These are further subdivided into distribution current limiting fuses and distribution fuse
cutouts, as described below.
Current-limiting fuses interrupt in less than _ cycle when subjected to currents in their current-limiting range. This
is an advantage as it limits the peak fault current to a value less than the prospective fault current as described
above for low voltage fuses. This provides current-limiting fuses with high interrupting ratings and allows them to
protect downstream devices with lower short-circuit ratings in some cases. However, the same technologies that
combine to give medium voltage current-liming fuses their current-limiting characteristics can also produce thermal
issues when the fuses are loaded at lower current levels. For this reason, the following definitions apply to
current-limiting fuses [3]
Backup current-limiting fuse: A fuse capable of interrupting all currents from its maximum rated interrupting
current down to its rated minimum interrupting current.
General purpose current-limiting fuse: A fuse capable of interrupting all currents from the rated interrupting
current down to the current that causes melting of the fusible element in no less than 1h.
Full-range current-limiting fuse: A fuse capable of interrupting all currents from its rated interrupting current
down to the minimum continuous current that causes melting of the fusible elements.
Due to the limitations of backup and general purpose current limiting fuses, current-limiting power fuses have
melting characteristics defined as E or R, defined as follows:
E-Rating: The current-responsive element for ratings 100 A or below shall melt in 300 s at an RMS current within
the range of 200% to 240% of the continuous-current rating of the fuse unit, refill unit, or use link. The currentresponsive element for ratings above 100 A shall melt in 600 s at an RMS current within the range of 220% to
264% of the continuous-current rating of the fuse unit, refill unit, or fuse link.
R-Rating: The fuse shall melt in the range of 15 s to 35 s at a value of current equal to 100 times the R number.
Similarly, distribution current-limiting fuses are defined by given characteristic ratings, one of which is the C rating,
defined as follows:
C-Rating: The current-responsive element shall melt at 100 s at an RMS current within the range of 170% to
240% of the continuous-current rating of the fuse unit.
A typical time-current curve for an E-rated current-limiting power fuse is shown in figure 7-13. The fuse in figure
7-13 is a 125E-rated fuse. Note that the curve starts at approximately 250 A for a minimum melting time of 1000 s.
Care must be taken with backup and general-purpose current-limiting fuses so that the load current does not to
exceed the E- or R-rating of the fuse. Failure to do this can result in the development of a hot-spot and
subsequent failure of the fuse and its mounting. For fuses enclosed in equipment, this can have disastrous
consequences since failure of the fuse and/or its mounting can lead to an arcing fault in the equipment. Note that
the boundary of the characteristic, denoting the minimum-melting current, should be further derated to take into
account pre-loading of the fuse (consult the fuse manufacturer for details). Note that, as with low voltage fuses,
the current-limiting fuse characteristic does not extend below .01 seconds since the fuse would be in its currentlimiting range below this interrupting time.
18
100K
10K
1K
100
10
1000
1000
100
10
10
0.10
T IME IN S E C ONDS
100
0.10
100K
10K
1K
0.01
10
100
0.01
Figure 7-13: Typical E-rated current-limiting power fuse time-current characteristic.
A current-limiting power fuse consists of a fuse mounting (typically fuse clips) plus the fuse unit itself. These are
frequently mounted in metal-enclosed switchgear. A distribution current-limiting fuse may consist of a
disconnecting-style holder or clips, plus the fuse unit. Distribution current-limiting fuses may also be provided with
under-oil mountings for use with distribution transformers. They are frequently used for capacitor protection as
well, with clips designed to mount to the capacitor.
Figure 7-14: Current-limiting power fuses and mountings
Current-limiting power fuses are typically used for short-circuit protection of instrument transformers, power
transformers, and capacitor banks. Table 7-7 gives maximum ratings for medium voltage current-limiting power
fuses from 2.75 through 38 kV.
19
Table 7-7: Maximum ratings for current-limiting power fuses 2.75 - 38 kV

(Same as [3] table 6-3)
Rated Maximum Voltage (kV)
Continuous-Current Ratings (A),

Maximum
Short-Circuit maximum
interrupting ratings
(kA RMS symmetrical)
2.75
225,450a,750a, 1350a
50.0, 50,0, 40.0, 40.0
2.75/4.76
450a
50.0
5.5
225,400,750a,1350a
50.0, 62.5, 40.0, 40.0
8.25
125,200a
50.0, 50.0
15.5
65,100,125a,200a
85.0, 50.0, 85.0, 50.0
25.8
50,100a
35.0, 35.0
38.0
50,100a
35.0, 35.0
Parallel Fuses
During interruption current-limiting fuses produce significant arc voltages. These must be taken into account in
selecting equipment. They are typically compared to the BIL level of the equipment, including downstream
equipment at the same voltage level. The maximum permissible overvoltages for current-limiting power fuses are
shown in table 7-8 [3]:
Table 7-8: Maximum permissible overvoltages for current-limiting power fuses
Maximum Peak Overvoltages (kV, crest)
0.5A to 12A
Over 12A
2.8
13
5.5
25
18
8.3
38
26
15.0
68
47
15.5
70
49
22.0
117
70
25.8
117
81
27.0
123
84
38.0
173
119
In practice, the arc voltages for current-limiting fuses generally indicate the use of the smallest available fuse
voltage class for the given system voltage, for example, 5.5 kV fuses instead of 8.3kV fuses for a 4160 V system.
After a fault interruption, in a three-phase set of current-limiting fuses all three fuses will be replaced, even if
only one fuse interrupted the fault. This is due to the possibility of damage to the other two fuses due to the
fault, which could change their time-current characteristics and make them unsuitable to carry load current
without failure.
Because medium voltage current-limiting fuses interrupt short circuits without the expulsion of gas or flame,
they are widely utilized in a variety of applications.
20
Power expulsion fuses generally consist of an insulating mounting plus a fuse holder which accepts the fuse
refills. The fuse holder may be of the disconnecting or non-disconnecting type. Only the refill is replaced when a
fuse interrupts an overcurrent, and if only one phase of a three-phase set interrupted the fault only that fuse need
be replaced. Power expulsion fuses are typically used in substations and enclosed equipment.
Distribution expulsion fuses are generally distribution fuse cutouts, which are adapted to pole or cross arm
mounting. They consist of the fuse holder and refill unit. The fuse holder is usually of the disconnecting type.
These are typically used as pole-mounted fuses on utility distribution systems.
Expulsion fuses use the liberation of de-ionizing gasses to interrupt overcurrents. Boric acid is typically used as
the interrupting medium for power expulsion fuses and bone fiber is typically used for distribution fuse cutouts.
When an expulsion fuse interrupts an overcurrent the interrupting medium liberates de-ionizing gas, interrupting
the overcurrent. The exhaust gasses are then emitted from the fuse, accompanied by noise. The exhaust
gasses for a boric acid fuse may condensed by an exhaust control device (commonly called an exhaust filter,
silencer, or snuffler).
Unlike current-limiting fuses, expulsion-type fuses interrupt high overcurrents at natural current zeros. They are
therefore non-current-limiting, and as a result typically have lower interrupting ratings than current-limiting fuses.
Table 7-9 shows the maximum continuous current and short-circuit interrupting ratings for refill-type boric-acid
expulsion-type power fuses [3]. Because expulsion-type fuses are non-current-limiting, they do not produce the
significant arc voltages that current-limiting fuses produce, and thus it is permissible to use a fuse with a larger
voltage class than the system, for example, a 14.4 kV-rated fuse on a 4160 V system. This makes expulsion-type
fuses particularly useful on systems which may be upgraded in the future to a higher voltage. However, the lower
interrupting ratings of expulsion-type fuses are often an issue vs. current-limiting fuses in light of the fact that the
largest expulsion-type fuse interrupting ratings require larger physical dimensions which cannot always be easily
accommodated in enclosed equipment. Further, in some cases the expulsion-type fuses prohibit some spacesaving mounting configurations in enclosed equipment that are available with current-limiting fuses.
Table 7-9: Maximum continuous current and short circuit interrupting ratings for refill type
boric-acid expulsion-type power fuses (Same as [3] table 6-6)
Continuous-Current Ratings (A),

maximum
Short-Circuit maximum
interrupting ratings
(kA, RMS symmetrical)
2.8
200,400,720a
19.0, 37.5, 37.5
4.8
200,400,720a
19.0, 37.5, 37.5
5.5
200,400,720a
19.0, 37.5, 37.5
8.3
200,400,720a
16.6, 29.4, 29.4
14.4
200,400,720a
14.4, 29.4, 29.4
15.5
200,400,720a
14.4, 34.0, 29.4
17.0
200,400,720a
14.4, 34.0, 25.0
25.8
200,300,540a
10.5, 21.0, 21.0
27.0
200,300
12.5, 20.0
38.0
200,300,540a
8.45, 17.5, 16.8
Parallel Fuses
21
100K
10K
1K
10
1000
100
E-ratings are used for power expulsion fuses. A typical time-current characteristic for a 125E boric-acid fuse is
given in figure 7-15.
1000
100
10
10
0.10
T IME IN S E C ONDS
100
0.10
100K
10K
1K
0.01
10
100
0.01
Figure 7-15: Typical boric acid power expulsion fuse time-current characteristic
Note that the characteristic extends to the available fault current (in this case, 29.4 kA), unlike that of the
current-limiting fuse. It is common practice to treat these as current-limiting fuses so far as the E-rating is
concerned, i.e., the maximum load current is usually kept below the E-rating. However, the boric-acid fuse is
not subject to damage when loaded above its E-rating, and they are often referred to in the industry as
non-damageable due to this fact.
When applying medium voltage fuses, the voltage rating and the interrupting rating are of importance.
The maximum line-to-line voltage of the system should not exceed the fuse voltage rating. The published
interrupting rating for power fuses is typically for a test X/R ratio of 15, and for distribution fuses the test X/R ratio
is typically 8; the fuse manufacturer should be consulted for derating factors for X/R ratios above these values.
The manufacturer should also be consulted if the test X/R is in doubt.
Medium voltage fuses provide economical short-circuit protection when applied within their ratings, particularly for
transformers, cables, and capacitors. For more sophisticated protection at the medium voltage level, other means
must be employed.
Medium voltage circuit breakers

The medium voltage circuit breaker is the device of choice when sophisticated system protection at the
medium voltage level is required.
Most modern medium voltage circuit breakers use a vacuum as the interrupting means, although older
sulfur-hexafluoride (SF6)based units still exist. As with medium voltage fuses, the same standards are used for
both medium and high voltage circuit breakers. The applicable standards are ANSI/IEEE C37.04-1999,
ANSI/IEEE C37.06-2000, and ANSI/IEEE C37.09 1999. In addition, ANSI/IEEE C37.010-1999 and
ANSI/IEEE C37.011-1994 give valuable application advise for these devices.
22
Medium voltage circuit breakers are generally not equipped with integral trip units as low voltage circuit breakers
are. Instead, protective relays must be used to sense abnormal conditions and trip the circuit breaker accordingly.
Most modern medium voltage circuit breakers are rated on a symmetrical current basis. The following rating
definitions apply [6]:
Rated Maximum Voltage: The highest RMS phase-to-phase voltage for which the circuit breaker is designed.
Rated Power Frequency: The frequency at which the circuit breaker is designed to operate.
Figure 7-16: Medium voltage circuit breaker, for use In metal-clad switchgear
Rated Dry Withstand Voltage: The RMS voltage that the circuit breaker in new condition is capable of
withstanding for 1 minute under specified conditions.
Rated Wet Withstand Voltage: The RMS voltage that an outdoor circuit breaker or external components in new
condition are capable of withstanding for 10s.
Rated Lightning Impulse Withstand Voltage: The peak value of a standard 1.2 x 50 s wave, as defined in
IEEE Std 4-1978, that a circuit breaker in new condition is capable of withstanding.
Rated Continuous Current: The current in RMS symmetrical amperes that the circuit breaker is designed to
carry continuously.
Rated Interrupting Time: The maximum permissible interval between the energizing of the trip circuit at rated
control voltage and the interruption of the current in the main circuit in all poles.
Rated Short Circuit Current (Required Symmetrical Interrupting Capability): The value of the symmetrical
component of the short-circuit current in RMS amperes at the instant of arcing contact separation that the circuit
breaker shall be required to interrupt at a specified operating voltage, on the standard operating duty cycle, and
with a DC component of less than 20% of the current value of the symmetrical component.
Required Asymmetrical Interrupting Capability: The value of the total RMS short-circuit current at the instant
of arcing contact separation that the circuit breaker shall be required to interrupt at a specified operating voltage
and on the standard operating duty cycle. This is based upon a standard time constant of 45ms (X/R ratio =17 for
60 Hz and 14 for 50 Hz systems) and an assumed relay operating time of _ cycle.
Rated closing and latching capability: The circuit breaker shall be capable of closing and latching any power
frequency making current whose maximum peak is equal to or less than 2.6 (for 60 Hz power frequency; 2.5 for
50 Hz power frequency) times the rated short-circuit current.
23
Rated Short-Time Current: The maximum short-circuit current that the circuit breaker can carry without tripping
for a specified period of time.
Maximum Permissible Tripping Delay: The maximum delay time for protective relaying to trip the circuit
breaker during short-circuit conditions, based upon the rated short-time current and short-time current-carrying
time period.
Rated Transient Recovery Voltage (TRV): At its rated maximum voltage, a circuit breaker is capable of
interrupting three-phase grounded and ungrounded terminal faults at the rated short-circuit current in any circuit in
which the TRV does not exceed the rated TRV envelope. For a circuit breaker rated below 100kV, the rated TRV
is represented by a 1-cosine wave, with a magnitude and time-to-peak dependent upon the rated maximum
voltage of the circuit breaker.
Rated Voltage Range Factor K: Defined in earlier versions of [6] as the factor by which the rated maximum
voltage may be divided to determine the minimum voltage for which the interrupting rating varies linearly with the
interrupting rating at the rated maximum voltage by the following formula:
V
I vop = I v max max
Vop
(7-8)
where
Ivmax
Vmax
Vop
Ivop
is the rated short-circuit current at the maximum operating voltage

is the rated maximum operating voltage
V
is the operating voltage where Vop Kmax
is the short-circuit current interrupting capability where Ivop Iv max x K.
For values of Vop below (Vmax K) the short-circuit interrupting capability was considered to be equal to
(Iv max x K). This model was more representative of older technologies such as air-blast interruption. Because
most modern circuit breakers employ vacuum technology, the current version of [6] assumes that K = 1., which
gives the same short circuit rating for all voltages below the rated voltage. However, in practice designs with K > 1
still exist and are in common use.
Table 7-10 shows the preferred ratings for circuit breakers from [7] where K=1. Table 7-11 shows the preferred
ratings for circuit breakers where K > 1.
24
Table 7-10: Preferred ratings for indoor circuit breakers with K=1.0
(Essentially same as [7] table 1)
Rated
Max.
Voltage,
(kV)
Rated
Voltage
Range
Factor K
Rated
Continuous
Current
(A RMS)
Rated
Rated TRV
ShortRated Peak Rated
Circuit and
Voltage E2 Time to
Short-Time
(kV peak) Peak T2,
Current
(s)
(kA RMS)
Rated
Interrupting
Time
(ms)
Rated Max.
Permissible
Tripping
Time Delay Y
(s)
4.76
1.0
1200, 2000
31.5
8.9
4.76
1.0
1200, 2000
40
8.9
4.76
1.0
1200, 2000, 3000
50
8.25
1.0
1200, 2000, 3000
15
1.0
1200, 2000
15
1.0
15
1.0
15
15
Rated
Closing
and
Latching
Current,
(kA Peak)
50
83
82
50
83
104
8.9
50
83
130
40
15.5
60
83
104
20
28
75
83
52
1200, 2000
25
28
75
83
65
1200, 2000
31.5
28
75
83
82
1.0
1200, 2000, 3000
40
28
75
83
104
1.0
1200, 2000, 3000
50
28
75
83
130
15
1.0
1200, 2000, 3000
63
28
75
83
164
27
1.0
1200
16
51
105
83
42
27
1.0
1200,2000
25
51
105
83
65
38
1.0
1200
16
71
125
83
42
38
1.0
1200,2000x
25
71
125
83
65
38
1.0
1200, 2000, 3000
31.5
71
125
83
82
38
1.0
1200, 2000, 3000
40
71
125
83
104
It should be noted that although 83 ms or 5 cycles is the preferred value per [6] for the rated interrupting time,
3-cycle designs are common.
Other related preferred ratings, such as dielectric ratings and capacitance switching ratings, are also given in [7].
Table 7-11: Preferred ratings for indoor circuit breakers with voltage range factor K > 1.0
(Essentially same as [7] table A1)
Rated
Max.
Voltage,
(kV)
Rated
Voltage
Range
Factor K
Rated
Continuous
Current at
60Hz
(A RMS)
Rated ShortRated
Rated Max.
Max.
Closing and
Circuit
Interrupting
Voltage
Symmetrical
Latching
Current at
Time,
Divided by Interrupting
Capability
Rated Max.
Cycles
K, kV RMS Capability and 2.7K Times
kV
Rated Short- Rated Short(kA RMS)
Time Current
Circuit
(kA, RMS)
Current
(kA Crest)
4.76
1.36
1200
8.8
3.5
12
32
4.76
1.24
1200, 2000
29
3.85
36
97
4.76
1.19
1200, 2000, 3000
41
4.0
49
132
8.25
1.25
1200, 2000
33
6.6
41
111
15.0
1.30
1200, 2000
18
11.5
23
62
15.0
1.30
1200, 2000
28
11.5
36
97
15.0
1.30
1200, 2000, 3000
37
11.5
48
130
38.0
1.65
1200, 2000, 3000
21
23.0
35
95
38.0
1.0
1200, 3000
40
38.0
40
108
In order to apply medium voltage circuit breakers, it is important to understand how the system X/R ratio affects
the circuit breaker interrupting rating. As stated above, for 60Hz systems the asymmetrical interrupting capability is
based upon an X/R ratio of 17. Thus, for systems where the X/R ratio is 17 or lower the circuit breaker will have
adequate asymmetrical interrupting capability so long as 100% of the symmetrical short-circuit current rating is
25
equal to or above the available RMS symmetrical fault current. For X/R ratios above 17, the available RMS
symmetrical fault current must be compared to the short-circuit current rating of the circuit breaker multiplied by a
multiplying factor determined from [8]. Because the multiplying factors from [8] do not usually exceed 1.25, the
fault current may be compared to 80% of the circuit breaker interrupting rating regardless of X/R ratio in most
cases. The close and latch rating is evaluated using equation (7-9) to obtain the asymmetrical fault current at the
circuit breaker. Reference [8] contains a full method for determining the suitability of a circuit breaker for duty on a
given system, and along with the requirements for low voltage short-circuit calculations from [5] forms the basis for
what the industry terms as ANSI short-circuit analysis. Capacitance switching and generator applications are also
areas of concern when applying medium voltage circuit breakers. Preferred capacitance switching values are
given in [7] and must not be exceeded. Generator applications, for generators rated above 3MVA, must be
approached with caution due to the high X/R ratios encountered. Often, breakers with longer interrupting times are
desirable in large generator applications in order to allow the fault current to decay to the point that there is a
natural current zero for interruption.
As stated above, medium voltage circuit breakers are typically provided without integral trip units. For this reason,
custom protection must be provide via protective relays, discussed in the next section. Circuit breakers are
equipped with tripping and closing coils to allow tripping and closing operations via protective relays, manual
control switches, PLCs, SCADA systems, etc. The circuit breaker internal control circuitry is arranged per IEEE
C37.11-1997. Circuit breakers are also equipped with a number of auxiliary contacts to allow interlocking and
external indication of breaker position.
For medium voltage protection applications, circuit breakers offer flexibility that cannot be obtained with fuses.
Further, they do not require a separate switching device as fuses do. These benefits are gained at a price: Circuit
breaker applications are more expensive than fuse applications, both due to the inherent cost of the circuit
breakers themselves and due to the protective relays required. For many applications, however, circuit breakers
are the only choice that offers the flexibility required. Large medium voltage services and distribution systems and
most applications involving medium voltage generation employ circuit breakers.
Protective relays
For medium voltage circuit breaker applications, protective relays serve as the brains that detect abnormal
system conditions and direct the circuit breakers to operate. They also serve to provide specialized protection in
low voltage power circuit breaker applications for functions not available in the circuit breaker trip units.
Most modern protective relays are solid-state electronic or microprocessor-based devices, although older
electromechanical devices are still available. Solid-state electronic or microprocessor-based relays offer more
flexibility and functionality than electromechanical relays, including the ability to interface with common
communications protocols such as MODBUS for integration into a SCADA environment. However, they do require
reliable control power to maintain operation during abnormal system conditions. This reliable control power is
most often provided by a DC battery system, although AC UPS-based systems are also encountered.
Electromechanical relays are typically single-phase devices. Solid-state electronic relays are typically available in
single-phase or three-phase versions. Microprocessor-based relays are typically three-phase devices. While
electromechanical and solid-state electronic relays typically incorporate one relay function per device,
microprocessor-based relays usually encompass many functions in one device, making a single microprocessorbased relay capable of performing the same functions that would require several electromechanical or solid-state
relays. This functionality usually makes microprocessor-based relays the best choice for new installations.
Figure 7-17: Microprocessor-based protective relay
26
Protective relays are not rated for direct connection to the power system where they are applied. For this reason,
instrument transformers are used to reduce the currents and voltages to the levels for which the relays are
designed. Instrument transformers generally fall into one of two broad categories: Current Transformers (CTs)
and Voltage Transformers (VTs). The loads on instrument transformers, such as relays and meters, are known as
burdens to distinguish them from power system loads.
A current transformer consists of a coil toroidally-wound around a ferromagnetic core. The conductor for which the
current is to be measured is passed through the center of the toroid. The magnetic field generated by the current
through the conductor causes current to flow in the coil. In essence, a CT may be thought of as a conventional
transformer with one primary turn.
CTs in the United States typically have 5 A-rated secondaries, with primary ratings from 10 - 40,000 A and larger.
For relaying applications in industrial facilities, CT ratios are typically 50:5 - 4000:5. IEEE Std. C57.13-1993
designates certain ratios as standard, as well as a classification system for relaying performance. The
classification system consists of a letter and a number. The letter may be C, designating that the percent ratio
correction may be calculated, or T, denoting that the ratio correction has been determined by test. The number
denotes the voltage that the CT can deliver to a standard burden (as described in IEEE Std. C37.13-1993) at 20
times the rated secondary current without exceeding 10% ratio error. As a more accurate alternative,
manufacturer-published CT excitation curves may be used to determine the accuracy. For relaying application, the
issue at hand is the performance of the relay during worst-case short-circuit conditions, when the CT secondary
currents are the largest and may cause the secondary voltage to exceed the CTs rating due to the voltage
developed across the relay input coil. This condition will cause the CT to saturate, significantly changing the ratio
and thus the accuracy of the measurement. For cases of severe CT saturation the relay may respond in an
unpredictable manner, such as not operating or producing chatter of its output contacts.
CT's where the power conductor passes through the window formed by the toroidal CT winding are known as
window-type CTs. CTs which are designed with an integral bus bar running through device are known as bus-bar
type CTs. Other designs, such as wound primary CTs for metering applications and non-saturating air-core CTs,
are available. Additional information on CT application can be found in [3].
Quasi-Physical
Arrangement
Ip
Is
Circuit
Representation
Ip
Is
H1
X1
X2
H2
POWER
CONDUCTOR
Figure 7-18: Current transformer
Voltage transformers (VTs) are used to step the power system voltage down to a level that the relay can utilize.
The operation of voltage transformers is essentially the same as for conventional power transformers discussed in
section 2 of this guide, except that the design has been optimized for accuracy. Like current transformers, voltage
transformers are assigned accuracy classes by IEEE Std. C57.13-1996. VT accuracy classes are designated W,
X,M Y, Z, and ZZ in order of increasing burden requirements. Refer to [3] for more information regarding the
application of voltage transformers.
Protective relays are classified by function. To make circuit representations easier, each function has been defined
and assigned a number by IEEE Std. C37.2-1996. The IEEE standard function numbers are given in table 7-12.
Table 7-13 gives the commonly-used suffix letters to further designate protective functions [3].
These designations can be combined in various ways. For example, 87T denotes a transformer differential relay,
51N denotes a residual ground time-overcurrent relay, 87B denotes a bus differential relay, etc.
27
Table 7-12: Commonly used protective relay device function numbers (Same as [3] table 4-1)
Relay Device
Function Number
Protection Function
21
Distance
25
Synchronizing
27
Undervoltage
32
Directional Power
40
Loss of Excitation (field)
46
Phase balance (current balance, negative sequence current)
47
Phase-Sequence Voltage (reverse phase voltage)
49
Thermal (generally thermal overload)
50
Instantaneous Overcurrent
51
Time-overcurrent
59
Overvoltage
60
Voltage balance (between two circuits)
67
Directional Overcurrent
81
Frequency (over and underfrequency)
86
Lockout
87
Differential
Table 7-13: Commonly used suffix letters applied to relay function numbers
Suffix Letter
Relay Application
Alarm only
Bus protection
Ground fault protection [relay current transformer (CT) in a system neutral circuit] or
generator protection]
GS
Ground-fault protection (relay CT is toroidal or ground sensor)
Line Protection
Motor Protection
Ground fault protection (relay coil connected in residual CT circuit)
Transformer protection
Voltage
Several commonly-used protective functions are described below. It must be noted that where a protective
function is described it may be a dedicated relay (electromechanical, solid-state electronic, or microprocessorbased) or a single protective function contained within a microprocessor-based relay. In some manufacturers
literature the individual functions are referred to as elements.
A.) Overcurrent relays (Devices 50, 51)

Overcurrent relays are the most commonly-used protective relay type. Time-overcurrent relays are available with
various timing characteristics to coordinate with other protective devices and to protect specific equipment.
Instantaneous overcurrent relays have no inherent time delay and are used for fast short-circuit protection.
Figure 7-19 shows the timing characteristics of several typical 51 time-overcurrent relay curve types, along with
the 50 instantaneous characteristic.
28
100K
1K
10K
100
10
1000
1000
100
100
51, MOD. INV.
51, ST. INV.

51, INVERSE
10
10
51, EXT. INV.
1
50
0.10
0.01
100K
10
1K
100
0.01
10K
0.10
T IME IN S E C ONDS
51, VERY INV.
Figure 7-19: 50 and 51 overcurrent relay characteristics
The pickup level is set by the tap setting, which is usually set in CT secondary amperes but may be set in primary
amperes on some microprocessor-based relays.
Each relay curve has a time dial setting which allows the curve to be shifted up or down on the time-current
characteristic curve. In figure 7-19, the time dial settings are different to give enough space between the curves to
show their differences.
The above are IEEE-standard curves; others are available, depending upon the relay make and model. A solidstate electronic or microprocessor-based relay will have all of these curves available on one unit;
electromechanical relays must be ordered with a given characteristic that cannot be changed.
The 50 instantaneous function is only provided with a pickup setting. The 30ms delay shown in figure 7-19 for the
50 function is typical and takes into account both the relay logic operation and the output contact closing time.
Most microprocessor-based units will also have an adjustable delay for the 50 function; when an intentional time
delay is added the 50 is referred to as a definite-time overcurrent function. On solid-state electronic and
microprocessor-based relays, the 50 function may be enabled or disabled. On electromechanical relays, the 50
function can be added as an instantaneous attachment to a 51 time-overcurrent relay. If a relay has both 50 and
51 functions present and enabled is referred to as a 50/51 relay.
Typically, overcurrent relays are employed as one per phase. In solidly-grounded medium voltage systems, the
most common choice for ground fault protection is to add a fourth relay in the residual connection of the CTs to
monitor the sum of all three phase currents. This relay is referred to as a residual ground overcurrent or 51N (or
50/51N) relay.
The CT arrangement for 50/51 and 50/51N relays for a solidly-grounded system is shown in figure 7-20.
SOURCE
A
RELAY CURRENT INPUTS

50/51-A
50/51-B
50/51-C
50/51-N
LOAD
Figure 7-20: Overcurrent relay arrangement with CTs, including 50/51N
29
For a low-resistance-grounded system, the use of an overcurrent relay connected to a CT in the service
transformer or generator neutral is usually the best option. This CT should have a ratio smaller than the phase
CTs, and the relay pickup range in conjunction with the neutral CT should allow a pickup as low as 10% of the
neutral resistor rating. For a feeder circuit downstream from the service transformer, a zero-sequence CT is
recommended, again with a ratio small enough to allow a pickup as low as 10% of the neutral resistor rating.
When an overcurrent relay is utilized with a zero-sequence CT it is referred to as a 50G, 51G or 50/51G relay
depending upon relay type used. Figure 7-21 shows typical arrangements for both these applications.
TRANSFORMER NEUTRAL
C
RELAY CURRENT INPUT
51N
ZERO-SEQUENCE
A
RELAY CURRENT INPUT

50G
Figure 7-21: Transformer neutral and zero-sequence ground relaying applications for
resistance-grounded systems
For ungrounded systems, the ground detection methods in Section 6 are recommended since little ground current
will flow during a single phase-to-ground fault. Low voltage solidly-grounded systems are discussed below.
The typical application of phase and residual neutral ground overcurrent relays in one-line diagram form is
shown in figure 7-22.
[3]
CT
600:5
[3]
51
51N
52
Figure 7-22: Typical application of overcurrent relays
In figure 7-22, the designation 52 is the IEEE Std. C37.2-1996 designation for a circuit breaker. The phase
relays are designated 51 and the residual ground overcurrent relay is designated 51N (both without instantaneous
function). The bracketed [3] denotes that there are three phase overcurrent relays and three CTs. The dotted
line from the relays to the circuit breaker denotes that the relays are wired to trip the circuit breaker on an
overcurrent condition.
Another type of overcurrent relay is the voltage-restrained overcurrent relay 51 V and the voltage-controlled relay
51C. Both are used in generator applications to allow the relay to be set below the generator full-load current due
to the fact that the fault contribution from a generator will decay to a value less than the full-load current of the
generator. The 51C relay does not operate on overcurrent unless the voltage is below a preset value. The 51 V
relay pickup current shifts as the voltage changes, allowing it to only respond to overcurrents at reduced voltage.
Both require voltage inputs, and thus require voltage transformers for operation.
30
B.) Directional overcurrent relays (Devices 67, 67N)

When fault currents can flow in more than one direction with respect to the load current it is often desirable
to determine which direction the fault current is flowing and trip the appropriate devices accordingly. This is
usually due to the need to de-energize only those parts of the power system that must be de-energized to contain
a given fault.
Standard overcurrent relays cannot distinguish the direction of the current flow. Directional relays (67, 67N) are
required to perform this function.
An important concept in the application of directional overcurrent relays is polarization. Polarization is the method
used by the relay to determine the direction of current flow. For phase directional overcurrent relays, this is
accomplished by the use of voltage transformers, which provide a voltage signal to the relay and allow it to
distinguish the current direction. The details of polarization methods are not discussed here, but can be found in
[3]. Because the voltage on a faulted phase can be unreliable, each phase is restrained via the voltage from a
different phase. Care must be used when defining CT polarities as each manufacturer typically defines a preferred
polarity to match the their standard connection diagrams.
Polarization for a 67N relay is more difficult. They must be polarized with zero-sequence current or zero-sequence
voltage. Electromechanical 67N relays must be polarized via either a CT in the source transformer neutral
(zero-sequence current polarization) or three VTs connected with a wye-connected primaries and broken-delta
connected secondaries (refer to figure 6-9 in Section 6 for an example of the wye-broken delta connection with
ferroresonance-swamping resistor). Solid-state 67N relays usually must be polarized the same way but do
sometimes offer a choice of either method. Microprocessor-based relays typically offer a choice of either method
and, in some cases, can self-polarize by calculating the zero-sequence voltage from the measured three-phase
line voltage.
As an example of the effectiveness of directional overcurrent relays, consider the primary-selective system
arrangement from Section 5 of this guide. The primary main and tie circuit breakers and an example of protective
relaying for those circuit breakers are shown in figure 7-23.
UTILITY
FEED
#1
UTILITY
FEED
#2
VT [3]
FAULT
CURRENT
FLOW
VT [3]
FAULT
FAULT
CURRENT
FLOW
[3]
CT
600:5
[3]
AUX
VT.
[3]
AUX
VT.
[3]
FAULT
CURRENT
FLOW
[3]
51
51
51N
51N
52-M1
CT
600:5
[3]
52-M2
67, 67N
TRIP
DIRECTION
67, 67N
TRIP
DIRECTION
FAULT
CURRENT
FLOW
CT
600:5
[3]
FAULT
CURRENT
FLOW
67N
52-T
N.C.
67
[3]
67N
67
CT
600:5
[3]
[3]
[3]
51
51N
Figure 7-23: Example protective relaying arrangement for closed-transition primary-selective system
In figure 7-23 the bus tie circuit breaker is normally-closed, paralleling the two utility feeds. Each main circuit
breaker and the bus tie circuit breaker are protected via 51 and 51N relays. The mains also have 67 and 67N
relays. Note that the 67 relays are polarized via the line voltage transformers, and auxiliary voltage transformers
connected in wye-broken delta are supplied for polarization of the 67N relays. The polarization results in the
indicated tripping directions for these relays. The need for the 67 and 67N relays can be demonstrated by
considering a fault on one of the utility feeds. Should utility feed #2, for example, experience a fault, the fault
current will be supplied both from the upstream system feeding utility feed #2 and from utility feed #1 through
circuit breakers 52-M1, 52-T, and 52-M2. Because the 51 and 51N relays for 52-M1 and 52-M2 are likely set
identically, they will both respond to the fault at the same time, tripping 52-M1 and 52-M2 and de-energizing the
entire downstream system. To avoid this, the 67 and 67N relays are set to coordinate with the 51 and 51N relays,
respectively, so that the 67 and 67N relays trip first. For a fault on utility feed #2, the 67 and 67N relays for 52-M1
31
will not trip due to the fact that the current is flowing in the direction opposite to the tripping direction. However, the
67 and 67N relays on 52-M2 will sense current in the tripping direction and trip 52-M2. The downstream system is
still energized by 52-M1 and 52-T after 52-M2 trips.
C.) Directional power relays (Device 32)

Directional power relays function when the measured real power flow in the tripping direction is exceeded. They
are used when the power flow in a given direction is undesirable or harmful to system components.
One common use of 32 relays is at the utility service when onsite generators are paralleled with the utility. Under
normal conditions, the incoming power from the utility is measured and generator power output is controlled so
that power is not exported to the utility. Should the generator controls malfunction, the generator may begin to
source power to the utility. The 32 relay would then trip the service breaker offline, isolating the system from the
utility. Of course, this would not apply if surplus power is to be intentionally sold to the utility in a co-generation
arrangement.
Another use of the 32 relay is for the anti-motoring of generators, should prime-mover power be lost.
32 relays are most often supplied as single-phase devices, or as a single-phase function in the case of
microprocessor-based relays. They require both current and voltage sensing to function.
In most applications, 32 relays should be time-delayed to allow the system to ride through momentary power swings.
D.) Undervoltage relays (Device 27)

Undervoltage relays operate when the system voltage falls below a pre-determined level. They are used in
multiple applications. The most common application is as a bus undervoltage relay, which alarms or trips a bus
offline if the system voltage becomes unacceptably low. In this application, the relay should be time-delayed to
ride through momentary dips in voltage, for example for a fault downstream from the relay.
27 relays are commonly used as the signal to an automatic bus transfer system to initiate the transfer from
a failed source to an active source. In this application, also, they should be delayed to ride through momentary
dips in voltage.
27 relays may also be used as permissive devices, for example to disallow closure of a circuit breaker if the
system voltage is not above a pre-defined level. In this application the relay is typically configured to have
instantaneous pickup, with time-delayed drop-out to insure that the system voltage has been above the preset
level for a specified period of time before circuit breaker closure is allowed.
E.) Phase-sequence voltage relays (Device 47)

47 relays generally detect the negative-sequence component of the system voltage, and are thus inherently
three-phase devices. They may be set in terms of voltage balance or in terms of negative sequence voltage. They
are used in a variety of applications, usually in conjunction with 27 and/or 59 relays.
For protection of motors, 47 relays are useful since a loss of one phase may not be detected for a running motor.
This is due to the fact that a lightly-loaded motor (or group of lightly-loaded motors) may keep the voltage on the
lost phase high enough to avoid pickup by a 27 relay on that phase. This fact usually justifies the use of the 47
relay whenever 27 relays are used for bus or motor protection.
47 relays may also be used for permissive functions in conjunction with a 27 relay, as described above. In this role
the 47 relay helps insure that a circuit breaker does not close if the system phase rotation is reversed, such as by
the swapping of phase cables.
As with the 27 relay, the 47 relay should have a time delay to allow the system to ride through transient
conditions. When a 27 and 47 relay are combined into the same electromechanical or solid-state electronic
device, the device is referred to as a 27/47 relay.
32
F.) Overvoltage relays (Device 59)

59 relays respond to voltages above a pre-determined level. They are most often used in conjunction with 27
relays in generator applications to protect voltage-sensitive devices from overvoltage. They may also be used as
permissive devices, usually in conjunction with 27 relays. Either application gives a voltage window within which
the system is allowed to operate. In this application 59 relays should be time-delayed just as 27 relays are.
59 relays may also be used for ground-fault detection on high-resistance grounded or ungrounded systems.
Application for a high-resistance grounded system is shown in Section 6 figure 6-12. For an ungrounded system
the 59 relay may be used across the broken-delta secondary of a ground-detection VT circuit, such as the circuit
shown in Section 6 figure 6-9.
When an electromechanical or solid-state electronic relay includes both 27 and 59 functions it is referred to as a
27/59 relay. When an electromechanical or solid-state electronic relay includes 27, 47, and 59 functions it is
referred to as a 27/47/59 relay.
G.) Lockout relays (Device 86)

The lockout relay is used to trip a device and prevent its reclosure until the lockout relay is reset. In most cases
the lockout relay is essentially a switch, and in fact is typically mounted in close proximity to circuit breaker control
switches. The relay is spring-loaded, and a trip coil, when energized, causes the lockout relay to trip the
connected devices and prevent them from reclosing. There is typically a conspicuous target on the lockout relay to
alert operating personnel that it has tripped. When the lockout relay is reset, the opening springs are compressed
and the relay is ready for the next tripping operation.
86 relays are commonly used where one protective relay must trip several protective devices, and where
reclosure of the tripped devices needs to be controlled to avoid closing onto a fault.
H.) Differential relays (Device 87)

Differential relays operate on the principle that if the current flowing into a device does not equal the current
flowing out, a fault must exist within the device.
Differential relays generally fall within one of two broad categories: Current-differential or high-impedance
differential.
Current-differential relays are typically used to protect large transformers, generators, and motors. For these
devices detection of low-level winding-to-ground faults is essential to avoid equipment damage. Current differential
relays typically are equipped with restraint windings to which the CT inputs are to be connected.
For electromechanical 87 current differential relays, the current through the restraint windings for each phase is
summed and the sum is directed through an operating winding. The current through the operating winding must
be above a certain percentage (typically 15%-50%) of the current through the restraint windings for the relay to
operate. For solid-state electronic or microprocessor-based 87 relays the operating windings exist in logic only
rather than as physical windings.
A typical application of current-differential relays for protection of a transformer is shown in figure 7-24. In
figure 7-24, the restraint windings are labeled as R and the operating windings are labeled as O. Because
the delta-wye transformer connection produces a phase shift, the secondary CTs are connected in delta to
counteract this phase shift for the connections to the relays. Under normal conditions the operating windings will
carry no current. For a large external fault on the load side of the transformer, differences in CT performance in
the primary vs. the secondary (it is impossible to match the primary and secondary CTs due to different current
levels) are taken into account by the proper percentage differential setting. Because the CT ratios in the primary
vs. secondary will not always be able to match the current magnitudes in the relay operating windings during
normal conditions, the relays are equipped with taps to internally adjust the current levels for comparison. The
specific connections in this example apply to a delta primary/wye secondary transformer or transformer bank only.
The connections for other winding arrangement will vary, in order to properly cancel the phase shift. For many
solid-state electronic and microprocessor-based relays, the phase shift is made internally in the relay and the CTs
may be connected the same on the primary and secondary sides of the transformer regardless of the transformer
winding connections. The manufacturers literature for a given relay make and model must be consulted when
planning the CT connections.
33
SOURCE
A
RELAY CURRENT INPUTS
A-PHASE
R
R
O
B-PHASE
R
R
O
C-PHASE
R
R
O
LOAD
Figure 7-24: Typical application of current-differential relays for delta-wye transformer protection
Percentage-differential characteristics are available as fixed-percentage or variable percentage. The difference is

that a fixed-percentage relay exhibits a constant percentage restraint, and for a variable-percentage relay the
percentage restraint increases as the restraint current increases. For an electromechanical relay, the percentage
characteristic must be specified for each relay; for solid-state electronic or microprocessor-based relays these
characteristics are adjustable. For transformers relays with an additional harmonic restraint are available.
Harmonic restraint restrains the relay when certain harmonics, normally the 2nd and 5th, are present. These
harmonics are characteristic of transformer inrush and without harmonic restraint the transformer inrush may
cause the relay to operate.
An important concept in the application of differential relays is that the relay typically trips fault interrupting devices
on both sides of the transformer. This is due to the fact that motors and generators on the secondary side of the
protected device will contribute to the fault current produced due to an internal fault in the device. An example
one-line diagram representation of the transformer differential protection from 7-24 is given in figure 7-25:
87T ZONE OF
PROTECTION
CT
[3]
52
[3]
86T
87T
CT
[3]
Figure 7-25: Transformer differential relay application from figure 7-24 in one-line diagram format
Note that the secondary protective device is shown as a low voltage power circuit breaker. It is important that the
protective devices on both sides of the transformer be capable of fault-interrupting duty and suitable for relay
tripping.
In figure 7-25 a lockout relay is used to trip both the primary and secondary overcurrent devices. The lockout relay
is designated 86T since it is used for transformer tripping, and the differential relay is denoted 87T since it is
protecting the transformer. The wye and delta CT connections are also noted.
34
An important concept in protective relaying is the zone of protection; a zone of protection is the area that a given
protective relay and/or overcurrent device(s) are to protect. While the zone of protection concept applies to any
type of protection (note the term zone selective interlocking as described earlier in this section), it is especially
important in the application of differential relays because the zone of protection is strictly defined by the CT
locations. In figure 7-25 the zone of protection for the 87T relay is shown by the dashed-line box around the
transformer. For faults within the zone of protection, the currents in the CTs will not sum to zero at the relay
operating windings and the relays will operate. Outside the zone of protection the operating winding currents
should sum to zero (or be low enough that the percentage restraint is not exceeded), and therefore the relays
will not operate.
The other major category of differential relays, high-impedance differential relays, use a different principle for
operation. A high-impedance differential relay has a high-impedance operating element, across which the voltage
is measured. CTs are connected such that during normal load or external fault conditions the current through the
impedance is essentially zero. But, for a fault inside the differential zone of protection, the current through the
high-impedance input is non-zero and causes a rapid rise in the voltage across the input, resulting in relay
operation. A simplified schematic of a high-impedance differential relay is shown in figure 7-26 to illustrate the
concept. Note that the relay only has one set of input terminals, without restraint windings. This means that any
number of CTs may be connected to the relay as needed to extend zone of protection, so long as the CT currents
sum to zero during normal conditions. Also note that a voltage-limiting MOV connected across the high-impedance
input is shown. This is to keep the voltage across the input during a fault from damaging the input.
ZONE OF PROTECTION
HIGH-IMPEDANCE
DIFFERENTIAL RELAY
MOV
Z
Figure 7-26: High-impedance differential relay concept
High-impedance differential relays are typically used for bus protection. Bus protection is an application that
demands many sets of CTs be connected to the relays. It is also an application that demands that that relay be
able to operate with unequal CT performance, since external fault magnitudes can be quite large. The highimpedance differential relay meets both requirements.
Figure 7-27 shows the application of bus differential relays to a primary-selective system. Note that in figure 7-27
the zones of protection for Bus #1 and Bus #2 overlap. Here the 86 relay is extremely useful due to the large
number of circuit breakers to be tripped. Note that all circuit breakers attached to the protected busses are
equipped with differential CTs and are tripped by that busses respective 86 relay. The 87 relays are denoted 87B
since they are protecting busses. The same applies for the 86B relays. Note also that the protective zones
overlap; this is typical practice to insure that all parts of the bus work are protected.
The high-impedance differential relay is typically set in terms of voltage across the input. The voltage setting is
typically set so that if one CT is fully saturated and the others are not the relay will not operate. By its nature, the
high-impedance differential relay is less sensitive than the current-differential relay, but since it is typically applied
to protect bussing, where fault magnitudes are typically high, the additional sensitivity is not required.
35
BUS #1 ZONE OF
PROTECTION
CT
[3]
52-M1
BUS #2 ZONE OF
PROTECTION
[3]
[3]
87B
87B
86B
86B
CT
[3]
52-T
N.C.
BUS #1
CT
[3]
CT
[3]
CT
[3]
52-M2
BUS #2
CT
[3]
CT
[3]
CT
[3]
CT
[3]
CT
[3]
Figure 7-27: High-impedance differential relaying applied to a primary-selective system
I.) Other Protective Relay Types

Only a small selection of the most commonly-used protective relay types are given here. For more in-depth
descriptions of their application, and for descriptions of other protective relay types, see reference [3]. Those
protective relay functions that are typically used for motors are described in section 8 of this guide.
Ground-fault protection for solidly-grounded systems 600 V and below

Because the ground fault is the most common type of system fault, and because low voltage systems are
necessarily the largest portion of most industrial and commercial facilities, low voltage ground-fault protection has
become a specialized area of development for system protection. Unlike the relayed ground-fault protection
systems shown in VIII above, these systems are specially designed to provide sensitive protection for four-wire
systems with imbalanced loads.
As noted in Section 6, the National Electrical Code [1] requires ground-fault protection for most solidly-grounded
electrical systems 1000 A or more and above 150 volts to ground but not exceeding 600 V phase-to-phase. For
this reason, the ground-fault systems described herein are prevalent in systems meeting these criteria.
The low-ground fault protection methods in this section are for solidly-grounded systems only and augment
the ground detection methods given in Section 6 for ungrounded and high-resistance-grounded systems.
Low-resistance grounded systems at the low voltage level are uncommon but can be protected per the guidelines
given above for relayed ground fault protection.
A.) Ground-fault protection for radial systems

Ground-fault protection for low voltage radial systems is straightforward. For electronic trip units the tripping logic
is typically built into the circuit breaker, and only the neutral CT or sensor must be connected to complete the
ground fault protection system. Such an arrangement is illustrated in figure 7-28:
SOURCE
A
N
IF
ELECTRONIC-TRIP
CIRCUIT BREAKER
MAIN OR SYSTEM
BONDING
JUMPER
TRIP
UNIT W/
GFT
(LSG,
LSIG)
CB CURRENT
SENSORS
IF
NEUTRAL SENSOR
OR CT
SAFETY GROUND MAY

OR MAY NOT
EXIST ON
NEUTRAL
SENSOR/CT
CIRCUIT
IF
GROUND
FAULT
IF
LOAD
Figure 7-28: Low voltage ground fault protection for 4-wire radial system with electronic-trip circuit breaker
36
In figure 7-28 the neutral sensor may be an air-core CT or a conventional iron-core CT. Note that the ground
fault current is diverted around the neutral sensor when it is placed on the load-side of the main or system
bonding jumper (see Section 6 for the definition of main and system bonding jumpers and related discussion).
Under normal unbalanced-load conditions the neutral sensor will detect the neutral current and prevent the
circuit breaker from tripping. Note that if the system is a 3-wire system without a system neutral the neutral
CT is omitted.
If the circuit breaker is not equipped with an electronic trip system, an external ground fault relay may be used
with a zero-sequence sensor to trip the circuit breaker. The circuit breaker must be equipped with a shunt trip
attachment in this case. Figure 7-29 shows an example of this arrangement. In figure 7-29 the external ground
fault relay is noted as GS. In low voltage systems this is the typical notation rather than 51G, although 51G
could be used also. Note that in a 3-wire system the neutral is omitted, and the zero-sequence sensor includes
the phase conductors only.
SOURCE
A
N
IF
BONDING
JUMPER
IF
Circuit Breaker
without Electronic Trip
ST
ZEROSEQUENCE
SENSOR
OR CT
GS
IF
GROUND
FAULT
IF
SAFETY GROUND MAY

OR MAY NOT
EXIST ON
ZERO-SEQUENCE
SENSOR/CT
CIRCUIT
LOAD
Figure 7-29: Low voltage ground fault protection for 4-wire radial system without electronic trip
circuit breaker
These methods provide sensitive ground fault protection for solidly-grounded radial systems. However, if multiple
sources are involved a more involved system is required in order to obtain reliable ground-fault protection.
B.) Modified-differential ground fault systems

Because 4-pole circuit breakers are not in common use in the United States, the issue of multiple ground current
return paths has a large effect upon ground-fault protection in 4-wire systems. To illustrate this point, consider a
secondary-selective system as shown in figure 7-30.
A ground fault on one bus has two return paths: Through its source-transformer main/system bonding jumper or
the other source-transformer main/system bonding jumper neutral. How much ground fault current flows in each
path is dependent upon the ground or zero-sequence impedances of the system, which is difficult to evaluate.
Therefore, let us assume a factor of A x the total ground-fault current flows through the source transformer
main/system bonding jumper neutral and B x the total ground-fault current flows through the other transformer
main/system bonding jumper, where A + B = 1. As can be seen from figure 7-30, the ground-fault protection for
the faulted bus can be de-sensitize or, worse, the wrong circuit breaker(s) may trip.
37
SOURCE
#1
A
SOURCE
#1
N
IF
MAIN OR SYSTEM
BONDING
JUMPER
B x IF
A x IF
CB-M1
B x IF
CB-M2
B x IF
IF
GROUND
FAULT
IF
CB-T
B x IF
SOURCE
#1
LOADS
SOURCE
#2
LOADS
Figure 7-30: Secondary-selective system with radial ground-fault protection of figure 7-28 applied
The solution is the modified-differential ground fault system. A typical example of such a system is shown in
figure 7-31:
SOURCE
#1
A
SOURCE
#1
N
IF
MAIN OR SYSTEM
BONDING
JUMPER
B x IF
A x IF
A x IF
GT
IF
B x IF
A x IF
CB-M1
GM1
B x IF
IF
IF
B x IF
B x IF
CB-M2
GM2
B x IF
B x IF
GROUND
FAULT
IF
CB-T
B x IF
SOURCE
#1
LOADS
SOURCE
#2
LOADS
Figure 7-31: Modified-differential ground-fault protection for secondary-selective system
In figure 7-31 the breaker internal sensors are shown, but the trip units are omitted for clarity. The ground-fault
function for CB-M1 is noted as GM1, for CB-M2 is noted as GM2, and for CB-T is noted as GT. In this
arrangement, regardless of the ground current dividing factors A and B the correct circuit breakers will sense the
ground fault and trip. Note that this system works regardless of whether CB-T is normally-open or normally-closed.
Non-electronic circuit breakers could also be used, but external CTs and ground relays would have to be utilized.
For unusual system arrangements or arrangements with more then two sources, the system of figure 7-31 can be
expanded. These are usually custom-engineered solutions.
C.) Four-pole circuit breakers

Another possible option, in lieu of the modified differential ground fault system, is the use of four-pole circuit
breakers. These switch the neutral as well as the phase conductors, separating the neutrals of multi-source
circuits. This method will not work if sources are paralleled. Four-pole circuit breakers are not common in the
United States for this reason, as well the increased physical equipment sizes they necessitate.
38
D.) Ground fault time-current characteristics

1K
10
100
1
10K
Figure 7-32 shows typical time-current characteristics for the ground fault function of an electronic-trip circuit breaker.
1000
100
100
10
10
0.10
0.10
10K
10
1K
0.01
1
100
0.01
0.5
T IME IN S E C ONDS
1000
Figure 7-32: Typical electronic-trip circuit breaker ground-fault protection time-current characteristic
This characteristic is adjustable both for pickup and time delay. Discrete relays for use with non-electronic circuit
breakers are also available with similar characteristics.
Care must be taken when coordinating ground-fault protection if multiple levels of ground-fault protection do not exist
downstream from the service or source of a separately-derived system. The NEC Article 230.95 (A) service-entrance
requirement [1] for a maximum of 1200 A pickup and maximum 1-second delay at 3000 A ground-fault current can
lead to a lack of coordination for downstream feeder and branch-circuit ground faults. This is one of the reasons for
the use of other than solidly-grounded systems where maximum system reliability is to be achieved.
Surge protection
Surge protection is protection of conductors and equipment against the effects of voltage surges. These are
usually due to lightning, although switching transients can also cause damaging overvoltages. Unlike overvoltage
relaying, surge protection is directly connected to the power circuit, and for the best protection is usually located
as close as physically practical to the protected equipment.
A.) Medium voltage surge protection

Medium voltage surge protection is generally accomplished with surge arresters. A surge arrester exhibits an
impedance which decreases with the line voltage. Older technologies included spark-gap arresters, which
provided surge protection when the voltage became high enough to ionize the air in an internal spark gap.
Modern surge arrestors employ metal-oxide varistor (MOV) technology, which exhibit a non-linear resistance
which changes with the applied voltage. These generally provide better protection than spark-gap arrestors,
although they do have limits on the continuous voltage that may be applied to the arrester without damage.
Surge arrestors are connected phase-to-grounded, even on ungrounded systems. When MOV surge arrestors are
used, they must be sized for the maximum anticipated phase-to-ground voltage. MOV surge arrestors have a
duty-cycle voltage rating and a maximum continuous operating voltage (MCOV) rating; the MCOV rating is the
quantity to be compared to the phase-to-ground voltage. Also of importance is the arresters classification as
distribution-class, intermediate-class, or station-class; these classes are defined in terms of the energy the
39
surge arrestor can absorb without damage, in ascending order as listed. Table 7-14 gives commonly-applied MOV
surge arrestor ratings vs. the system voltage. In general, use of surge arrestors with the lowest MCOV exceeding
the anticipated line-to-ground voltage provides the best protection. Detailed insulation coordination studies can
also be performed with the use of transient analysis software. For low-resistance-grounded systems, selection of
the lowest acceptable surge arrestor rating involves comparing the overvoltage vs. time characteristic of the surge
arrestor to the maximum time a ground fault will remain on the system prior to tripping.
For motor circuits, surge capacitors are also often employed. These provide dV/dt protection for the motor
windings. Care must be used when sizing surge capacitors and the effects of harmonic currents must be
evaluated to insure the capacitors will not rupture.
Both surge capacitors and surge arresters are applied without dedicated overcurrent protection. For this reason,
failure of these devices will result in some equipment damage. In the case of surge arresters, use of polymer
housings will result in minimal damage should the arrester fail; the housing will simply split to relieve the internal
overpressure. Use of porcelain housings which can sustain large internal overpressures can result in severe
damage should the arrester fail. In the case of surge capacitors, since they are typically filled with dielectric fluid
and have steel housings they can sustain high internal overpressures, and failure of the housing due to internal
overpressure can result in catastrophic equipment damage and risk to personnel.
Applicable standards include IEEE Std. C62.11 and IEEE Std. C62.22.
Table 7-14: Commonly-applied ratings for metal-oxide surge arrestors
Duty-Cycle Voltage (kV)`
4-Wire EffectivelyGrounded Neutral

System1
MCOV (kV)
3-Wire Grounded and

Resistance-Grounded
Wye Systems2
2.6
4160 Y/2400
2400
5.1
8320 Y/4800
4160
4800
7.7
12000 Y/6930
6900
12470 Y/7200
10
8.4
12
10.2
15
12.7
13200 Y/7620
13800Y/7970
20780 Y/12000
12000
12470
18
40
15.3
22860 Y/13200
13200
24940 Y/14400
13800
34500 Y/19920
20780
21
17.0
24
19.5
27
22.0
30
24.4
22860
36
29.0
24940
Use of this system category requires a solid ground conductor (non-earth) path back to the upstream transformer
or generator neutral.
Includes grounded-wye systems where the path to the upstream transformer neutral includes an earth path
B.) Low voltage surge protection

Strictly speaking, low voltage surge protection is via smaller versions of the MOV arresters used in medium
voltage systems. Various versions of these are available, including mountings which fit the circuit-breaker spaces
in panelboards. These are generally manufactured to the same standards as their medium voltage counterparts.
A more commonly-used device is the Transient Voltage Surge Suppressor (TVSS). The TVSS is a device
classification unique to UL and is intended to provide a degree of protection for sensitive utilization equipment
against the effects of transient voltages. TVSSs usually include MOVs for voltage clamping as well as filtering for
surge attenuation. They are designed and manufactured to UL Std.1449, which requires consideration for device
failure modes due to the intended installation location. UL 1449 defines three general installation classifications:
Permanently Connected, Cord Connected, and Direct Plug-In. TVSSs are intended for use on the load side of an
overcurrent protective device such as a circuit breaker. They often include status indication for identification of a
failed unit. Transient Voltage Surge Suppressors are generally considered the preferred means of low voltage
surge protection.
Protection of specific system components

A.) Power cables
For low voltage power cables, so long as they are protected at their ampacities per NEC Article 240.5 [1] they can
be considered adequately protected.
For medium voltage power cables, proper protection requires both sizing per the NEC ampacity tables and
comparing the overcurrent protective device time-current characteristic with the cable damage characteristic per
IPCEA Publication P-32-382. The cable damage characteristic for a #1/0 AWG copper conductor is shown in
figure 7-33 as compared with the time-current characteristic of a 300E fuse. Per NEC Article 240.100 (C) [1], the
factor to be considered is the short-circuit performance of the cable; indeed, NEC Article 240.101 (A) allows the
rating of a fuse protecting the conductor to be up to 3 times the ampacity of the conductors (6 times for a circuit
breaker or electronically-actuated fuse). As can be seen in figure 7-33, the 300E fuse indeed does not provide
overload protection for the cable at its ampacity (200 A), but is within the limits of NEC 240.101 (A). Even though
the fuse will allow the cable ampacity to be exceeded, the continuous load on the cable should not exceed the
published conductor ampacity. The 300E fuse does, however, adequately protect the cable for short-circuits, as
evidenced by the cable damage characteristic being to the right and above the fuse characteristic.
When overcurrent relays are used to protect medium voltage power cables the procedure is the same, but a
51 pickup of no more than 125%-150% of the maximum load on the cable.
B.) Transformer protection

Transformer protection consists of both overload protection and short-circuit protection.
Overload protection is usually accomplished via proper selection of the secondary overcurrent protective device.
NEC Article 450 [1] gives specific primary and secondary overcurrent device ratings that may not be exceeded.
These vary depending upon the accessibility of the transformer to unqualified persons and the impedance of the
transformer. The smallest protective device that allows the rated full-load current of the transformer gives the best
practical overcurrent protection. Increasing the secondary overcurrent device size beyond this may be necessary
for short-term overloads or for coordination with downstream devices, but in any case the requirements of NEC
Article 450 must be met.
41
100
100K
10K
1K
100
10
1000
1000
#1/0 AWG CABLE
100
300E C/L FUSE
#1/0 AWG CABLE
10
0.10
T IME IN S E C ONDS
10
0.10
100K
10K
1K
0.01
10
100
0.01
Figure 7-33: Medium voltage cable protection example
Short-circuit protection involves comparison of the transformer damage curve per IEEE Std. C57.109-1993 with
the primary overcurrent device time-current characteristic. In general, the damage curve must be to the right and
above the primary overcurrent device characteristic. Another constraint on the primary overcurrent device is that it
must be capable of withstanding the inrush of the transformer without tripping (and without damage for currentlimiting fuses). An example time-current characteristic showing protection for a 1000 kVA 13.2 kV Delta:
480 Y/277 V, 5.75%Z dry-type transformer is shown in figure 7-34. The transformer is protected with a 65E
current-limiting primary fuses and a 1200 A electronic-trip secondary circuit breaker. As can be seen from the
figure, the fuses do withstand the inrush without damage since the inrush point is to the left and below the fuse
minimum melt curve. The transformer is protected from short-circuits by the primary fuses. The secondary circuit
breaker provides overload protection at the full-load current of the transformer. Note that the primary fuse and
secondary circuit-breaker characteristics overlap for high fault currents; this is unavoidable and is considered
acceptable. Note also that the fuse curve and the transformer damage curve overlap; this is unavoidable but these
should overlap at the lowest current possible. For currents below the fuse/transformer damage curve overlap the
secondary circuit breaker must protect the transformer; the lower the point of overlap, the more likely the fault is
an external fault on the load side of the secondary circuit breaker and therefore greater chance the secondary
circuit breaker will effectively protect the transformer for faults in this region.
Also note that the transformer damage characteristic is shown twice. Because transformer is a delta-wye
transformer, a ground-fault on the secondary side of the transformer will result in only 57.7% of the maximum
three-phase primary fault current while one secondary winding experiences the full fault current. This is illustrated
in Figure 7-35, as well as the corollary for delta-delta transformers. The damage characteristic has therefore been
shifted to 57.7% of its published value to account for secondary line-to-ground faults. Also, the shifted curve has
another, more conservative curve shown; this is the frequent-fault curve and is applicable only to the secondary
overcurrent device since faults between the transformer secondary and the secondary overcurrent protective
device should not be frequent.
Additional devices, such as thermal overload alarms/relays and sudden-pressure relays, are also available for
protection of transformers. These are typically specified with the transformer itself and can provide very good
protection. However, even if these devices are installed the primary and secondary overcurrent devices must be
coordinated with the transformer as described above.
42
1000
10K
10
1K
100
1
1000
XFMR
PRI. FUSE
100
100
XFMR
SEC. CB
10
TX Inrush
0.10
0.10
10K
10
1K
0.01
1
100
0.01
0.5
T IME IN S E C ONDS
10
Figure 7-34: Example protection for a 1000 kVA, 13.2 kV Delta: 480 Y/277V, 5.75%Z dry-type transformer
Differential protection for transformers, as described above, is very effective for transformer internal faults.
If differential protection is supplied it is the primary protection for internal faults and will operate before the
primary overcurrent device. The primary overcurrent device serves as a backup protective device for internal
faults in this case.
Figure 7-35: Fault-current flow for delta-wye transformer L-N faults and delta-delta transformer L-L faults
C.) Generator protection

The subject of generator protection is a complex one, and due to this fact it is not presented here. Please refer to
[3] for detailed descriptions of generator protection methods, as well as descriptions of protective relay types that
are not discussed above that used for generator protection.
D.) Motor protection

The protection of motors is presented in AC motors, motor control and motor protection section (section 8 of
this guide).
43
E.) Other devices and additional information

For protection other devices, refer to [3] and/or the applicable standards for the device in question. For additional
information on the protection of cables and transformers, please refer to [3].
Protection selectivity
The selectivity of protection refers to its ability to isolate an abnormal condition to the smallest portion of the
system possible. In most cases selectivity is a function of how well-coordinated the overcurrent protective devices
in the system are. As an example, consider the system of figure 7-36:
UTILITY SERVICE
A
B
C
D
E
F
G
Figure 7-36: Example system for selectivity discussion
Figure 7-36 shows a small radial system with a medium voltage utility service, a service substation consisting of a
primary switch step-down transformer protected by a primary fuse, and a secondary switchboard. One of the
switchboard feeder circuit breakers is shown feeding al lighting panel and other loads.
For optimum selectivity, a fault at point G should only cause its lighting panel feeder circuit breaker to trip. The
panel main circuit breaker and all devices upstream should not be affected. If the lighting panel feeder circuit
breaker time-current characteristic does not coordinate with that of the lighting panel main, the main may trip, deenergizing the entire panelboard.
Going upstream, a fault at point F should only cause the panelboard main circuit breaker to trip and a fault at point
E should only cause the switchboard main circuit breaker to trip. A fault at point D may cause the switchboard
main circuit breaker to trip or the primary fuse to blow, but the effect on the system is the same since all of the
loads will be de-energized in either event. A fault at point C should only cause the transformer primary fuse to blow.
Lack of selectivity causes more of the system to be de-energized for a fault in a given location. The severity of the
outage increases as the fault location is considered farther and farther upstream. In this example, if the
transformer primary fuses and the upstream utility recloser, protective relays, or fuses are not coordinated the
entire utility distribution line, or a segment of the line, could be de-energized, affecting other customers.
To analyze system selectivity, a time-current coordination study must be performed. This study analyzes the
time-current coordination characteristics of the protective devices in the system and plots them on time current
curves such as those illustrated in this section. Coordination is considered to be achieved between two devices if
their time-current bands show sufficient clear space between them on the time-current curve or, in the case of
protective relays, if sufficient margin for overtravel, manufacturing tolerances, circuit breaker speed, and safety are
achieved.
44
Coordination is not always possible to maintain in the high fault-current ranges. However, in most cases an
acceptable compromise can be reached since high-level faults are a rare occurrence.
Another important concept is that of backup protection. In this case, for a fault at point G if the lighting panel
feeder circuit breaker fails to trip the panelboard main circuit breaker should trip as dictated by its time-current
curve. If selective coordination exists between the panelboard main circuit breaker and the switchboard feeder
circuit breaker, then the switchboard feeder circuit breaker will not trip. So, backup protection must consider one
level upstream vs. primary protection unless additional backup protective devices are installed.
References
[1]
[2]
Alan Greenwood, Electrical Transients in Power Systems, New York, John Wiley and Sons Inc., 1971.
[3]
IEEE Recommended Practice for Protection and Coordination of Industrial Power Systems, IEEE Std. 2422001, December 2001.
[4]
Molded-Case Circuit Breakers, Molded Case Switches and Circuit-Breaker Enclosures, UL 489, Underwriters
Laboratories Inc., April 25, 2002.
[5]
IEEE Standard for Low Voltage AC Power Circuit Breakers Used in Enclosures, ANSI/IEEE Standard C37.131990, October 1990, Reaff. April 8, 1996.
[6]
IEEE Standard Rating Structure for AC High Voltage Circuit Breakers, ANSI/IEEE Standard C37.04-1999, June
1999.
[7]
AC High Voltage Circuit Breakers Rated on a Symmetrical Current Basis Preferred Ratings and Related
Required Capabilities, ANSI Standard C37.06-2000, May 2000.
[8]
IEEE Application Guide for AC High Voltage Circuit Breakers Rated on a Symmetrical Current Basis, IEEE Std
C37.010-1999, September 1999.
[9]
Swindler, D.L., Fredericks, C.J., Modified Differential Ground Fault Protection for Systems Having Multiple
Sources and Grounds, Industry Applications, IEEE Transactions on, Volume 30, Issue 6, Nov. 1994.
45
Section 8:
AC Motors, motor control and motor protection

Introduction
Electric motors are an important part of any electrical system. Because they convert electrical energy to
mechanical energy, they are the interface between the electrical and mechanical systems of a facility.
This creates unique challenges for control and protection which have, in turn, led to unique solutions.
This section gives background on various AC motor types, and the control and protection practices commonly
used for these.
AC motor types
Motors generally consist of two basic assemblies: The stator, or stationary part, and the rotor, or rotating part.
Motors have two sets of windings: armature windings, to which the power is applied, and field windings , which
produce a magnetic field that interacts with the magnetic field from the armature windings to produce torque on
the rotor. This torque causes the rotor to rotate. For most AC motors, the armature windings are located on the
stator, and the field windings are located on the rotor (one exception is the field exciter for a brushless
synchronous motor, as described below). For this reason, in most cases the armature windings are known
synonymously as the stator windings.
AC motors in common use today may be divided into two broad categories: Induction (asynchronous) or
synchronous. These two types of motors differ in how the rotor field excitation is supplied. For induction motors,
there is no externally-applied rotor excitation, and current is instead induced into the rotor windings due to the
rotating stator magnetic field. For synchronous motors, a field excitation is applied to the rotor windings. This
difference in field excitation leads to differences in motor characteristics, which leads in turn to different protection
and control requirements for each motor type.
A.) Induction motors

Induction motors are the workhorses of modern industry. Because they have no applied field excitation, the rotor
windings can be made to be very simple and rugged. The most common motor type is the squirrel-cage motor,
which has rotor windings consisting of copper or cast-aluminum bars solidly connected to conducting end rings on
each end, forming a structure which resembles a squirrel cage [1]. Due to the simple rotor construction, the
squirrel cage motor is rugged and durable, and is the most common type. Wound-rotor motors are also available,
usually for special application where external resistance is applied to the rotor for speed control, as described later
in this section.
An important concept in the application of induction motors is the fact that due to the lack of field excitation, the
motor speed will vary with the torque of the load. Synchronous speed for a given motor is given by the equation:
(8-1)
where
ns
f
p
is the synchronous speed, in RPM

is the frequency in Hz
is the number of poles of the motor, which can be defined as 2 x the number of different magnetic
field orientations around the stator per phase. The minimum number of poles is 2 and the number
of poles is always even.
For an induction motor, the speed will always be less than synchronous speed by a factor known as the slip of the
motor. The motor speed can be expressed as:
(8-2)
where
n
s
ns
is the speed of the motor, in RPM

is the slip
is the synchronous speed of the motor per (8-1) above
Induction motors are classified by application with a design letter which gives an indication of key performance
characteristics of the motor. Table 8-1 gives typical design letter characteristics for induction motors. These are
typical characteristics only for further details consult the specific performance standards for the complete
requirements [2,3].
B.) Synchronous motors

Synchronous motors have a DC field excitation applied to the field windings on the rotor. This has the effect
of allowing the motor to run at synchronous speed. However, the motor produces torque only at synchronous
speed, so for starting the rotor is also equipped with damper windings that allow the motor to be started as an
induction motor.
Table 8-1: Typical characteristics and applications of fixed frequency medium AC squirrel
cage motors (Essentially same as [2] table 10-1 and [3] table 3)
Polyphase
Characteristics
LockedPull-up Breakdown Lockedrotor

torque
torque
rotor
torque
(percent
(percent
current
(percent rated load rated load
(percent
rated load torque)
torque)
rated load
torque)
current)
Slip
(%)
Typical Applications
Relative
Efficiency
Design A
Normal locked
rotor torque and
high locked-rotor
current
70-275a
65-190a
175-300
Not
Defined
0.5-5
Fans, blowers, centrifugal

pumps and compressors,
motor-generator sets, etc.,
where starting torque
requirements are relatively low
Medium
or High
Design B
Normal lockedrotor torque and
normal lockedrotor current
70-275a
65-190a
175-300a
600-800
0.5-5

Medium
or High
Design C
High locked-rotor
torque and
normal lockedrotor current
200-285a
140-195a
190-225a
600-800
1-5
Conveyors, crushers, stirring

machines, agitators,
reciprocating pumps and
compressors, etc., where
starting under load is required
Medium
Design D
High locked-rotor
torque and
high slip
275
Not
defined
275
600-800
High peak loads with or without

flywheels such as punch
presses, shears, elevators,
extractors, winches, hoists, oilwell pumping and wire-drawing
machines
Medium
IEC Design H
High locked rotor
torque and high
locked rotor
current
200-285a
140-195a
190-225a
800-1000
1-5
Conveyors, crushers, stirring

machines, agitators,
reciprocating pumps and
compressors, etc., where
starting under load is required
Medium
IEC Design N
Normal lockedrotor torque and
high locked rotor
current
70-190a
60-140a
160-200a
800-1000
0.5-3

Medium
or High
Higher values are for motors having lower horsepower ratings
Synchronous motors may be further classified as brush or brushless type. The field exciter for a brush-type motor
is typically a DC generator with its rotor mounted on the motor shaft. The output of the DC generator is fed via
brushes and slip rings to the motor field windings. The field exciter for a brushless synchronous motor typically
consists of an AC generator with the field windings on its stator, armature windings on its rotor, and with its rotor
mounted on the motor shaft. The output of the generator is rectified by solid-state rectifier elements also mounted
on the rotor shaft and fed directly to the motor field windings without the need for brushes or slip rings. Because of
the proliferation of solid-state power electronic technology, and because the brushless-type motors require less
maintenance almost all new synchronous motors are brushless-type [1], although many existing installations do
have older brush-type motors in service. In either design the field excitation to the exciter may be varied to vary
the power-factor operation of the motor, and in fact power factor correction is one common use of synchronous
motors since they can be made to operate at leading power factors.
C.) Enclosure types, cooling methods and other general application information
Please refer to [3] for more information on motor enclosure types and cooling methods, as well as additional
general application information for motors.
Motor torque and driven load characteristics

Motors are rated in horsepower (hp; 1hp = 746W) or, occasionally, in watts or kilowatts. In either case, this is the
rated output power of the motor at the motor shaft when the motor is running at full speed. Due to losses in the
motor, the input power will be higher. Due to the motor power factor and these losses, the full-load current of the
motor will be larger than would be otherwise anticipated by looking only at the hp or kW rating. This will be
discussed further later in this section.
At the motor shaft, the rated output power is related to the shaft rotational speed as follows:
(8-3)
where
P
T
n
is the shaft output power in hp

is the shaft output torque in ft-lbf
is the motor speed in RPM
Further, the shaft rotational acceleration is related to the output torque and the inertia of the load as follows:
(8-4)
where
T
J
is the output accelerating torque in ft-lbf

is the total moment of inertia of the motor shaft and rotor plus the driven load, in lb-ft2 (also referred to
as wk2)_ is the shaft acceleration in rpm/min.
Because a= dn/dt, the speed of the motor shaft can be written as:
(8-5)
The inertia of the load (and rotor), then, is crucial to the acceleration rate of the motor shaft (and the load) and
thus to the output speed of the shaft. A typical design B induction motor torque-speed characteristic is as shown in
figure 8-1, along with pertinent characteristics from table 8-1 labeled:
Torque (pu breakdown)
1.2
1
0.8
0.6
0.4
LockedRotor
Torque
0.2
0.
06
0.
12
0.
18
0.
24
0.
3
0.
36
0.
42
0.
48
0.
54
0.
6
0.
66
0.
72
0.
78
0.
84
0.
9
0.
96
Breakdown
Torque
Pull-Up
Torque
Speed (pu synchronous)
Figure 8-1: Typical NEMA design B induction motor speed-torque characteristic
Figure 8-1 shows the motor output torque as a function of shaft speed with full rated voltage applied to the motor.
To show the performance of a motor when connected to a load, a typical speed-torque-characteristic for a fan is
plotted along with the motor speed-torque characteristic in figure 8-2. The load speed-torque characteristic is a
plot of the torque required to drive a load at a given speed. Several points can be made regarding the motor and
load of figure 8-2:
I
The motor locked-rotor torque is greater than the load torque at zero speed. This means the motor can start with
the load connected.
The motor pull-up torque is greater than the load torque during the acceleration period. This means that the
motor can successfully accelerate the load.
The steady-state speed of the motor is where the motor-torque and load-torque curves cross the steady-state
operating point approximately 98.5% synchronous speed. The motor slip is therefore approximately 1.5%
The difference between the motor output torque and the load torque is the accelerating torque for the motor-load
system. The accelerating torque is the same as given in eq. (8-1) above. A plot of the accelerating torque is given
in figure 8-3.
Torque (pu breakdown)
1.2
1
Steady-State Operating
Point
0.8
Motor
Torque
0.6
0.4
0.2
Load
Torque
Accelerating Torque
0.98
0.91
0.84
0.7
0.77
0.63
0.56
0.49
0.42
0.35
0.28
0.21
0.14
0.07
Figure 8-2: Example motor and load speed-torque characteristics
Accelerating Torque (pu motor

breakdown torque)
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.
06
0.
12
0.
18
0.
24
0.
3
0.
36
0.
42
0.
48
0.
54
0.
6
0.
66
0.
72
0.
78
0.
84
0.
9
0.
96
Figure 8-3: Accelerating torque for motor and load of figure 8-2
With the accelerating torque known, the motor (and load) shaft speed can be calculated from eq. (8-4) and (8-5).
In practice, this is best left to computer simulation. A typical plot of the approximate shaft speed, is shown in
figure 8-4.
As can be seen from figure 8-4, for the example shown the motor accelerates to steady-state speed in
approximately 14 seconds. The motor breakdown torque and load and motor moments of inertia (typically
referred to as motor wk2 and load wk2, respectively) must be known to obtain this speed vs. time characteristic.
1.2
1
0.8
0.6
0.4
0.2
72
63
67
.5
54
58
.5
45
49
.5
36
40
.5
27
31
.5
18
22
.5
9
13
.5
4.
5
0
0
Motor Speed (pu synchronous)
The importance of the above analysis lies in the fact that for successful motor starting the motor must be able to
successfully support the load torque during acceleration. If the motor cannot do this, it will stall during starting.
Proper motor selection, considering both the HP and torque characteristics, is essential for proper starting.
Further, for an induction motor the slip is determined by the torque characteristics of the motor and load.
Time (s)
Figure 8-4: Typical motor speed vs. time for the example above
For a synchronous motor, starting analysis is similar since the damper windings of the motor give a speed-torque
characteristic similar to that of an induction motor. For a synchronous motor the steady-state speed is the
synchronous speed of the motor, which is achieved by applying the field excitation once the motor has
accelerated to a speed close to synchronous speed on the damper windings.
Motor starting methods

Several methods exist for starting motors. The most common methods are outlined here.
In addition, a discussion of motor-starting and control devices is given.
A.) Motor starting devices

The most common motor starting device is the low voltage motor-starting contactor. A contactor is defined in [4]
as a two-state ON-OFF device for repeatedly establishing and interrupting an electric power circuit. Contactors
are designed for optimum performance and lifetime when switching loads; they are not designed for interrupting
short-circuit currents and therefore motor circuits require separate short-circuit protection. Because contactors are
closed magnetically via their control coils, the use of contactors is typically referred to as magnetic control. For
small motors, typically fractional-horsepower, manual control switches are also available. Motor starting contactors
and switches in the United States are typically designed and manufactured per NEMA ICS-1 [4], NEMA ICS-2 [5],
and UL 508.
A controller is defined by [6] as a device or group of devices that serves to govern, in some pre-determined
manner, the electric power delivered to the apparatus to which it is connected. Motor starting contactors are
available as integral units with externally-operable switching means, defined by [4] as a combination controller.
A starter is defined by [4] as a form of electric motor controller that includes the switching means necessary to
start and stop a motor in combination with suitable overload protection.; a combination starter, which includes
the motor switching contactor as well as overload protection (described further below) and an integral
disconnecting device, is a type of combination controller. Low Voltage manual and magnetic controllers are
classified by [5] as Class A, B, or V according to their interrupting medium and their ability to interrupt currents:
Class A: Class A controllers are AC air-break, vacuum break, or oil-immersed manual or magnetic controllers for
service on 600 V or less. They are capable of interrupting operating overloads but not short circuits or faults
beyond operating overloads.
5
Class B: Class B controllers are DC air-break manual or magnetic controllers for service on 600 V or less. They
are capable of interrupting operating overloads but not short circuits or faults beyond operating overloads.
Class V: Class V controllers are AC vacuum-break magnetic controllers for service on 1500 V or less, and
capable of interrupting operating overloads but not short circuit or faults beyond operating overloads.
Low voltage NEMA-rated contactors are designated in sizes 00 (smallest) through 9 (largest) for various duty
applications per [5]. Figure 8-5 shows a NEMA-rated low voltage contactor along with a manual motor starting
switch, a starter, and a combination starter.
a.)
b.)
c.)
d.)
Figure 8-5:
a.) Motor starting contactor,
b.) Manual motor starter,
c.) Motor starter with contactor and overload relay,
d.) Combination starter with magnetic-only circuit breaker, contactor, thermal overload relay and
pilot devices
Control of contactors using maintained-contact devices is referred to as two-wire control. Use of momentarycontact devices in the control of contactors is referred to as three-wire control. Three-wire control has the
advantage of allowing the contactor to open and remain open if the line voltage should fail; this arrangement is
typical to provide undervoltage protection for motors and prevent inadvertent re-energization after a power failure.
Two-wire and three-wire control are shown in figure 8-6.
Figure 8-6: Low voltage contactor control: (full-voltage non-reversing control shown):
a.) Contactor nomenclature,
b.) Two-wire control,
c.) Three wire control
Medium voltage contactors are typically use vacuum as the interrupting means. Unlike a circuit breaker, a medium
voltage vacuum contactor is specifically designed for long life in load-interrupting duty rather than for short-circuit
interrupting duty. However, unlike their low voltage counterparts a medium voltage contactor may be able to
interrupt short-circuit currents beyond operating overloads.
Medium voltage air-break, vacuum, or oil-immersed controllers are classified by [7] as class E. Class E controllers
are further divided into class E1 and E2 as follows:
Class E1: Class E1 controllers employ their contacts for both starting and stopping the motor and interrupting
short circuits or faults exceeding operating overloads.
Class E2: Class E2 controllers employ their contacts for starting and stopping the motor and employ fuses for
short circuits or faults exceeding operating overloads.
Above 7200 V, motor control is generally accomplished using circuit breakers.
B.) Across-the-Line Starting

The most common method for starting an induction motor is across-the-line starting, where the motor is started
with full voltage applied to the stator windings. Across-the-line starting uses contactors to energize the motor at full
line voltage. The motor acceleration will be as described above, and is dependent upon the line voltage, motor
output torque, and load torque characteristics. Across the line-starting is also known as full-voltage starting.
Across-the-line starting at 600 V or less employs a single low voltage contactor, connected as shown in figure 8-7.
Note that the short-circuit protection and disconnecting devices are not shown.
Figure 8-7: Low voltage across-the-Line starting implementation
Another form of across-the-line starting is full voltage reversing starting, in which the motor may be made to turn
in either direction. This arrangement utilizes a full voltage reversing contactor with six poles, interlocked so that
only one set of contacts may be closed at a given time. The contacts are connected so that in the reverse
direction the motor has two phases swapped, forcing it to run in the opposite direction.
Across-the-line starting is the least expensive method, but it has the disadvantage that the full locked-rotor current
will be drawn during starting. This can cause voltage sags. Also, since the motor acceleration is dependent only
upon the motor output torque and load torque characteristics (along with the line voltage level), the acceleration is
not as smooth as can be attained with other starting methods.
C.) Reduced-voltage autotransformer starting

Reduced-voltage autotransformer starting consists of initially starting the motor with an autotransformer, then
removing the autotransformer from the circuit as the motor accelerates. This method results in a lower inrush
current as the motor starts, but also results in less available output torque when the autotransformer is in the
circuit. Autotransformer windings typically are tapped at 80, 65, and 50% voltage levels; the available output
torque is related to the output torque when at full voltage by the equation:
(8-6)
where
TRV
TFV
is the motor output torque at reduced voltage when the autotransformer is in the circuit
is the motor output torque with full voltage applied
Therefore, the motor output torque at the 80% autotransformer tap is 64% of the full-voltage torque value, at 65%
tap the torque is 42.25% of the full-voltage torque, and at 50% tap the torque is 25% of the full-voltage value.
Care must be taken to insure that the motor can be started at the tap value selected. Also, the thermal duty
capabilities of the autotransformer per NEMA ICS-9 must be taken into account; these will generally limit the
lowest tap to which the motor may be connected without damage to the autotransformer during starting.
A typical low voltage implementation of the reduced-voltage autotransformer starter is shown in figure 8-8.
Figure 8-8: Low voltage implementation of a reduced-voltage autotransformer starter
In figure 8-8 there are three contactors, labeled R, 1S, and 2S. The control scheme is designed so that the first
contactor to close is 1S, connecting the two autotransformers in open-delta. Once contactor 1S is closed,
contactor 2S closes, connecting the motor to the output of the autotransformer, in this case set to 50%. After a
pre-set time delay or current transition level, contactor 1S opens, leaving the motor energized through the noncommon autotransformer windings. Once contactor 1S is open, contactor R closes, energizing the motor at full
voltage. This is a closed-transition scheme; open transition schemes exist also.
D.) Reduced-voltage reactor or resistor starting

Reduced-voltage reactor or resistor starting consists of adding a resistor or reactor in series with the motor at the
beginning of the starting cycle, then shorting the resistor or reactor as the motor accelerates. This is generally a
less expensive method than the reduced-voltage autotransformer method, but suffers the same limitations due to
the reactor or resistor thermal limits.
E.) Wye-delta starting

Motors that have windings in which both ends of each stator windings are brought to terminals and are
configurable in either wye or delta candidates for wye-delta starting. Wye-delta starting starts the motor in a wye
configuration, which supplies 57.7% of the line-to-line voltage to each winding. During the starting process the
motor is connected in delta, supplying 100% of the line-to-line voltage to each winding. Both normally-open and
normally-closed transition schemes are available. This starting method typically uses three contactors.
F.) Part-winding starting

Motors which have stator windings in two parts with at least six terminal leads may be started with part-winding
starting. Part-winding starting energizes part of the transformer windings, typically 1/2 or 2/3 of the entire winding
per phase, to allow a lower inrush and smoother acceleration. This scheme typically uses two contactors and is a
closed-transition scheme. Separate overload relays are provided for each part of each winding.
G.) Solid-state soft-starting

Solid-state soft-starting ramps the voltage at the motor terminals linearly, producing a smooth acceleration.
Recent innovations include motor output torque-control models which linearly ramp the motor output torque, which
can result in even smoother, almost linear, acceleration.
Central to the operation of a solid-state soft-starter is the silicon-controlled rectifier, or SCR (also known as a
thyristor). An SCR is a device which conducts current in one direction when current is injected into its gate
terminal, and blocks current in the other direction. The circuit symbology and nomenclature for an SCR, including
the direction of current flows, is given in figure 8-10. In figure 8-10, if ig flows in the direction shown and Vak has
the polarity shown, the SCR current i will flow in the indicated direction. If the gate current becomes zero, the SCR
will turn off; this is known as natural commutation. To stop the flow of current in the absence of a natural current
zero, the SCR must be reversed-biased by applying Vak with a polarity the opposite to that shown in the figure.
This is referred to as forced commutation. For AC circuits, SCRs are employed in back-to-back pairs. Reference
[8] contains much background material on SCR operating theory and application.
A typical low voltage implementation of a solid-state soft-start controller is shown in figure 8-9. In figure 8-9 the inline contactor IC closes first. The firing circuit then causes the SCRs to vary the motor voltage as required by the
starting parameters. Once the motor has accelerated to full speed, the shorting contactor SC closes, bypassing
the SCRs and connecting the motor directly to full voltage.
Figure 8-9: Circuit symbology and nomenclature for an SCR
Figure 8-10: Solid-state soft-starter, low voltage implementation
The soft-start controller can also decelerate the motor in the same manner using the SCRs.
Because the SCRs dissipate heat, the equipment heat dissipation and the ambient temperature are concerns
when applying soft-start controllers and must be considered carefully.
Dedicated motor power factor correction capacitors must be switched out of the circuit during starting when
adjustable-speed drives are employed, due to the harmonic voltage interactions that could cause them to fail.
Surge capacitors should not be used at motors which are soft-started for the same reason.
Because soft-starters are microprocessor-based devices, they are typically supplied with communications and
internal diagnostic capabilities, making them a truly cutting-edge motor starting (and stopping) solution.
9
H.) Rotor resistance starting

Applicable only to wound-rotor motors, this starting method employs an adjustable external resistance which is
connected to the rotor via brushes and slip rings. The resistance in the rotor circuit dramatically alters the speedtorque characteristics of the motor during starting. The resistance in the rotor circuit is generally adjusted to be
highest during starting and is gradually lowered throughout the starting process.
A variation on this method is the use of solid-state circuitry to switch the rotor current and vary the effective value
of the external rotor circuit resistance.
These methods offer running speed control also. However they have been supplanted in recent years by
adjustable-speed drives where speed control is required and by soft-starters where speed control is not required.
I.) Adjustable-speed drive starting

Adjustable-Speed Drives, discussed in more detail below, have the benefit of providing soft-starting for a motor,
with starting advantages similar to the soft-starter described above. Unless speed control of the motor is required,
the soft-starter is a more economical solution for starting. If speed control is required, however, an adjustablespeed drive is among the best solutions available.
Because an adjustable-speed drive is not bypassed after starting, unlike a soft-starter, the harmonic currents it
causes to flow can affect the system power quality on a continuous basis. Also, power factor correction capacitors
must not be used at the drive output to the motor.
J.) Medium voltage starting method implementations

All of the starting methods mentioned above are generally available at the medium voltage level. However, the
disadvantages attributed to a given starting method are often exponentially more so when that method is applied
at the medium voltage level.
Because medium voltage contactors generally employ vacuum technology, they are more expensive than
low voltage contactors. They are also larger, and the vacuum interruption technology has a tendency to
produce larger transients than air-break technology. This can create issues in such starting methods as the
reduced-voltage autotransformer, which at the medium voltage level typically employs a three-phase
autotransformer with three windings connected in open-wye rather than two in open-delta as shown for the
low voltage implementation in figure 8-8. The voltage transients developed during starting force the use of surge
arrestors to protect the autotransformer when the motor voltage is switched from reduced-voltage to full voltage.
K.) Which starting method to use?

The selection of a motor starting method will be dictated both by the requirements of the driven machinery and the
requirements of the electrical system. Across-the-line starting will be sufficient in a great number of situations.
However, in some cases one of the starting methods discussed above must be employed. Table 8-2 gives a list of
the general advantages and disadvantages for each of the starting methods discussed above.
Motor speed control methods

It is often desirable to control the motor speed, usually for reasons process control for such variables as flow or
pressure. Such applications as fans and pumps often have varying output requirements, and control of the motor
speed is more efficient than mechanically limiting the process output with such devices as throttling valves or
dampers. The reason for this is due to the fact that for centrifugally-based processes (such as fans and
centrifugally-based pumps), the following relationships exist [1]:
(8-7)
(8-8)
10
So, for these types of processes the torque required to turn them is proportional to the square of the speed. But,
the power required to turn them is proportional to the cube of the speed, and this is what makes motor speed
control economically attractive [3]. To further this argument, consider the energy wasted when mechanical means
such as the throttling valves or dampers are used to control a process which is being driven from a motor running
at full speed. It is clear that motor speed control can be used to save energy by reducing wasted energy used to
mechanically control the process.
A.) Adjustable-speed drives

By far the most commonly-used AC motor control method is the use adjustable-speed drives. In most commercial
and industrial environments these have supplanted virtually every other motor speed control method.
An adjustable-speed drive works on the principle of varying the frequency to vary the speed of the motor. Recall
that from eq. (8-1) the synchronous speed of a motor is a function of both the system frequency and the number
of poles of the motor. By varying the frequency, the motor speed may be varied so long as the motor is equipped
to dissipate the heat at reduced speeds. Unlike soft-starting, specialized definite-purpose inverter-rated motor
designs are preferred since reduced-speed operation can cause thermal issues and
Table 8-2: Motor starting methods summary
Method
Advantages
Disadvantages
Across-the -Line
Simple
High Current Inrush
Cost-Effective
High Starting Torque

Abrupt Start
Reduced-voltage
autotransformer
High output torque vs. starting current
Limited duty cycle
Reduced-Voltage
Resistor or
Reactor
High output torque vs. starting current
Wye-Delta
Relatively low inrush current
Relatively low output torque vs. starting current
Relatively simple starter construction
Limited flexibility in starting characteristics
Good for long acceleration times
Requires special motor construction
Relatively Simple starter construction
Relatively low output torque vs. starting current
Some Flexibility in starting characteristics due adjustable Large equipment size due to autotransformers
taps on autotransformers
Limited duty cycle
Limited flexibility in starting characteristics
Higher inrush current than with reduced-voltage
autotransformer
Large equipment size due to resistors/reactors
Part-Winding
Not suitable for frequent starts

Requires special motor construction
Solid-state
soft starter
Smooth Acceleration
Relatively Expensive
Low inrush current
Sensitive to power quality
High flexibility in starting characteristics
Heat dissipation and ambient temperature are a concern
Typically offers deceleration control also

Typically integrates with industrial automation
infrastructure
Rotor Resistance
Smooth acceleration available
Complicated controller design
Good flexibility in starting characteristics
Requires expensive wound-rotor motor construction
Can be used for speed control also

Adjustable
Speed Drive
Smooth Acceleration
Cost-prohibitive unless speed control is required also
Low inrush current
Sensitive to power quality
High flexibility in starting characteristics
Heat dissipation and ambient temperature are a concern
Offers deceleration and speed control also
Continuous harmonic currents can create power quality

issues
Typically integrates with industrial automation

infrastructure
11
overspeed operation can result in safety issues. Further, pulse-width modulated (PWM) drive outputs can cause
repetitive voltage overshoots referred to as ringing, which can reduce the life expectancy of a general-purpose
motor. As per [3], the motor manufacturer should be consulted before applying a general-purpose motor in an
adjustable-speed drive application.
Various designs exist for adjustable-speed drives, however for low voltage drives the most prevalent is the
voltage-source pulse-width modulated design. As its name implies, the output is pulse-width modulated to reduce
the output harmonic and noise content. The AC input to the drive is typically a diode rectifier. A simplified circuit
topology for al voltage-source PWM drive is given in figure 8-11.
Figure 8-11: Voltage-source PWM adjustable-speed drive: simplified circuit topology for low voltage
implementation
The output stage for the circuit in figure 8-11 consists of Insulated-Gate Bipolar Transistors (IGBTs), which are
commonly used in low voltage PWM adjustable-speed drives instead of SCRs due to their superior switching rate
capability.
Adjustable speed drives offer superior speed control for motors through 10,000hp, depending upon the system
voltage [1]. They usually incorporate protection for the motor as well, allowing the omission of separate motor
protective relays if desired. Due to the high switching frequencies involved and their interaction with the cable
capacitance, the length of the cable runs between the output of the drive and the motor are limited, and, as
mentioned above for soft-starters, power factor correction capacitors and surge capacitors should not be used at
the output of an adjustable speed drive. Also due to the high switching frequencies, common-mode noise on the
grounding conductors can be an issue when these drives are employed.
On the incoming line, adjustable speed drives produce harmonics which must be taken into account in the over-all
system design. This topic is addressed in a later section of this guide.
Adjustable speed drives, like soft-starters, are microprocessor-based devices. Therefore, they can interface with
the automation infrastructure of a facility.
With the exception of a few isolated cases, for most industrial and commercial facilities adjustable speed drives
are the speed control of choice for AC motors.
B.) Older methods

Various other methods exist for AC motor speed control. A few of these are:
I
Rotor-resistance speed control similar to rotor-resistance starting, this method consists of varying the effective
resistance in the rotor of a wound-rotor induction motor to vary the speed. Variants of this method include rotor
power recovery systems using a second machine or an auxiliary solid-state rectifier and converter.
Multi-speed motor This type of motor is typically a squirrel-cage motor which has up to four fixed speeds.
Primary voltage adjustment using saturable reactors This method is only applicable to NEMA Design D motors
and offers a very narrow range of speed control.
Because of the limitations of these methods and the fact that they do not fit a wide range of motors, the adjustable
speed drive is typically the solution of choice for most commercial and industrial facilities.
12
Motor stopping devices

Several methods for motor stopping exist. Which, if any, is to be used is dependent upon the application.
A.) Dynamic braking

On form of dynamic braking involves the disconnection of AC power an induction motor and connecting DC power
to one stator phase. The kinetic energy of the motor and load is dissipated in the rotor circuit resistance.
An alternative method, which is frequently used in adjustable-speed drives, allows the motor to supply energy
back to the drive, where it is dissipated via a braking resistor.
B.) Plugging
Plugging is the reversal of the phase sequence on an induction motor via switching two phase connections to the
motor, which will cause the motor to come to a very rapid stop due to torque developed on the rotor in the
opposite direction from the current running direction. A zero-speed switch should be used to prevent reversal of
the motor.
C.) Mechanical braking

Unlike the methods mentioned above, mechanical brakes can hold a motor at standstill after power is removed.
Various forms of mechanical braking, such as DC solenoid, AC solenoid, AC torque-motor, and AC thrustor type,
are available. These are typically spring-set and electrically released, allowing them to be fail-safe in the event of
an electrical power failure.
D.) Adjustable speed drive and soft-start controller deceleration

For those applications not require fast deceleration times but a controlled deceleration is required, the speed
ramping capabilities inherent in most adjustable speed drives and soft-start controllers may be used. For faster
stopping with an adjustable-speed drive dynamic braking is required.
Motor protection
Motor protection involves protection of a motor from abnormal conditions. The most common abnormal condition
is an overload, which can produce damaging heating effects in the motor. For this reason, overload relays are the
primary means of motor protection. However, short-circuit protection is also required to minimize damage to the
motor from an internal short-circuit. Other protective devices are also available, their use depending upon the size
of the motor and the cost of protection vs. the cost of the motor.
A.) Nameplate values

While these are not the only values marked on a motor nameplate, the following are the nameplate values most
important to motor protection:
Rated Volts: The rated voltage of the motor
Rated Full Load current (FLA): The rated full-load current of the motor when running at full output capacity.
Time Rating: 5, 15, or 30 minutes, or continuous.
Rated Horsepower or KW: This is the output rating of the motor, not the input rating
Code Letter or Locked-Rotor Current (LRA): The locked-rotor current is the current that will be drawn by the
motor at zero speed. It is the initial current upon full-voltage energization of the motor. If given as a code letter, the
code letter may be used to determine the locked-rotor kVA per hp of the motor from NEC Table 430.7(B) [6].
Service Factor (SF): This is the factor by which the rated hp or kW may be multiplied to determine the maximum
continuous output of the motor without exceeding a defined temperature rise in the motor.
13
B.) Low voltage motor protection

Low voltage motor protection typically involves overload and short-circuit protection.
Overload protection is the protection from the thermal effects of overloads. As mentioned above, the motor input
current is always larger than would normally be dictated by the output power due to losses and the motor power
factor. NEC Article 430 [6] gives typical full-load currents where a machines actual full-load current is not known.
However, for overload protection purposes the motor nameplate full-load current rating must be used ([6] Article
430.32 (A) (1)). The NEC basic requirement for overload protection is 125% of the nameplate rating for motors
with service factors of 1.15 or greater and with a marked temperature rise of 40 C or less, and 115% for all others.
This NEC requirement takes into account the maximum long-time setting of the overload relay, but for low voltage
motors fine-tuning of the relay selection should be made according to the motor manufacturers recommendations.
Typically, overload relays for low voltage motors are classified as melting alloy, bimetallic, or solid-state. In
general, melting alloy relays are hand-reset devices, whereas bimetallic relays can be self-resetting or handresetting. Bimetallic relays are available as temperature-compensated or non-compensated; non-compensated is
an advantage when the relay and motor are in the same ambient temperature since the relay opening time
changes with the temperature in a similar manner to the motor [2]. Temperature compensated relays are designed
for operation where the motor is at a constant ambient temperature but the relay is at a varying ambient
temperature. While melting-alloy and bimetallic overload relays must be selected to suit the motor, for a solid-state
relay the same physical relay may be used for several different types of motors, with the settings adjusted on the
relay to match the motor it is protecting. Some solid-state relay models also have the advantage of providing
phase-loss protection.
Note that NEMA ICS 2-2000 classifies motor overload relays into three classes, Class 10, 20, and 30, depending
upon the time delay to trip on locked-rotor current. NEMA Class 10 overload relays will trip in 10s at 6x the
overload rating of the relay, Class 20 will trip in 20s at 6x the overload rating, and Class 30 will trip in 30s at
6x the overload rating.
Short-circuit protection generally involves fuses or magnetic-only circuit breakers (also known as motor circuit
protectors). Short-circuit protection is constrained by NEC article 430-52 [6] gives limits for various types of
motor/protective device combinations. In general, however, the lowest rating that does not cause nuisance tripping
due to motor inrush will give the best protection.
In addition to protecting the motor, the short-circuit protection also protects the motor circuit conductors and the
contactor. Note that the motor circuit conductors, per NEC Article 430.22 (A), must be sized to have an ampacity
not less than the 125% of the motor full-load current as determined from the tables in Article 430, not from the
full-load current marked nameplate rating. The purpose of this is avoid undersized cables should the motor be
replaced in the future with a different make and model of the same hp rating, since the hp rating does not clearly
define the full-load current of the motor. Unlike conventional branch circuits, overcurrent protective devices in
motor branch circuits do not dictate the conductor sizes. For this reason, care must be taken to insure that the
motor branch circuit short-circuit protection protects the motor conductors for short-circuits
To show how the overload and short-circuit protective devices coordinate with the motor and motor cable damage
curves, consider a 480 V, 300 hp squirrel-cage induction motor with a full-load current nameplate rating of 355 A.
The NEC Table 430.250 [6] full-load current rating for sizing the motor branch-circuit conductors is 361A. From
NEC table 310-16 [6], (1) 500kcmil cable per phase, with an ampacity of 380 A, is selected to supply the motor.
A Class 10 melting-alloy overload relay, sized for the motor per the manufacturers recommendations, is selected
for overload protection. A magnetic-only circuit breaker, sized at 800 A, which is in accordance with the 800% of
motor full-load current per NEC Table 430.52 [6], is used for short-circuit protection. The motor switching devices
is a NEMA size 6 contactor. A one-line representation of this motor branch circuit is shown in figure 8-12.
14
Figure 8-12: Example low voltage motor protection branch circuit
100K
10K
10
1000
1K
100
The resulting time-current coordination is shown in figure 8-13. Note that in figure 8-13 the purple curve to the
right of the overload relay curve is the motor thermal damage curve, obtained from the motor manufacturer. If this
curve is not available the relay or motor manufacturers selection tables should be used for selection of the
overload relay. The thermal overload relay protects the motor from overloads, while at the same time not opening
the motor inrush current or full-load current, as denoted by the purple MOTOR current curve to its left. Note the
high-current region (with current equal to the motor locked-rotor current) with an acceleration time of
approximately 9 seconds; this curve is dependent upon the connected load and must also obtained from the
motor manufacturer unless the motor acceleration time can be determined. In many cases the motor starting
current curve will not be available; for most cases a Class 10 overload relay will clear the locked-rotor current,
with Class 20 relays applied for higher service-factor motors such as NEMA design T-frame motors and Class 30
applied for high-inertial loads [2]. The magnetic-only circuit breaker protects the cable for short-circuits, as
denoted by the red CABLE short-circuit characteristic to the right of the circuit breaker characteristic. Finally,
while it is not shown on the curve the contactor can break up to 10 x its motor FLA rating, or 5400A, up to 10
times without servicing per [5] which is more than the maximum trip current of the circuit breaker; the contactor is
therefore adequately protected. The motor and its branch circuit is, therefore, adequately protected.
1000
CB
100
100
O.L.
10
CABLE
10
MOTOR
0.10
T IME IN S E C ONDS
0.10
100K
10K
1K
0.01
10
100
0.01
Figure 8-13: Time-current coordination for circuit of figure 8-12
For larger motors on solidly-grounded systems low voltage ground-fault protective devices may also be required to
allow coordination with upstream ground-fault protective devices. The application of these falls under the same
guidelines as given in System protection scetion (section 7 in this guide).
15
In addition to the overload protection described above, thermostats are commonly installed in three-phase
industrial-service 460 V motors from 11 kW through 150 kW (14-200hp) [2]. These are bimetallic devices that
operate at one fixed temperature and serve to de-energize the motor if the temperature setpoint is exceeded.
Low voltage motors are also occasionally provided with undervoltage relays, either to trip or prevent energization
when an undervoltage condition exists.
C.) Medium voltage motor protection

The protection of medium voltage motors is typically more complex than for their low voltage counterparts.
Multi-function microprocessor-based relays are typically used, which provide overload and overcurrent protection
as well as a host of other protection features which protect the motor from other abnormal conditions. R-rated
fuses are typically used for short-circuit protection.
Some of the additional protective elements utilized for medium voltage motors include:
I
RTDs Resistance Temperature Detectors are typically made of platinum, nickel, or copper and exhibit an
increasing resistance with increasing temperature. The RTD resistance is used to monitor the temperature at
various points in the motor, typically in the stator windings. The temperature is used to provide precise overload
protection for the motor. Per [2], RTDs should be specified for all motors 370 kW (500 hp) and above.
Negative-Sequence Overcurrent (Device 46) This is used to protect against damaging negative-sequence
currents, which can be caused by unbalanced voltages.
Phase sequence (Device 47) This is used to prevent the single-phasing of three-phase motors, which can
cause thermal damage if not detected.
Differential (Device 87) This is used to provide sensitive, high-speed protection for motor internal faults.
Typically only larger motors are provided with differential protection. In addition to traditional differential
protection, motors can also be equipped with self-balancing differential protection in which only one CT is used
for each phase, with both ends of each winding passing through that phases CT. Both are shown in figure 8-14.
Note that both ends of each stator winding must be brought to terminals to utilize differential protection.
Traditional differential protection may utilize either percentage differential (preferred) or high-impedance
differential relays. Self-balancing differential protection typically utilizes a standard overcurrent relay element.
Ground Fault Protection (Device 50G): Almost all medium voltage motors on solidly-grounded or low-resistancegrounded systems are provided with ground-fault protection. This is accomplished with a zero-sequence CT and
is almost always instantaneous.
87M
87M
87M
87M
87M
87M
a.)
Figure 8-14: Motor differential protection:
a.) Traditional
b.) Self-balancing
16
b.)
Typical overload and short-circuit protection for a medium voltage motor may be illustrated by considering the
following: A 750 hp, 4160 V motor is to be protected. The motor has a nameplate full-load current value of 96 A
and a locked-rotor current of 576 A, and a service factor of 1.15. A microprocessor-based motor protection relay is
to be utilized. The motor is to be provided with R-rated fuses for short-circuit protection. NEC Article 430.224 [6]
states that for motors over 600V the conductors shall have an ampacity no less than that at which the motor
overload protective device(s) are to trip. The pickup value for the overload protection is to be set equal to the
service factor times the nameplate full-load current, which is 110.4 A; the cables are copper in underground
conduit, therefore the cable size selected is #2AWG, with an ampacity of 145A per NEC table 310.77 [6]. The CT
primary ratings for the motor protection relay are typically selected as no less than 1.5 times the motor full load
current to avoid saturation (must be checked!) in this case 200:5. To coordinate with the overload protection and
protect the motor branch circuit cables and motor, a 6R fuse is chosen (note that per NEC Article 430.225 [6] the
motor overcurrent protection must be coordinated to automatically interrupt overload and fault currents in the
motor, but there is not specific constraint given for the short-circuit protection, unlike the requirements for motors
under 600 V per above). The motor switching device is a vacuum contactor rated 5.5 kV with an interrupting
rating of 5000A. A one-line representation of the motor branch circuit is shown in figure 8-15, excluding
ground-fault protection.
Figure 8-15: Example medium voltage motor circuit, excluding ground-fault protection
The resulting time-current coordination for this circuit is shown in figure 8-16. Note that the purple MV MOTOR
load current curve is to the left and below the green overload relay characteristic, therefore the motor inrush and
full-load current does not trip the overload relay. Unlike the case for a low voltage motor, this is typically available
from the manufacturer, who has analyzed the motors performance when connected to the driven load. Note also
that the purple motor thermal damage curve is to the left and above the relay overload curve, indicating the motor
is protected for overloads. The same applies for the red MV CABLE rated full-load current marker at the top of
the plot. The motor thermal damage curve is obtained from the motor manufacturer; if the entire curve is not
available, the motor hot safe-stall time provides one point on the curve. The red MV CABLE short-circuit
damage curve is to the right and above the blue MV FUSE characteristic, therefore the cable is protected for
short-circuits by the fuses. Finally, note that the fuse total clearing and overload relay curves cross at
approximately 900 A; this is well above the inrush of 576 A, but well below the contactor rating of 5000 A. The
fuse will therefore clear faults above the contactors 5000 A rating before the contactor opens.
NEC requirements for motors

The following are highlights from the NEC [6] requirements for motors. This is not intended to list all NEC
requirements for motors, but to illustrate the major points that apply in the most common motor installations and
affect the power system design. For the full text of the complete NEC requirements for motors, consult the NEC.
A.) General
The NEC basic requirements for motors, motor circuits, and controllers are given in NEC Article 430 [6], and are
supplemented by additional articles for specific motor-driven equipment. Article 430 is divided into 14 parts. The
requirements which apply to each part of a motor circuit are illustrated in figure 8-17.
17
100
100K
10K
1K
100
10
1000
1000
MV CABLE
RELAY
MV CABLE
100
MV FUSE
RELAY
10
MV MOTOR
10
10K
0.01
1K
0.10
100
0.10
T IME IN S E C ONDS
0.01
100K
10
Figure 8-16: Time-current coordination for circuit of figure 8-15
Figure 8-17: Illustrated NEC article 430 contents [6]
One of the main premises of Article 430 is the fact that the hp or kW rating of a motor is the output rating, as
discussed above. The motor electrical input characteristics will vary based upon the motor design. The
requirements in Article 430 are designed around this fact, as will be illustrated below.
Several definitions are given in Article 430 for terms unique or have meanings unique to that article:
Adjustable Speed Drive: A combination of the power converter, motor, and motor mounted auxiliary devices
such as encoders, tachometers, thermal switches, and detectors, air blowers, heaters, and vibration sensors.
18
Adjustable-Speed Drive System: An interconnected combination of equipment that provides a means of

adjusting the speed of a mechanical load coupled to a motor. A drive system typically consists of an adjustable
speed drive and auxiliary electrical apparatus.
Controller: For purposes of Article 430, this is any switch or device that is normally used to start and stop the
motor by making and breaking the motor circuit current.
Motor Control Circuit: The circuit of a control apparatus or system that carries the electric signals directing the
performance of the controller but does not carry the main power current.
System Isolation Equipment: A redundantly monitored, remotely operated contactor-isolating system, packaged
to provide the disconnection/isolation function, capable of verifiable operation from multiple remote locations by
means of lockout switches, each having the capability of being padlocked in the off position.
B.) Ampacity and motor rating determination (Article 430.6)

NEC motor ampacity table values, rather than motor nameplate full-load current values, must be used to
determine the motor ampacity for all purposes other than for overload protection, per Article 430.6. This does not
apply to low-speed (less than 1200 RPM) motors, motors built for high torques, multispeed motors, equipment
employing a shaded-pole or permanent-split capacitor type fan or blower motor and marked with the motor type,
or listed factory-wired motor-operated appliances marked with both hp and full-load current. The NEC motor
ampacity tables (Tables 430.248, 430.248, 430.29, and 430.250) are referenced in hp; for motors marked in
amperes, the horsepower must be determined by finding the hp corresponding with the nameplate ampacity, using
interpolation if necessary.
The motor nameplate full-load current rating must be used for determination of overload protection.
The basis of this requirement is the fact that the motor horsepower alone is not enough to define the input current
requirements of the motor, yet it is possible that a replacement for a given motor would be selected based only
upon horsepower.
C.) Motor circuit conductors [Article 430 Part II]

For a single motor used in a continuous duty application, Article 430.22 (A) dictates that the motor circuit
conductors be rated not less than 125% of the motors full-load current rating per the motor ampacity tables.
For a multispeed motor, the requirement is that the branch circuit conductors on the line side of the controller must
be based upon the highest of the full-load current ratings shown on the motor nameplate, with the branch-circuit
conductors between the controller and the motor based upon the current rating of the winding(s) that the
conductors energize. [430.22 (B)]
For a wye-start, delta-run connected motor, the selection of branch circuit conductors must be based upon the
motor full-load current. The conductors between the controller and the motor must be based upon 58% of the
motor full-load current. [430.22 (C)]
For a part-winding connected motor, the selection of branch-circuit conductors on the line side of the controller
must be based upon the motor full-load current. The selection of conductors between the controller and the motor
must be based upon 50% of the motor full-load current. [430.22 (D)]
For motors with other than continuous duty, the motor branch circuit conductors must have a rating not less than
as shown in table 430.22 (E), which gives percentages of the nameplate full-load current rating for various
classifications of service and motor duty ratings. [430.22(E)]
For continuous-duty motors with wound-rotor secondaries, conductors between the secondary (rotor) to the
controller must have an ampacity no less than 125% of the full-load secondary current of the motor. For other than
continuous duty motors, the percentages in table 430.22 (E) apply. If there is a resistor separate from the
controller, Table 430.23 (C), which is based upon the resistor duty classification, contains percentages of full-load
secondary current to which the conductors between the controller and the resistor must be compared. [Article
430.23).
19
Conductors that supply several motors or a motor(s) and other loads must have an ampacity not less than 125%
of the full-load current rating of the largest motor in the group plus the sum of all the full-load currents of all other
motors in the group, plus the ampacity required for other loads. Various exceptions apply, and the authority having
jurisdiction may grant permission for a lower ampacity, provided the conductors have the ampacity for the
maximum load determined in accordance with the sizes and number of motors supplied and the character of their
loads and duties. [430.24, 430.25]
Where a motor installation includes a capacitor connected on the load side of the motor overload device, the
effect of the capacitor must be disregarded in sizing the motor circuit conductor. [460.9, referenced in 430.27]
D.) Motor and branch-circuit overload protection (Part III)

Continuous-duty motors over 1hp must be protected against overload by means of a separate overload device
that is responsive to motor current or a thermal protector integral with the motor that will protect from dangerous
overheating and failure to start. Motors that are part of an approved assembly that does not subject the motor to
overloads may be protected by an integral device that protects the motor against failure to start. Motors larger
than 1500 hp must be protected by a protective device with imbedded temperature detectors that cause current to
the motor to be interrupted when the motor attains a temperature rise greater than marked on the motor
nameplate in an ambient temperature of 40C. [430.32]
The separate overload device per the above can be recognized to be an overload relay as discussed above.
The protective device having imbedded temperature detectors typically refers to RTDs and their associated
relay(s).
If overload relays are used, for motors with a marked service factor of 1.15 or greater or a marked temperature
rise of 40C or less may be set at a maximum of 125% of the motor nameplate full-load current rating. For all
other motors the maximum overload relay setting is 115% of the motor nameplate full-load current rating. If these
values do not allow the motor to start or carry the load, higher-size sensing elements or incremental settings may
be permitted to be used so long as they do not exceed 140% of the motor nameplate full-load current rating for
motors with a service factor of 1.15 or greater or a temperature rise of 40C or less, or 130% for all other motors.
Part-winding motors must have overload protection for each winding, set to half of these values.[430.32(A),
430.32(C), 430.4]
Overload requirements for motors 1hp or less vary depending upon whether the motor is automatically started or
non-automatically started. [430.32(B), 430.32 (D)]
Motors for intermittent and similar duty may be protected against overload by the branch-circuit and ground-fault
protective device. [430.33]
Motor overload devices for non-automatically started motors may be shunted or cut out of the circuit during
starting. With some exceptions, an automatically-started motor cannot have their overload devices shunted or cut
out of the circuit. [430.35]
A motor controller may be permitted to serve as overload protection. [430.38]. This allows a separate relay
which causes the motor contactor to be used, or the use of adjustable-speed drive built-in overload
protection capabilities.
A motor overload device that can restart a motor manually after overload tripping shall cannot installed unless
approved for use with the motor it protects, and must not be installed if it causes injury to persons [430.43].
This requirement is to insure that proper cooling time is given before a motor is automatically restarted.
If immediate automatic shutdown of a motor by a motor overload protective device(s) would introduce additional or
increased hazard(s) to a person(s) and continued motor operation is necessary for a safe shutdown of equipment
or process, the motor overload protection may be permitted to be connected to a supervised alarm rather than
causing an immediate shutdown. [430.44]
Part-winding motors must have overload protection for each winding, set to half of that specified by
430.52. [430.4]
Where a motor installation includes a capacitor connected on the load side of the motor overload device, the
rating or setting of the motor overload device must be based upon the improved power factor of the motor circuit.
[460.9, referenced in FPN to 430.32].
20
E.) Motor branch circuit short-circuit and ground fault protection (Part IV.)
Motor branch circuit short-circuit and ground fault protective devices must comply with Table 430.52, which gives
maximum ratings, in percentage of the motor full-load current, which can be used for various motor and protective
device types. If the value for the protective device rating does not correspond with a standard size for a fuses,
nonadjustable circuit breakers, thermal protective devices, or possible settings of adjustable circuit breakers, the
next higher standard size is permitted. If the value for the protective device rating is no sufficient for the starting
current of the motor, various exceptions apply depending upon the protective device type. [430.52 (C) (1)].
Where maximum branch-circuit short-circuit and ground-fault protective device ratings are shown in the
manufacturers overload relay table for use with a motor controller or are otherwise marked on equipment, they
must not be exceeded even if otherwise permitted by Table 430.52. [Table 430.52 (C) (2)]
Instantaneous-trip circuit breakers may only be used if adjustable and if part of a listed combination motor
controller having coordinated motor overload and short-circuit and ground-fault protection in each conductor.
Exceptions apply. [Table 430.52 (C) (3)]
For a multi-speed motor, a single short-circuit and ground-fault protective device is permitted for two or more
windings, so long as the rating of the protective device does not exceed the percentage per Table 430.52 of the
smallest winding protected. Exceptions apply. [430.52 (C) (4)]
So long as the replacement fuse size is marked adjacent to the fuses, suitable fuses are permitted in lieu of the
devices listed in Table 430.52 for power electronic devices in a solid state motor control system. [430.52 (C) (5)]
A listed self-protected combination controller is permitted in lieu of the devices specified in Table 430.52 so long
as the adjustable instantaneous trip settings do not exceed 1300% of full-load motor current for other than Design
B energy-efficient motors and 1700% of full-load current for Design B energy-efficient motors. The same applies
for a motor short-circuit protector, so long as it is part of a listed motor controller having coordinate motor overload
protection and short-circuit and ground-fault protection. [430.52 (C) (6), 430.52 (C) (7)]
Torque motors must be protected at the motor nameplate current rating in accordance with 240.4 (B). [430.52 (D)]
Two or more motors or one or more motors and other loads are permitted to be connected to the same
branch circuit if:
I
The motors are not over 1hp, the branch circuit is 120 V and protected at not over 20 A 600 V or less protected
at not over 15 A, the full-load rating of each motor does not exceed 6 A, the rating of the branch-circuit shortcircuit and ground-fault protective device marked on any of the controllers is not exceeded, and individual
overload protection confirms to 430.32, OR,
If the branch circuit short-circuit and ground-fault protective device is selected not to exceed the requirements of
430.52 for the smallest rated motor, each motor has individual overload protection, and it can be determined that
the branch-circuit short-circuit and ground-fault protective device will not open under the most severe normal
conditions of service that might be encountered, OR,
The motors are part of a group installation complying with 430.52 (C) and (D).
[430.53]
For multimotor and combination load equipment, the rating of the branch-circuit and ground-fault protective device
must not exceed the rating marked on the equipment. [430.54]
Motor branch circuit and ground-fault protection may be combined into a single protective device where the rating
or setting of the device provides the overload protection specified in 430.32. [430.55]
F.) Disconnecting means (Part IX.)

An individual disconnecting means must be provided for each controller. The disconnecting means must be in
sight from the controller location unless the circuit is over 600 V, in which case a controller disconnecting means
capable of being locked in the open position is permitted to be out of sight of the controller if it is marked with a
warning label giving the location of the disconnecting means. A single disconnecting means is permitted for a
21
group of coordinated controllers that drive several parts of a single machine or piece of apparatus. In this case the
disconnecting means must be located in sight from the controllers, and both the disconnecting means and the
controllers must be located in sight from the machine or apparatus. [430.102]
A motor disconnecting means must be located in sight from the motor location and the driven machinery location
unless the controller disconnecting means is individually capable of being locked in the open position and either
a.) such a location of the disconnecting means is impracticable or introduces additional or increased hazards to
persons or property, or b.) the motor is in an industrial installation where conditions of maintenance and
supervision ensure that only qualified persons service the equipment. The controller disconnecting means per
above may be permitted to serve as the disconnecting means for the motor if it is located in sight from the motor
location and the driven machinery location. [430.102]
The disconnecting means must open all ungrounded supply conductors and must be designed so that no pole can
be operated independently. The disconnecting means is permitted to be in the same enclosure with the controller.
The disconnecting means must clearly indicate whether it is in the open (off) or closed (on) position. The
disconnecting means may be a listed motor circuit switch rated in horsepower, a listed molded case circuit
breaker, a listed molded case switch, or an instantaneous trip circuit breaker that is part of a listed combination
controller. Listed manual motor controllers additionally marked as suitable as motor disconnect are permitted as
a disconnecting means where installed between the final motor branch-circuit short-circuit protective device and
the motor. [430.103, 430.104, 430.109 (A)]
System isolation equipment must be listed for disconnection purposes. Where system isolation equipment is used
it must be installed on the load side of the overcurrent protection and its disconnecting means. The disconnecting
means must be a listed motor-circuit switch rated in horsepower, a listed molded case circuit breaker, or a listed
molded-case switch. [430.109 (A) (7)]
Stationary motors of 1/8 hp or less may use the branch-circuit overcurrent device as the disconnecting means.
Stationary motors rated 2hp or less and 300V or less may use a general-purpose switch with an ampere rating not
less than twice the full-load current rating of the motor, a general-use AC snap switch for use only on AC, or a
listed manual motor controller with a hp rating not less than the motor hp and marked suitable as motor
disconnect as the motor disconnecting means. [430.109 (B), 430.109 (C)]
For stationary motors rated at more than 40hp up to and including 100hp, the disconnecting means is permitted to
be a general-use or isolating switch where plainly marked do not operate under load. [430.109(E)]
Cord-and-plug connected motors with a horsepower-rated attachment plug and receptacle having ratings no loess
than the motor ratings may use the attachment plug and receptacle and the disconnecting means. Cord-and-plug
connected appliances per 422.32, room air conditioners per 440.63, or a portable motor rated 1/3 hp or less do
not require the hp-rated attachment plug and receptacle. [430.109 (F)]
The ampere rating of the disconnecting means must not be less than 115% of the full load current rating of the
motor, unless it is rated in hp and has a hp rating not less than the hp of the motor. For torque motors the
disconnecting means must have a an ampere rating of at least 115% of the motor nameplate current. A method
for determining the required disconnect rating for combination loads is given in 430.110 (C). [430.110]
Each motor must be provided with its own disconnecting means, unless a number of motors drive several parts of
a single machine or piece of apparatus, a group of motors is under the protection one set of branch-circuit
protective devices as permitted by 430.53 (A), or where a group of motors is in a single room within sight from the
location of the disconnecting means. [430.112]
Where a motor or motor-operated equipment receive electrical energy from more than one source, each source
must be provided with a disconnecting means from each source of electrical energy immediately adjacent to the
equipment served. The disconnecting means for the main power supply to the motor is not required to be
immediately adjacent of the controller disconnecting means can be locked in the open position.
G.) Motor controllers and control circuits (Parts VI and VII)

Each controller must be capable of starting or stopping the motor it controls and must be capable of interrupting
the locked-rotor current of the motor. An autotransformer controller must provide an off position, a running
position, and at least one starting position, and designed so that it cannot rest in the starting position or in any
22
position that will render the overload device in the circuit inoperative. Motor starting rheostats must be designed
so that the contact arm cannot be left on intermediate segments. [430.82]
Stationary motors of 1/8 hp or less which are normally left running and is constructed so that it cannot be
damaged by overload or failure to start, the branch-circuit protective device is permitted to serve as the controller.
Portable motors rated 1/3 hp or less may have an attachment plug and receptacle serve as the controller. [430.81]
Controllers, other than inverse time circuit breakers and molded case switches, must have horsepower ratings at
the application voltage not lower than the horsepower rating of the motor. A branch circuit inverse time circuit
breaker or molded case switch is permitted as a controller for all motors. For stationary motors 2hp or less and
300 V or less, a general-use switch having an ampere rating not less than twice the full-load current rating of the
motor or an AC only snap switch where the motor full-load current rating is not more than 80% of the ampere
rating of the switch may serve as the controller. For torque motors, the controller must have a continuous-duty,
full-load current rating not less than the nameplate current rating of the motor. [430.83]
A controller with a straight voltage rating, for example 240 V or 480 V, is permitted to be applied in a circuit in
which the nominal voltage between any two conductors does not exceed the controllers voltage rating.
A controller with a slash rating, for example, 480 Y/277 V, may only be applied on a solidly-grounded circuit where
the nominal voltage to ground from any conductor does not exceed the lower of the two values of the controllers
voltage rating and the nominal voltage between any two conductors does not exceed the higher value of the
controllers voltage rating. [430.83 (E)]
A controller need not open all conductors to the motor, unless it also serves as a disconnecting means [430.84].
The controller must only open enough conductors as is necessary to stop the motor.
A controller is permitted to disconnect the grounded conductor, so long as the controller is designed so that the
pole which disconnects the grounded conductor cannot open without simultaneously opening all conductors of the
circuit. [430.85]
Each motor must have its own individual controller, unless a number of motors drive several parts of a single
machine or piece of apparatus, a group of motors is under the protection one overcurrent device as permitted by
430.53 (A), or where a group of motors is in a single room within sight from the location of the disconnecting
means. An air-break switch, inverse time circuit breaker, or oil switch may be permitted to serve as the controller
and disconnecting means if it complies with the requirements for controllers in 430.83, opens all ungrounded
conductors to the motor, and is protected by an overcurrent device in each ungrounded conductor. An
autotransformer type controller must be provided with a separate disconnecting means. Inverse-time circuit
breakers and oil switches are permitted to be both hand and manually operable. [430.111]
Motor control circuits must be provided with overcurrent protection in accordance with 430.72. [430.72]
Motor control circuits must be arranged so that they will be disconnected from all sources of supply when the
disconnecting means is in the open position. The disconnecting means may be two separate adjacent devices,
one to disconnect the motor circuit, the other to disconnect the control circuit. Various exceptions apply to the
requirement to the need for the two disconnecting means to be adjacent to each other. Control transformers in
controller enclosures must be connected to the load side of the disconnecting means for the motor control
circuit. [430.74]
Where damage to a motor control circuit would constitute a hazard, all conductors of such a remote motor control
circuit that are outside the control device itself must be installed in a raceway or otherwise suitably protected from
physical damage. Where one side of the motor control circuit is grounded, the motor control circuit must be
arranged so that an accidental ground in the control circuit remote from the motor controller will not start the motor
or bypass manually operated shutdown devices. [430.73]
H.) Adjustable-speed drive systems

Branch/feeder circuit conductors that supply power conversion equipment included as part of an adjustable-speed
drive system must have an ampacity not less than 125% of the rated input to the power conversion equipment.
For an adjustable speed drive system that utilizes a bypass device, the conductor ampacity must not be less than
required by 430.6 (see above). [430.122]
23
Where the power conversion equipment is marked to indicate that motor overload protection is included, additional
overload protection is not required. If a bypass circuit is utilized, motor overload protection as described in part III
(see above) must be provided in the bypass circuit. For multiple-motor applications individual motor overload
protection per part III is required. [430.124]
Adjustable speed drive systems must protect the motor against overtemperature conditions by means of a motor
thermal protector per 430.32, an adjustable speed drive controller with load and speed-sensitive overload
protection and thermal memory retention upon power loss, overtemperature protection relay utilizing thermal
sensors embedded in the motor and meeting the requirements of 430.32 (A)(2) or (B)(2), or a thermal sensor
embedded in the motor that is received and acted upon by an adjustable speed drive. Motors that utilize external
forced-air or liquid cooling systems must be provided with protection that will be continuously enabled or enabled
automatically if the cooling system fails. For multiple motor applications, individual motor overtemperature
protection must be provided. The provisions of 430.43 and 430.44 apply to motor overtemperature protection
means. [430.24]
The disconnecting means is permitted to be in the incoming line conversion equipment and must have a rating of
not less than 115% of the rated input current of the conversion unit. [430.128]
I.) Motor control centers (Part VIII.)

Motor control centers must be provided with overcurrent protection with parts I, II, and IX of article 240. The
ampere rating or setting of the overcurrent protective device must not exceed the rating of the common power
bus. This overcurrent protection may be provided by an overcurrent protective device located ahead of the motor
control center or a main overcurrent device located within the motor control center. [430.94]
J.) Motor feeder short-circuit and ground-fault protection (Part V.)

A feeder supplying a specific fixed motor load(s) and consisting of conductor sizes based upon 430.24 must be
provided with a protective device having a rating or setting not greater than the largest rating or setting of the
branch-circuit short-circuit and ground-fault protective device for any motor supplied by the feeder, plus the sum of
the full-load currents of the other motors of the group. The largest rating or setting of the branch-circuit shortcircuit and ground-fault protective device is based upon the maximum permitted size per Article 430.52. and Table
430.52. Where one or more instantaneous trip circuit breakers or motor short-circuit protectors are used for motor
branch-circuit and ground fault protection as permitted in 430.52(C), each instantaneous trip circuit breaker or
motor short-circuit protector must be assumed to have a rating not exceeding the maximum percentage of motor
full-load current permitted by Table 430.52 for the type of feeder protective device employed. Where the feeder
overcurrent protective device also provides overcurrent protection for a motor control center, the provisions of
430.94 apply. [430.62]
Where a feeder supplies a motor load and, in addition, a lighting or lighting and appliance load, the feeder
protective device must have a rating sufficient to carry the lighting and appliance load, plus the rating permitted by
430.52 for a single motor, the rating permitted by 440.22 for a single hermitic refrigerant motor-compressor, or the
rating permitted by 430.62 for two or more motors. [430.63]
K.) Over 600 V, nominal (Part XI.)

Certain requirements mentioned above are amended or added to above 600V, as follows:
Conductors supplying motors must have an ampacity not less than the current at which the motor overload
protective device(s) is selected to trip [430.224]
Each motor circuit must include coordinated protection to automatically interrupt overload and fault currents in the
motor, the motor circuit conductors, and the motor control apparatus. This may be a thermal protector integral to
the motor or external current-sensing devices, or both. The secondary circuits of wound-rotor AC motors are
considered to be protected against overcurrent by the motor overload protection means. Operation of the overload
interrupting device must simultaneously disconnect all ungrounded conductors. Automatic reset of overload
sensing devices is prohibited after trip unless resetting does not cause automatic restarting or there is no hazard
to persons due to automatic restarting. Where a motor is vital to operation of the plant and the motor should
24
operate to failure if necessary to prevent a greater hazard to persons, the sensing device(s) are permitted to be
connected to a supervised annunciator or alarm instead of interrupting the motor circuit [430.225]
Fault current protection must be provided by either a circuit breaker, arranged so that it can be serviced without
hazard, or fuses. A circuit breaker must open each ungrounded conductor. Fuses must be placed in each
ungrounded conductor and must be furnished with a disconnecting means (or be of the type that can serve as the
disconnecting means) and arranged so that they cannot be serviced while energized. Automatic reclosing of the
fault-current interrupting device is not permitted unless the circuit is exposed to transient faults and such
automatic reclosing does not create a hazard to persons. Overload and fault-current protection may be provided
by the same device. [430.225]
The ultimate trip current of overload relays or other motor-protective devices must not exceed 115% of the
controllers continuous current rating. Where the motor branch-circuit disconnecting means is separate from the
controller, the disconnecting means current rating must not be less than the ultimate trip setting of the overcurrent
relays in the circuit.
The controller disconnecting means must be capable of being locked in the open position.
L.) Other articles

Other NEC [6] articles which apply to motors and augment or amend the provisions in article 430 are given in
table 430.5. Chief among these are Article 440, Air-Conditioning and Refrigerating Equipment, Article 610
Cranes and Hoists, Article 620 Elevators, Dumbwaiters, Escalators, Moving Walks, Wheelchair Lifts and
Stairway Chair Lifts and Article 695 Fire Pumps.
References
[1]
Donald G. Fink, F. Wayne Beaty, Standard Handbook for Electrical Engineers, New York: McGraw-Hill, 2000.
[2]
IEEE Recommended Practice for Protection and Coordination of Industrial Power Systems, IEEE Std. 242-2001,
December 2001.
[3]
Safety Standard and Guide for Selection, Installation, and Use of Electric Motors and Generators, NEMA
Standards Publication MG 2-2001
[4]
Industrial Controls and System: General Requirements, NEMA Standards Publication ICS 1-2000.
[5]
Industrial Control and Systems: Controllers, Contactors, and Overload Relays Rated 600 Volts, NEMA
Standards Publication ICS 2-2000.
[6]
[7]
Industrial Control and Systems: Medium Voltage Controllers Rated 2001 to 7200V AC.
[8]
P.C. Sen, Principles of Electric Machines and Power Electronics, New York: John Wiley & Sons, 1989.
25
Section 9:
Power Distribution Equipment
Introduction
Power Distribution Equipment is a term generally used to describe any apparatus used for the generation,
transmission, distribution, or control of electrical energy. This section concentrates upon commonly-used power
distribution equipment: Panelboards, Switchboards, Low Voltage Motor Control Centers, Low Voltage Switchgear,
Medium Voltage Power and Distribution Transformers, Medium Voltage Metal Enclosed Switchgear, Medium
Voltage Motor Control Centers, and Medium Voltage Metal-Clad switchgear. Each has its own unique standards
and application guidelines, and one facet of good power system design is the knowledge of when to apply each
type of equipment and the limitations of each type of equipment. All of these equipment described herein are
typically custom-engineered on a per-order basis.
NEMA enclosure types

One common characteristic of all of the equipment types covered in this section is that they are all enclosed for
safety. The enclosures for enclosed equipment generally follow the guidelines set forth in NEMA 250-2003 [1],
and, although this standard is intended for equipment less than 1000 V, this is true of medium voltage power
equipment also.
The most common NEMA enclosure types are described as follows [1]:
Type 1: Enclosures constructed for indoor use to provide a degree of protection to personnel against access to
hazardous parts and to provide a degree of protection of the equipment inside the ingress of solid foreign objects.
Type 3R: Enclosures constructed for either indoor or outdoor use to provide a degree of protection to personnel
against access to hazardous parts; to provide a degree of protection of the equipment inside the enclosure
against ingress of solid foreign objects (falling dirt and windblown dust); to provide a degree of protection with
respect to harmful effects on the equipment due to the ingress of water (rain, sleet, snow); and that will be
undamaged by the external formation of ice on the enclosure.
Type 4: Enclosures constructed for either indoor or outdoor use to provide a degree of protection to personnel
against ingress of solid foreign objects (falling dirt and windblown dust); to provide a degree of protection with
respect to harmful effects on the equipment due to the ingress of water (rain, sleet, snow, splashing water, and
hose directed water); and that will be undamaged by the external formation of ice on the enclosure.
Type 4X: Enclosures constructed for either indoor or outdoor use to provide a degree of protection to personnel
against ingress of solid foreign objects (windblown dust); to provide a degree of protection with respect to harmful
effects on the equipment due to the ingress of water (rain, sleet, snow, splashing water, and hose directed water);
that provides an additional level of protection against corrosion; and that will be undamaged by the external
formation of ice on the enclosure.
Type 5: Enclosures constructed for indoor use to provide a degree of protection to personnel against access to
hazardous parts; to provide a degree of protection of the equipment inside the enclosure against the ingress of
solid foreign objects (falling dirt and settling airborne dust, lint, fibers, and flyings); and to provide a degree of
protection with respect of harmful effects on the equipment due to the ingress of water (dripping and light
splashing).
Type 12: Enclosures constructed (without knockouts) for indoor use to provide a degree protection to personnel
against ingress of solid foreign objects (falling dirt and circulating dust, lint, fibers, and flyings); and to provide a
degree of protection with respect to harmful effects on the equipment due to the ingress of water (dripping and
light splashing).
Panelboards
Table 9-1: Quick reference Panelboards
Available voltage ratings
120-600 V
Available current ratings
30-1200 A
Available short-circuit ratings
Through 200 kA
Major industry standards
UL 50, UL 67, CSA C22.2 No. 29, CSA C22.2 No. 94,
NEMA PB 1, Federal Specification W-P-115C, NEC
Typical enclosure types
1, 3R, 5, 12
Primary NEC requirements
Article 408
Panelboards are the most common type of power distribution equipment. A panelboard is defined as a single
panel or group of panel units designed for assembly in the form of a single panel, including buses and automatic
overcurrent devices, and equipped with or without switches for the control of light, heat, or power circuits;
designed to be placed in a cabinet or cutout box placed in or against a wall, partition, or other support; and
accessible only from the front [2]. It typically consists of low voltage molded-case circuit breakers arranged
with connections to a common bus, with or without a main circuit breaker. Figure 9-1 shows typical examples
of panelboards.
Figure 9-1: Panelboards
Panelboards are used to group the overcurrent protection devices for several circuits together into a single piece
of equipment. In small installations they may serve as the service equipment. The NEC [2] divides panelboards
into two categories:
Lighting and Appliance Branch-Circuit Panelboard: A panelboard having more than 10 percent of its
overcurrent devices protecting lighting and appliance branch circuits.
Power Panelboard: A panelboard having 10 percent or fewer of its overcurrent devices protecting lighting and
appliance branch circuits.
Lighting and appliance branch-circuit panelboards are limited to a maximum of 42 overcurrent devices, excluding
mains. UL 67 [3] designates Class CTL Panelboard as the marking for appliance and branch circuit panelboards;
CTL stands for circuit limiting. In some manufacturers literature lighting and appliance branch-circuit
panelboards for residential or light commercial use are referred to as loadcenters.
Panelboards are available with built-in main devices or as main lugs only (MLO). The NEC [2] requires appliance
and branch circuit panelboards to be individually protected on the supply side by not more than two main circuit
breakers or two sets of fuses having a combined rating no greater than the rating of the panelboard. Lighting and
appliance branch circuit panelboards are not required to have individual protection if the feeder overcurrent device
is no greater than the rating of the panelboard. Power panelboards must be protected by an overcurrent device
with a rating not greater than that of the panelboard [2].
Various methods for attaching the circuit breakers to the panelboard bus are available, such as plug-on, bolt on,
etc. The circuit breakers are typically purchased separately. Often, the enclosure, interior, and trim assemblies for
the panelboard itself are purchased separately as well. This is typically true of larger panelboards and gives a
great deal of flexibility with regard to use of the same interior with different enclosures and trims.
Panelboards are available with a number of accessories. Subfeed lugs allow taps directly from the panelboard
bus without the need for overcurrent devices. Circuit breaker locking devices allow locking of circuit breakers in
the open or closed position (note that the breakers will still trip on an overcurrent condition). Various types of trims
are available, with various locking means available for trims that are equipped with doors.
Switchboards
Table 9-2: Quick reference Switchboards
120-600 V
800-5000 A
Through 200 kA
UL 891, NEMA PB 1, NEC
1, 3R
Article 408
The definition of a switchboard is a large single panel, frame, or assembly of panels on which are mounted on
the face, back, or both, switches, overcurrent and other protective devices, buses, and usually instruments [2].
Switchboards are free-standing equipment, unlike panelboards, and are generally accessible from the rear as well
as from the front. They may consist of multiple sections, connected by a common through-bus. Unlike
panelboards, the number of overcurrent devices in a switchboard is not limited.
Switchboards generally house molded case circuit breakers or fused switches. They are generally the next level
upstream from panelboards in the electrical system, and in some small to medium-size electrical systems they
serve as the service equipment. Figure 9-2 shows an example of a switchboard.
Figure 9-2: Switchboards
Switchboards are available with a main circuit breaker or fusible switch, or as main-lugs only. The available
ampacities and multi-section availability makes them more flexible than panelboards. They are generally available
utilizing either copper or aluminum bussing, and with a variety of bus plating options. Custom bussing for retrofit
applications is also possible.
Switchboard circuit breakers may be stationary-mounted (also referred to as fixed-mounted), where they can be
removed only by unbolting of electrical connections and mounting supports, or drawout-mounted, where they can
be without the necessity of removing connections or mounting supports. It is possible to insert and remove
drawout devices with the main bus energized. The section which contains the main circuit breaker(s) or service
disconnect devices is referred to as a main section. A section containing branch or feeder circuit breakers is
referred to as a distribution section.
Devices mounted in the switchboard may be either panel mounted (also referred to as group mounted), where
they are mounted on a common base or mounting surface, or individually mounted, where they do not share a
common base or mounting surface. Individually mounted devices may or may not be in their own compartments.
A device which is segregated from other devices by metal or insulating barriers and which is not readily accessible
to personnel unless special means for access are used is referred to an isolated device. Figure 9-3 shows
examples of sections with group-mounted individually-mounted device.
The main through-bus is often referred to as the horizontal bus. The bussing in a section which connects to the
through-bus is referred to as the section bus (also known as vertical bus). The bussing that connects the section
bus to the overcurrent devices is referred to as the branch bus. Section and branch busses may be smaller than
the main through-bus; if this is the case UL 891 [2] gives the required section bus size as a function of the number
of overcurrent devices connected to it.
a.)
b.)
Figure 9-3:
a.) Group-mounted devices
b.) Individually-mounted devices
Switchboards are available with a number of accessories, including custom-engineered options such as utility
metering compartments, automatic transfer schemes, and modified-differential ground fault for switchboards with
multiple mains. However, the internal barriering requirements are minimal.
Low voltage motor control centers

Table 9-3: Quick reference Low voltage motor control centers
120-600 V
600-2500 A
Through 100 kA
NEMA ICS-18, UL 845, NEC
1, 3R, 12
Article 430
A motor control center (MCC) is defined as a floor-mounted assembly of one or more enclosed vertical sections
typically having a common power bus and typically containing combination motor control units [5].
Motor control centers are used to group a number of combination motor controllers together at a given location
with a common power bus. Figure 9-4 shows an example of a motor control center.
Figure 9-4: Low voltage motor control center
MCCs are classified into two classes by [5] and [6]:

Class I Motor Control Centers: Mechanical groupings of combination motor control units, feeder tap units,
other units, and electrical devices arranged in an assembly.
Class II Motor Control Centers: A Class I motor control center provided with manufacturer-furnished
electrical interlocking and wiring between units, as specifically described in overall control system diagrams
supplied by the user.
MCC wiring is classified by [5] and [6] into three types:
Type A Wiring: User (field) load and control wiring are connected directly to device terminals internal to the unit;
provided on Class I MCCs only.
Type B Wiring: User (field ) control wiring is connected to unit terminal blocks; the field load wiring is connected
either to power terminal blocks or directly to the device terminals.
Type C Wiring: User (field control wiring is connected to master terminal blocks mounted at the top or bottom of
vertical sections which contain combination motor control units or control assemblies; the field load wiring is
connected to master power terminal blocks mounted at the top or bottom of vertical sections or directly to the
device terminals.
MCCs generally consist of a common power bus and a vertical bus for each section to which combination motor
controllers are plugged on. The individual plug-in units are often referred to as buckets and may be inserted and
removed with the main bus energized so long as the disconnecting device for the individual unit is open.
A vertical wireway is supplied internal inter-unit connections and field connections within each section.
MCCs offer the opportunity to group several motor starters together in one location with a space-efficient footprint
vs. individual control cabinets, and like switchboards are available with many options. Removable plug-on units
allow quick change-outs if spare units are kept on hand for the most common sizes of starters in the facility.
Low voltage soft-starters and variable-speed drives may also be mounted within MCCs.
Low voltage switchgear

Table 9-4: Quick reference Low voltage switchgear
120-600 V
1600-5000 A
Through 200 kA
ANSI/IEEE C37.20.1, ANSI/IEEE C37.51, NEMA SG-5,

CAN/CSA C22.2 NO 31-M89, UL 1558
1, 3R
Low voltage switchgear, more properly termed metal enclosed low voltage power circuit breaker switchgear,
is defined per [7] as LV switchgear of multiple or individual enclosures, including the following equipment
as required:
I
Low voltage power circuit breakers (fused or unfused) in accordance with IEEE Std. C37.13-1990 or
IEEE C37.14-1999
Bare bus and connections
Instrument and control power transformers
Instruments, meters, and relays
Control wiring and accessory devices
Low voltage power switchgear is the preferred equipment for medium to large industrial systems where the
advantages of low voltage power circuit breakers, discussed in Section 7, can be utilized to enhance coordination
and reliability. It is typically used as the highest level of low voltage equipment in a facility of this type and, if the
utility service is a low voltage service, the service entrance switchgear as well. Figure 9-5 shows an example of
low voltage switchgear.
Low voltage switchgear, although it performs the same functions and has comparable availability of voltage and
ampacity ratings as switchboards, represents a different mode of development from switchboards and is, in
general, more robust, both due to the construction of the switchgear itself and due to the characteristics of
low voltage power circuit breakers vs. molded-case circuit breakers. For this reason it is preferred over
switchboards where coordination, reliability, and maintenance are a primary concern.
Figure 9-5: Low voltage Switchgear
Low voltage switchgear is compartmentalized to reduce the possibility of internal fault propagation. ANSI C37.20.1
[7] requires each breaker to be provided with its own metal-enclosed compartment. Optional barriers are usually
available to separate the main bus from the cable terminations, forming separate bus and cable compartments
within a section, as well as side barriers to separate adjacent cable and bus compartments.
All low voltage switchgear is required to pass an AC withstand test of 2.2 kV for one minute [7].
As with switchboards, low voltage switchgear is available with many options. The options are generally more
numerous than those for switchboards due to the nature of switchgear service conditions.
Medium voltage power and distribution transformers

Table 9-5: Quick reference Medium voltage power and distribution transformers
Available primary voltage ratings
2400 - 38 kV
Available secondary voltage ratings
120 - 15 kV
Available kVA ratings
Through 10,000 kVA
ANSI/IEEE C57 Series (All Types)

UL 1562 (Dry and Cast-Resin Types)
1, 3R
Medium voltage power and distribution transformers are used for the transformation of voltages for the distribution
of electric power. They can be generally classified into two different types:
Dry-Type: The windings of this type of transformer are cooled via the circulation of ventilating air. The windings
may be one of several types, including Vacuum Pressure Impregnated (VPI), Vacuum Pressure Encapsulated
(VPE), and Cast-Resin. The Cast-Resin types generally are more durable and less likely to absorb moisture in the
windings than the VPI or VPE types. In some cases the primary windings are cast-resin and the secondary
windings are VPI or VPE.
Liquid-Filled: The windings of this type of transformer are cooled via a liquid medium, usually mineral oil,
silicone, or paraffinic petroleum-based fluids.
Liquid-filled units have a generally low in first-cost, but the requirements in NEC [2] Article 450 must be reviewed
to insure that installation requirements can be adequately met, and maintenance must be taken under
consideration since fluid levels should be monitored and the condition of the fluid examined on a regular basis.
They have an expected service life of around 20 years. VPE and VPI dry-type transformers also generally have
low first-costs, have longer lifetimes than liquid-filled units, and are much easier than liquid-filled types to install
indoors; however, consideration should be given to the absorption of moisture by the windings if these are used
outdoors. Installed indoors, these have expected service lifetimes of around 30 years. Cast-resin, dry-type
transformers have generally high first-costs compared to the other types, but have the same installation
requirements as dry-type transformers and have the longest expected service life (around 40 years).
Enclosure styles may also be divided into two basic types: pad-mounted, which is a totally-enclosed type
generally mounted outdoors and with specific tamper-resistance features to prevent inadvertent access by the
general public, and unit substation type, which is an industrial-type enclosure suitable for close-coupling into an
integrated unit substation lineup with primary and secondary equipment (note that unit substation-style
transformers may also be equipped with cable termination compartments as well).
Figure 9-6 shows typical examples of medium voltage power and distribution transformers.
Medium voltage power and distribution transformer capacities may be increased with the addition of fans. Cooling
types are listed as AA (ambient air) for dry-type transformers without fans, and AA/FA (ambient air/forced air) for
dry-type transformers with fans, for an increase of 33% in kVA capacity. The cooling type for a liquid-filled
transformer is listed as OA for units without fans, OA/FA for units with fans, with an increase of 15% kVA capacity
for units 225-2000 kVA, and 25% for units 2,500-10,000 kVA. FFA (future forced air) options are usually
available for both dry and liquid-filled types, although experience has shown that the fans are almost never added
in the field.
a.)
b.)
d.)
c.)
Figure 9-6: Medium Voltage Power and Distribution Transformers

a.) Cast-Coil Dry Type with Unit Substation-Style Enclosure
b.) VPI Dry-Type with Unit Substation-Style Enclosure
c.) Liquid-Filled Type with Unit Substation-Style Enclosure
d.) Dry-Type with Pad-Mounted Enclosure
Table 9-6 gives typical BIL levels for medium voltage power and distribution transformers. These apply to both the
primary and secondary windings. Table 9-7 gives typical design temperature rises.
Table 9-6: Typical BIL levels for medium voltage power and distribution transformers
kV class
VPI/VPE dry-type BIL (kV)
Liquid-filled and cast-resin

dry-type BIL (kV)
1.2
10
30
2.5
20
45
5.0
30
60
7.2
30
60
8.7
45
75
15.0
60
95
25.0
110
125
35.0
N/A*
150
VPI/VPE dry-type transformers are typically not available above 25.0 kV Class
Table 9-7: Typical design temperature rises for medium voltage power and distribution
transformers (over A 30C average/ 40C maximum ambient)
Transformer type
Temperature rise (C)
VPI/VPE dry-type
80, 115, or 150
Cast-coil dry-type
80 or 115
Liquid-filled
55/65 or 65
Impedance levels vary; the manufacturer must be consulted for the design impedance of a specific transformer.
In general, units 1000-5000 kVA typically have 5.75% impedance 7.5% tolerance.
Medium voltage power and distribution transformers are typically available with several types of accessories,
including connections to primary and secondary equipment, temperature controllers and fan packages, integral
fuses for transformers with padmount-style enclosures, etc.
8
Medium voltage metal-enclosed interrupter switchgear

Table 9-8: Quick referencee: Medium voltage metal-enclosed switchgear
2400 V - 38 kV
600 - 2000 A
Through 65 kA
ANSI/IEEE C37.20.3
1, 3R
Metal-enclosed power switchgear is defined by [8] as a switchgear assembly enclosed on all sides and top with
sheet metal (except for ventilating openings and inspection windows) containing primary power circuit switching or
interrupting devices, or both, with buses and connections and possibly including control and auxiliary devices.
Access to the interior of the enclosure is provided by doors or removable covers. Metal-enclosed interrupter
switchgear is defined by [8] as metal-enclosed power switchgear including the following equipment as required:
I
Interrupter switches
Power fuses (current-limiting or noncurrent- limiting)
Bare bus and connections
Instrument Transformers
Control wiring and secondary devices
Metal-enclosed interrupter switchgear is typically used for the protection of unit substation transformers and as
service-entrance equipment in small- to medium- size facilities. Figure 9-7 shows an example of metal-enclosed
interrupter switchgear.
Figure 9-7: Metal-enclosed interrupter switchgear
As with all fusible equipment, overcurrent protection flexibility is limited, however with current-limiting fuses this
equipment has high (up to 65 kA rms symmetrical) short-circuit interrupting capability. The load interrupter
switches in this class of switchgear are designed to interrupt load currents only, and may use air as the
interrupting medium or SF6. They may be arranged in many configurations of mains, but ties, and feeders as
required by the application.
This type of switchgear is frequently used as the primary equipment of a unit substation line-up incorporating
primary equipment, a transformer, and secondary equipment.
Table 9-9 shows the BIL levels of metal-enclosed interrupter switchgear, per [8]. The power frequency withstand
is a one-minute test value. Momentary (10 cycle) and short-time (2s) current ratings are also assigned for this
type of switchgear.
Table 9-9: Voltage withstand levels for metal-enclosed interrupter switchgear, per [8]
Power Frequency Withstand

(rms) (kV)
Impulse Withstand (kV)
4.76
19
60
8.25
36
95
15.0
36
95
27.0
60
125
38.0
80
150
Internal barriering requirements for medium voltage areas within the switchgear are minimal. All low voltage
components are required to be separated by grounded metal barriers from all medium voltage components.
Interlocks are required to prevent access to medium voltage fuses while their respective switch is open and to
prevent closing their respective switch while they are accessible. In the rare case that this type of switchgear
contains drawout devices, shutters must be provided to prevent accidental contact with live parts when the
drawout element is withdrawn.
Available options for this type of switchgear include shunt trip devices for the switches, motor operators for the
switches, blown fuse indication, etc. Relaying of any type, including voltage relaying, must be carefully reviewed to
avoid exceeding the limits of the switches. The application of overcurrent relaying to this type of switchgear is not
recommended unless a short-circuit interrupting element is included, such as a vacuum interrupter.
Medium voltage motor control centers

Table 9-10: Quick referencee: Medium voltage motor conrol centers
2400 V 7.2 kV
Through 3000 A
Through 50 kA
NEMA ICS-3, UL 347
1, 3R
Medium voltage motor controllers are used to control the starting and protection for medium voltage motors. They
generally utilize vacuum contactors rated up to 400 A continuous, in series with a non-load-break isolation switch
and R-rated fuses, fed from a common power bus. The motor starting methods in Section 8 are all generally
supported, including soft-start capabilities. Class E2 units per [9], which employ fuses for short-circuit protection,
are generally the most common. Figure 9-8 shows a typical example of a medium voltage MCC.
Medium voltage MCCs are generally available with a number of options depending upon the manufacturer,
including customized control and multi-function microprocessor-based motor protection relays. The contactors are
generally of roll-out design to allow quick replacement.
Above 7200, metal-clad switchgear is generally used for motor starting.
Figure 9-8: Medium voltage MCC
10
Medium voltage metal-clad switchgear

Table 9-11: Quick referencee: Medium voltage metal-clad switchgear
2400 V 38 kV
Through 3000 A
Through 50 kA
ANSI/IEEE C37.20.2
1, 3R
Metal-clad switchgear is defined by [10] as metal-enclosed power switchgear characterized by the following
necessary features:
I
The main switching and interrupting device is of the removable (drawout type) arranged with a mechanism
for moving it physically between connected and disconnected positions and equipped with self-aligning and
self-coupling primary disconnecting devices and disconnectable control wiring connections.
Major parts of the primary circuit, that is, the circuit switching or interrupting devices, buses, voltage
transformers, and control power transformers, are completely enclosed by grounded metal barriers that have no
intentional openings between compartments. Specifically included is a metal barrier in front of, or a part of, the
circuit interrupting device to ensure that, when in the connected position, no primary circuit components are
exposed by the opening of a door.
All live parts are enclosed within grounded metal compartments.
Automatic shutters that cover primary circuit elements when the removable element is in the disconnected, test,
or removed position.
Primary bus conductors and connections are covered with insulating material throughout.
Mechanical interlocks are provided for proper operating sequence under normal operating conditions.
Instruments, meters, relays, secondary control devices, and their wiring are isolated by grounded metal barriers
from all primary circuit elements with the exception of short lengths of wire such as at instrument transformer
terminals.
The door through which the circuit interrupting device is inserted into the housing may serve as an instrument or
relay panel and may also provide access to a secondary or control compartment within the housing
Medium voltage metal-clad switchgear is generally used as the high-level distribution switchgear for medium- to
large-sized facilities. It is also the preferred choice for service entrance equipment for these types of facilities.
Figure 9-9 shows an example of metal-clad switchgear.
Figure 9-9: Metal-clad switchgear
11
Metal-clad switchgear uses high-voltage circuit breakers, as described in Section 7, fed from a common power
bus. It is configurable in many different arrangements of main, bus tie, and feeder devices to suit the application.
Relays are usually required since the circuit breakers generally do not have integral trip units. This type of
switchgear is the preferred means for accomplishing automatic transfer control and complex generator paralleling
applications; the control may be placed in the switchgear itself or in a separate panel, depending upon the
application and specific end-user preferences.
The construction requirements per [10] insure that metal-clad switchgear is the safest type of switchgear in terms
of operator safety.
The BIL and withstand voltage requirements for this switchgear are the same as for metal-enclosed switchgear as
given in table 9-9 above.
This type of switchgear has many options available to suit the application, such as electric racking for circuit
breakers, ground and test units that allow the grounding/testing of stationary contacts with a circuit breaker
withdrawn, etc.
References
12
[1]
Enclosures for Electrical Equipment (1000 Volts Maximum), NEMA Standards Publication 250-2003.
[2]
[3]
UL Standard for Safety for Panelboards, UL 67, Underwriters Laboratory, Inc., November 2003.
[4]
UL Standard for Safety for Switchboards, UL 891, Underwriters Laboratories, Inc., February 2003.
[5]
UL Standard for Safety for Motor Control Centers, UL 845, Underwriters Laboratories, Inc., August 2005.
[6]
Motor Control Centers, NEMA Standards Publication ICS 18-2001
[7]
IEEE Standard for Metal-Enclosed Low Voltage Power Circuit Breaker Switchgear, IEEE Std. C37.20.1-2001,
October 2002.
[8]
IEEE Standard for Metal-Enclosed Interrupter Switchgear, IEEE Std. C37.20.3-2001, August 2001.
[9]
Industrial Control and Systems: Medium Voltage Controllers Rated 2001 to 7200 Volts AC, NEMA Standards
Publication ICS 3-1993.
[10]
IEEE Standard for Metal-Clad Switchgear, IEEE Std. C37.20.2-1999, July 2000.
Section 10:
Emergency and Standby Power Systems

Introduction
Emergency and standby power systems are designed to provide an alternate source of power if the normal source
of power, most often the serving utility, should fail. As such, reliability of these types of systems is critical and good
design practices are essential.
Code and standards

A.) Classification of emergency and standby power systems
The classification of emergency and standby power systems is as follows:
Emergency Power System: Defined in IEEE Std. 446-1995 [1] as an independent reserve source of electric
energy that, upon failure or outage of the normal source, automatically provides reliable electric power within a
specified time to critical devices and equipment whose failure to operate satisfactorily would jeopardize the health
and safety of personnel or result in damage to property.
The NEC [2] gives a slightly different definition for Emergency Systems as those systems legally required and
classed as emergency by municipal, state, federal, or other codes, or by any governmental agency having
jurisdiction. These systems are intended to automatically supply illumination, power, or both, to designated areas
and equipment in the event of failure of the normal supply or in the event of accident to elements of a system
intended to supply, distribute, and control power and illumination essential for safety to human life.
Standby Power System: Defined in [1] as an independent reserve source of electric energy that, upon failure or
outage of the normal source, provides electric power of acceptable quality so that the users facilities may
continue in satisfactory operation.
The NEC [2] divides standby power systems into two categories, as follows:
Legally Required Standby Systems: Those systems required and so classed as legally required standby by
municipal, state, federal, and other codes or by any governmental agency having jurisdiction. These systems are
intended to automatically supply power to selected load (other than those classed as emergency systems) in the
event of failure of the normal source. FPN: Legally required standby systems are typically installed to serve loads,
such as heating and refrigeration systems, communications systems, ventilation and smoke removal systems,
sewage disposal, lighting systems, and industrial processes that, when stopped during any interruption of the
normal electrical supply, could create hazards or hamper rescue and fire-fighting operations.
Optional Standby Systems: Those systems intended to supply power to public or private facilities or property
where life safety does not depend on the performance of the system. Optional standby systems are intended to
supply on-site generated power to selected loads either automatically or manually. FPN: Optional standby systems
are typically installed to provide an alternate source of electric power for such facilities as industrial and
commercial buildings, farms, and residences and to serve loads such as heating and refrigeration systems, data
processing and communications systems, and industrial processes that, when stopped during any power outage,
could cause discomfort, serious interruption of the process, damage to the product or process, and the like.
B.) IEEE Standard 446-1995

IEEE Standard 446-1995, IEEE Recommended Practice for Emergency and Standby Power Systems for
Industrial and Commercial Applications [1], is a general engineering reference for the design of these systems.
C.) The National Electrical Code

The National Electrical Code [2] contains requirements for emergency systems in Article 700, Legally-Required
Standby Systems in Article 701, and Optional Standby Systems in Article 702. In addition, Article 445
(Generators), 517 (Health Care Facilities), 665 (Integrated Electrical Facilities), and 705 (Interconnected Electrical
Power Production Sources) are all of particular interest for emergency and standby power systems.
The NEC [2] requirements for emergency and standby power systems are discussed in further detail below.
D.) NFPA 110

NFPA 110 [3], Standard for Emergency and Standby Power Systems, defines how emergency and standby
power systems are to be installed and tested. It contains requirements for energy sources, transfer equipment,
and installation and environmental considerations. It divides emergency power systems into Types, Classes,
and Levels.
The Type refers to the maximum time that an emergency power system can remain unpowered after a failure of
the normal source. The Types are listed in table 10-1 [3]:
Table 10-1: NFPA 110 emergency power system types (essentially the same as [3]
table 4.1(B))
Type
Power restoration time
Basically Uninterruptible (UPS Systems)
10
10 sec
60
60 sec
120
120 sec
Manual stationary or nonautomatic no time limit
The Class of an emergency power system refers to the minimum time, in hours, for which the system is designed
to operate at its rated load without being refueled or recharged. The Classes for emergency power systems are
shown in table 10-2 [3]:
Table 10-2: NFPA 110 emergency power system classes (essentially the same as [3]
table 4.1(B))
Class
Power restoration time
0.083
0.083 hr. (5 min.)
0.25
0.25 hr. (15 min.)
2 hr.
6 hr.
48
48 hr.
Other time, in hours, as required by the application, code, or user.
The Level of an emergency power system refers to the level of equipment installation, performance, and
maintenance requirements. The Levels for emergency power systems are shown in table 10-3 [3]:
Table 10-3: NFPA emergency power system levels
Level
When Installed
When failure of the equipment to perform could result in loss of human life or serious injuries
When failure of the equipment to perform is less critical to human life and safety and where
the authority having jurisdiction shall permit a higher degree of flexibility than that provided by
a level 1 system
E.) NFPA 101

NFPA 101 [4], Life Safety Code, addresses those construction, protection, and occupancy features necessary to
minimize danger to life from fire, including smoke, fumes, or panic. It defines the requirements for what systems
the Emergency Power System will supply.
F.) NFPA 99
NFPA 99 defines establishes criteria to minimize the hazards of fire, explosion, and electricity in health care
facilities. It defines several specific features of electric power systems for these facilities.
Reasons for application

Emergency and standby power systems are generally designed into the over-all electrical system for one of the
following two reasons:
I
Legal Requirements As required by the NEC [2] NFPA 101 [4], NFPA 99 [5], and other local, state, and federal
codes and requirements. These are concerned with the safety of human life, protection of the environment, etc.
Economic Considerations Continuous process applications often require a continuous source of electrical
power to avoid significant economic loss. In some cases even a momentary loss of power can be disastrous.
Co-generation systems which are used to sell power back to the utility as part of an energy management strategy
are not discussed in this section.
Power sources
Generators are by far the most prevalent source of power for emergency and standby power systems. For most
commercial and industrial power systems these will be engine-generator sets, with the prime-mover and the
generator built into a single unit. For reciprocating engines, diesel engines are the most popular choice of primemover for generators, due to the cost of the diesel engines as compared to other forms of power and the relative
ease of application. Gasoline engine generator sets are also available and are generally less expensive than
diesel generator sets, but suffer from the disadvantages of higher operating costs, greater fuel storage hazards,
and shorter fuel storage life as compared to diesel. Diesel engines can also run on natural gas, although for
maximum efficiency specially-tuned engines for natural gas use are available.
The other alternative for generators is the turbine generator, typically powered by natural gas. Gas-turbine
generator sets are generally lighter in weight than diesel engine-generator sets, run more quietly, and generally
require less cooling and combustion air, leading to lower installation costs. However, gas-turbine generator sets
are more expensive than diesel engine-generator sets, and require more starting time (normally around 30 s
compared to the 10-15 seconds for diesels). The long starting-time requirement and lack of available small sizes
(< 500 kW) makes the gas-turbine generators infeasible in many applications.
Generator installations must consider the combustion and cooling air required by the generator and prime mover,
as well as the provisions for the removal of exhaust gasses. Noise abatement must also be considered.
These considerations increase the installation costs, especially for reciprocating-engine units such as diesel or
gasoline engines. Further, the fuel supply must be considered; building code and insurance considerations may
force the fuel storage tank to be well removed from the generator(s), usually forcing the addition of a fuel transfer
tank near the generator(s).
Care must be taken when sizing engine-generator sets for a given application since several ratings exist for the
output capability of a given machine. The continuous rating is typically the output rating of the engine-generator
set on a continuous basis with a non-varying load. The prime power rating is typically the continuous output
rating with varying load. The standby rating is typically the output rating for a limited period of time with varying
load. The manufacturer must be consulted to define the capabilities of a given unit.
A second alternative for emergency or standby system power is a second utility source. However, the
procurement of a second utility source which is sufficiently independent from the normal service may be
economically infeasible.
Solid-state converters that invert DC voltage from a battery system are another alternative, although they can be
difficult to apply and generally are not available in the larger sizes that may be needed for a medium to large
emergency or standby system.
Because motor starting and block loading can have a big effect on the output voltage and frequency of a small
generator such as the engine-generator sets described above, and also because power is not available during the
engine starting period, a buffer between the generators and sensitive load equipment is generally required. The
Uninterruptible Power Supply (UPS) is usually the buffer of choice for these applications. UPSs are available in
several different topologies, but the operational goals are the same regardless of topology: The supply of
uninterrupted power to sensitive, critical loads. The most popular topology for a UPS is the double conversion
topology, as shown in figure 10-1:
Figure 10-1: Double-conversion UPS topology
So long as the batteries are properly maintained, the AC output should not be affected by change in frequency
or voltage, or even a complete loss, at the input, so long as backup time of the UPS is not exceeded. Other
topologies exist, including the line interactive, double-conversion rotary, hybrid rotary, and line-interactive
rotary topologies, each with advantages and disadvantages of application. UPS systems do not alleviate the need
for a generator or second utility service power source, but they do serve to buffer critical loads from the effects of
generator starting time and voltage and frequency variations.
Switching devices
A means must be provided to switch the critical loads from the normal utility source to the standby power source.
Several types of device are available for this.
An automatic transfer switch is defined as self-acting equipment for transferring one or more load conductor
connections from one power source to another [1]. The automatic transfer switch is the most common means of
transferring critical loads to the emergency/standby power supply. An automatic transfer switch consists of a
switching means and a control system capable of sensing the normal supply voltage and switching over to the
alternate source should the normal source fail. Automatic transfer switches are available in ratings from 30-50 A,
and up to 600V [1]. Because automatic transfer switches are designed to continuously carry the loads they serve,
even under normal conditions, care must be used in sizing these so that the potential for failure is minimized.
Automatic test switches with adjustable pickup and dropout setpoints and integral testing capability are generally
preferred. An automatic transfer switch is generally an open-transition device that will not allow paralleling of the
two sources. Manual versions of transfer switches are also available. A one-line representation of an automatic
transfer switch is shown in figure 10-2.
Figure 10-2: Automatic transfer switch one-line diagram representation
Other options for transferring devices include electrically-operated circuit breakers, as described in System
protection section (section7 in this guide). For medium voltage transfers, medium voltage circuit breakers are
generally used. Manual versions of circuit-breaker transferring schemes are, of course, also available.
Bypass/isolation switches, as their name suggests, are used to bypass an automatic transfer switch (or other
switching means) and connect the source directly to the load and allow isolation of the transfer switch for
maintenance. Figure 10-3 shows a typical bypass/isolation switch arrangement along with the transfer switch:
Figure 10-3: Bypass/isolation switch application
In figure 10-3 the bypass blade B serves to bypass the automatic transfer switch, and isolating contacts I
serve to isolate the automatic transfer switch. Bypass/isolation switches are typically manually-operated devices.
Bypass/isolation switches are available with a test position in which only the ATS-to-load isolation contact
(marked with an asterisk [*] in figure 10-3) is open, allowing the transfer switch to be operated without
disconnecting the load.
Static Transfer Switches are typically used when high-speed (~4ms) operation is required. The most common
application is to bypass a UPS so that a UPS failure will not result in interruption of service to the load.
System arrangements
Various ways of arranging emergency and standby power systems exist. The most common arrangements
are given here.
A.) Basic arrangement radial system

The most basic arrangement for an emergency or standby power system is shown in figure 10-4. This can be
recognized as an extension of the single-source radial system from System arrangements section (section 5 of
this guide), figure 5-2, with the transformer omitted. The transfer switch transfers the emergency/standby loads
to the alternate source upon failure of the normal source. This arrangement extends the same inherent
weaknesses of the radial system to the emergency system, since a single failure of one piece of equipment can
result in loss of service to the emergency/standby loads. Note that the single generator shown may be several
engine-generator sets operating in parallel, if necessary. This simple system may be expanded to the other
system types in Section 5.
Figure 10-4: Simple emergency/standby system arrangement
B.) More complex systems

The basic arrangement from figure 10-4 may be extended to the other system arrangements shown in Section 5.
For example, the secondary-selective system, shown in figure 5-8, could be equipped with an emergency system
as shown in figure 10-5:
Figure 10-5: Example of a more complex emergency/standby system arrangement
In figure 10-5, the emergency/standby load at the bottom of the figure will always be supplied by one of the
normal sources if possible, and by the generator(s) if not. This will avoid the generator starting time for this load if
one utility source were to fail. The two emergency/standby loads in the middle of the figure will be supplied by
their respective switchboard busses or by the emergency source.
Emergency/standby systems are not limited to the low voltage level. For example, the primary selective/primary
loop/secondary selective system shown in figure 5-14 can be expanded to include an emergency system, as
shown in figure 10-6:
Figure 10-6: Medium voltage emergency/standby system implementation
In figure 10-6 there is a great deal of flexibility in the system operation. However, instead of automatic transfer
switches metal-clad switchgear is used increasing the complexity of the system.
C.) Hospital arrangements

NFPA 99 [5] and the NEC [2] have very unique requirements for the design of a hospital emergency system.
The emergency system is classified into the essential electrical system, which is comprised of alternate sources
of power and all connected distribution systems and ancillary equipment, designed to ensure continuity of
electrical power to designated areas and functions of a health care facility during disruption of disruption of
normal power sources, and the emergency system itself, which is a system of circuits and equipment intended to
supply alternate power to a limited number of prescribed functions vital to the protection of life and safety [2].
The emergency system is a part of the essential electrical system. The minimum arrangement, for hospitals
150 kVA or less, is shown in figure 10-7. The minimum requirement over 150 kVA is shown in figure 10-8.
Figure 10-7: Minimum requirement per NEC [2] and NFPA 99 [5] for essential electrical system for
hospitals 150 kVA or Less (same as [2] FPN figure 517.30 No.2)
Figure 10-7: Minimum Requirement per NEC [2] and NFPA 99 [5] for Essential Electrical system for
Hospitals over 150 kVA (same as [2] FNP figure 517.30 No. 1)
The essential electrical system supplies the equipment system, defined as a system of circuits and equipment
arranged for delayed, automatic, or manual connection to the alternate power source and that serves primarily 3phase power equipment [2]. The emergency system supplies, which itself part of the essential electrical system,
supplies the life safety branch, which is a subsystem of the emergency system consisting of feeders and branch
circuitsintended to provide adequate power needs to insure safety to patients and personnel [2]. The
emergency system also supplies the critical branch, which is a subsystem of the emergency system consisting of
feeders and branch circuits supplying energy to task illumination, special power circuits, and selected receptacles
serving areas and functions related to patient care [2]. For hospitals of 150 kVA and less the equipment system,
life safety branch, and critical branch may be on the same transfer switch. Note that the transfer switch(es) for the
equipment system above 150 kVA is required to be delayed.
NEC requirements
The following are highlights from the NEC [2] requirements for emergency and standby power systems. This is not
intended to list all NEC requirements for these systems, but to illustrate the major points that apply in the most
common installations and affect the power system design. For the full text of the complete NEC requirements for
these systems, consult the NEC.
A.) Emergency systems (Article 700)

The NEC definition for an emergency system was given at the beginning of this section. These requirements
apply to those systems meeting this definition:
Witness Test: The authority having jurisdiction must conduct or witness a test of the complete system and
periodically afterward. [700.4 (A)]
Emergency systems must be tested periodically on a schedule acceptable to the authority having jurisdiction to
ensure the systems are maintained in proper operating condition. A written record must be kept of these tests.
[700.4 (B), (C), and (D)]
Battery systems that are part of the emergency system must be periodically maintained. [700.4 (B)]
A means for testing all emergency lighting and power systems during maximum anticipated load conditions must
be provided. [700.4 (E)]
The alternate power source is required to be sized to supply all emergency loads simultaneously. [700.5 (A)]
The alternate power source is permitted to supply emergency, legally required standby, and optional standby
system loads where the source has adequate capacity or where automatic selective load pickup or load
shedding is provided to insure adequate power to the emergency, legally required standby, and optional standby
system loads. If these requirements are met the system may also be used for peak load shaving. Peak load
shaving operation may satisfy the requirement for periodic testing if acceptable to the authority having
jurisdiction. A portable or temporary alternate source must be available if the emergency generator is out of
service for repair. [700.5 (B)]
Transfer equipment must be automatic, identified for emergency use, and approved by the authority
having jurisdiction. Automatic transfer switches must be electrically operated and mechanically held.
[700.6 (A) and (C)]
Transfer equipment must supply only emergency loads. [700.6 (D)]
Audible and visual signal devices must be provided for indication of derangement of the emergency source, that
the battery is carrying load, that the battery is not functioning, and to indicate a ground fault in solidly-grounded
wye systems of more than 150 V to ground and over 1000 A. The sensor for ground-fault indication must be
located at or ahead of the main system disconnecting means for the emergency source. [700.7]
A sign must be placed at the service entrance equipment, indicating the type and location of on-site emergency
power sources. A sign is also required where the grounded circuit conductor connected to the emergency
source is connected to a grounding electrode conductor at a location remote from the emergency source.
[700.8]
All boxes and enclosures for emergency circuits must be permanently marked so that they will be readily
identified as a component of an emergency circuit or system. [700.9 (A)]
Wiring from an emergency source or emergency source distribution overcurrent protection to emergency loads
must be kept entirely independent of all other wiring and equipment. Exceptions apply where load equipment
must have wiring from two sources. [700.9 (B)]
For occupancies of not less than 1000 persons or in buildings above 75 ft. in height with assembly, educational,
residential, detention/correctional, business, or mercantile occupancy class the feeder circuit wiring must be 1.)
installed in spaces or areas that are fully protected by an approved automatic fire suppression system, or 2.) be
a listed electrical circuit protective system with a 1-hour fire rating, or 3.) be protected by a listed thermal barrier
system for electrical system components, or 4.) be protected by a fire-rated assembly listed to achieve a
minimum fire rating of 1 hour, or 5.) be embedded in not less than 50mm of concrete. Feeder circuit equipment
must be either in spaces fully protected by a approved automatic fire suppression systems or in spaces with
a 1-hour fire resistance rating. [700.9 (D)]
In the event of failure of the normal supply to, or within, the building or group of buildings concerned, emergency
lighting, power, or both, must be available within the time required by the application but not to exceed 10
seconds.
The alternate source of power must be a storage battery, generator set, UPS, separate service, or fuel cell
system, each with restrictions on its use. [700.12 (A), (B), (C), (D), and (E)].
Storage batteries must have sufficient capacity to supply and maintain the total load for a minimum period of one
hours, without the voltage applied to the load falling below 87% of nominal. The battery charging means must be
automatic. [700.12 (A)]
Generator sets must have a prime-mover acceptable to the authority having jurisdiction, and means of
automatically starting the prime mover on failure of the normal service. If the prime-mover is an internal
combustion engine, an on-premises fuel supply must be provided to allow not less than 2 hours full-demand
operation of the system. If power is required for operation of fuel transfer pump to deliver fuel to a generator set
day tank, this pump must be connected to the emergency power system. Generator sets must not be solely
dependent on a public utility gas system for their fuel supply for a municipal water supply for their cooling
systems. If dual supplies for these are used, means must be provided to automatically transfer from one supply
to the other. If a storage battery is used for control or signal power or as the means of starting the prime mover, it
must be equipped with an automatic charging means independent of the generator set. Where power is required
for the operation of the dampers used to ventilate the generator set, the dampers must be connected to the
emergency system. [700.12 (B) (1), (2), (3), and (4)]
10
If a generator set requires more than 10 seconds to develop power, an auxiliary power supply that energizes the
emergency system until the generator can pick up the load is permitted. [700.12 (B) (5)]
Outdoor generator sets do not require an additional disconnecting means where the ungrounded conductors
serve or pass through the building or structure, so long as they are equipped with a readily-accessible
disconnecting means located within sight of the building or structure supplied. [700.12 (B)(6)]
UPSs used to provide power for emergency systems must comply with the applicable provisions for battery
systems and generators.
An additional utility service is permitted to be the power source for the emergency system, if acceptable to the
authority having jurisdiction. A separate service drop or service lateral and service conductors sufficiently remote
electrically and physically from other service conductors to minimize the possibility of simultaneous interruption
of supply must be supplied. [700.12 (D)]
Fuel cell systems must be capable of supplying and maintaining the total load for not less then two hours of fulldemand operation. Fuel cell systems must meet the requirements of Parts II through VIII of Article 692 (Fuel Cell
Systems). A single fuel cell that serves as the normal source for the building or group of buildings concerned
cannot serve as the alternate source. [700.12(E)]
Individual unit equipment for emergency illumination must have a rechargeable battery, a battery charging
means, provisions for one or more lamps mounted on the equipment or terminals for remote lamps, and a
relaying device arranged to energize the lamps automatically upon failure of the supply to the unit equipment.
The battery must be capable of supplying the lamps for no less than one hours at not less than 60% of the initial
illumination level. [700.12 (F)]
Individual unit equipment for emergency illumination must be fixed in place. Flexible cord-and-plug installation is
permitted if the cord is no more than 3ft. in length. The branch circuit feeding the unit equipment must be the
same as that serving normal lighting in the area and connected ahead of any local switches, and must be clearly
identified at the distribution panel. Alternatively, for areas with at least three normal lighting branch circuits the
emergency illumination unit equipment may be supplied by a separate branch circuit with a lock-on feature.
[700.12 (F)]
No appliances or lamps, other than those specified for emergency use, are allowed on emergency
lighting circuits. [700.15]
Emergency illumination must include all required means of egress lighting, illuminated exit signs, and all other
lights specified as necessary to provide required illumination. Failure of any individual lighting element must not
leave in total darkness any space that requires emergency illumination. If HID lighting is used as emergency
illumination, it must operate until normal illumination has been restored. [700.16]
Emergency lighting must have either an emergency lighting supply, with provisions for automatically transferring
the emergency lights upon the event of failure of the general lighting system supply, or two or more separate
and complete systems with independent power supplies, each providing sufficient current for emergency lighting
purposes. If two systems are used, means must be provided for automatically energizing either system upon
failure of the other unless they are both kept lighted. [700.17]
All branch circuits that supply equipment classed as emergency equipment must have an emergency supply
source to which the load will be transferred upon the failure of the normal supply. [700.18]
Emergency lighting circuits must be arranged so that only authorized persons have control of emergency
lighting. Exceptions apply. [700.20]
Switches in series or 3- and 4-way switches cannot be used in emergency lighting circuits. [700.20]
Control switches for emergency lighting must be in convenient locations for authorized persons. In assembly
occupancies or theaters, audience areas of motion picture studios, and performance areas, a switch
for controlling emergency lighting systems must be in the lobby or at a place conveniently accessible
thereto. [700.21]
Emergency lighting on the exterior of a building that is not required for illumination when there is sufficient
daylight may be controlled by an automatic light-actuated device. [700.22]
The branch-circuit overcurrent devices in emergency circuit must be accessible to authorized persons
only. [700.25]
The alternate source for emergency systems is not required to have ground-fault protection of equipment.
Ground-fault indication is required. [700.26]
Emergency system(s) overcurrent devices must be selectively coordinated with all supply-side overcurrent
protective devices. [700.27]
B.) Legally required standby systems (Article 701)

I
System periodic testing and maintenance requirements are essentially the same as for emergency systems,
except that the authority having jurisdiction is only required to witness the test upon installation. [701.5]
The legally required standby system alternate power source is permitted to supply both legally required standby
system and optional standby system loads, provided that it either has enough capacity to handle all connected
loads or that automatic selective load pickup and load shedding is provided that will ensure adequate power to
the legally required standby circuits. [701.6]
Requirements for transfer equipment are essentially the same as for emergency systems, except that no
restriction is placed upon the use of transfer equipment use for other systems in addition to the legally required
standby system. [701.7]
Audible and visual signal devices must be provided for indication of derangement of the standby source, that the
standby source is carrying load, and that the battery charger is not functioning. [701.8]
Signage requirements are essentially the same as for emergency systems. [701.9]
Wiring for legally required standby systems is permitted to occupy the same raceways, cables, boxes, and
cabinets with other general wiring. [701.10]
In the event of failure of the normal supply to, or within, the building or group of buildings concerned,
legally required standby power must be available within the time required by the application but not to exceed
60 seconds.
The alternate source of power must be a storage battery, generator set, UPS, separate service, connection
ahead of the service disconnecting means, or fuel cell system, each with restrictions on its use.
[701.11 (A), (B), (C), (D), (E), and (F)]
The requirements for storage batteries, generator sets, UPSs, separate utility service, and fuels cells as the
standby power source are essentially the same as for emergency systems, except the requirements for fuel
transfer pumps and ventilation dampers to be connected to the system for generator sets. [701.11 (A)]
Where acceptable to the authority having jurisdiction, connections ahead of but not within the same cabinet,
enclosure, or vertical switchboard section as the service disconnecting means may serve as the standby power
source. This connection ahead of the normal service must be sufficiently separated from the normal main
service disconnecting means to prevent simultaneous interruption of supply. [701.11 (D)]
The requirements for individual unit equipment for legally required standby illumination are essentially the same
as for emergency illumination individual unit equipment. [701.11 (G)]
Legally-required standby system overcurrent protection requirements are essentially the same as for emergency
systems, except that ground-fault indication is not required. [701.15, 701.17, 701.18]
C.) Optional standby systems (Article 702)

I
Transfer equipment is required, except in the case of temporary connection of a portable generator where
conditions of maintenance and supervision ensure that only qualified persons service the installation and
where normal supply is physically isolated by a lockable disconnect means or by disconnection of supply
conductors. [702.6]
Audible and visual signal devices must be provided for indication of derangement of the standby source and to
indicate that the optional standby source is carrying load. [702.7]
11
Signage requirements are essentially the same as for emergency and legally required standby systems. [702.8]
Wiring for optional standby systems is permitted to occupy the same raceways, cables, boxes, and cabinets with
other general wiring. [702.9]
Where a portable optional standby source is used as a separately derived system, it must be grounded to a
grounding electrode in accordance with Article 250.30. Where a portable optional standby source is used as a
non-seperately derived system, the equipment grounding conductor must be bonded to the system grounding
electrode. [702.10]
Outdoor generator sets do not require an additional disconnecting means where the ungrounded conductors
serve or pass through the building or structure, so long as they are equipped with a readily-accessible
disconnecting means located within sight of the building or structure supplied. [702.11]
D.) Health care facility essential electrical systems (Article 517 part III)
I
The essential electrical system is required to serve a limited amount of lighting and power service, which is
considered essential for life safety and orderly cessation of procedures during the time normal service is
interrupted for any reason. This includes clinics, medical and dental offices, outpatient facilities, nursing homes,
limited care facilities, hospitals, ad other health care facilities serving patients. [517.25]
The essential electrical system must meet the requirements of Article 700 (Emergency Systems), except as
amended by Article 517. [517.26]
N Hospitals (Articles 517.30 517.35)
12
The essential electrical systems for hospitals must comprised of two separate systems: The emergency
system and the equipment system. The emergency system must be limited to circuits essential to life
safety and to critical patient care, designated as the life safety branch and the critical branch. The
equipment system must supply major electrical equipment necessary for patient care and basic hospital
operation. [517.30 (B) (1), (2), and (3)]
The number of transfer switches used must be based on reliability, design, and load considerations.
One transfer switch is permitted to serve one or more branches or systems in a facility with a maximum
demand on the essential electrical system of 150 kVA. [517.30 (B) (4)]
Other loads not specifically mentioned in Article 517 must be served with their own transfer switches.
These loads must not be transferred to the essential electrical system generating equipment if the
transfer will overload the equipment, and they must be automatically shed upon generating equipment
overloading. [517.30 (B)(5)]
The life safety and critical branch circuit wiring must be kept independent of all of other wiring and
equipment and must not enter the same raceways, boxes, or cabinets with each other or other wiring.
Exceptions apply where transfer or load equipment must have wiring from two sources. Wiring for the
equipment system is permitted to occupy the same raceways boxes, or cabinets of other circuits that are
not part of the emergency system. [517.30 (C)]
All wiring of the emergency system must be mechanically protected. Nonflexible metal raceways, type MI
cable, or Schedule 80 rigid nonmetallic conduit are permitted, except that nonmetallic raceways cannot
be used for branch circuits that supply patient care areas. Schedule 40 rigid nonmetallic conduit, flexible
nonmetallic or jacketed metallic raceways, or jacketed metallic cable assemblies listed for installation in
concrete may be used if encased in no less than 2 in. of concrete. Listed flexible metal raceways and
listed metal sheathed cable assemblies may be used under certain conditions. Flexible power cords of
appliances or other utilization equipment and secondary circuits of Class 2 or Class 3 communications or
signaling systems are exempted from being run in metal raceways. [517.30 (C) (3)]
Generator sizing may be based upon demand calculations rather than on the entire load operating
simultaneously as required in 700.5. [517.30 (D)]
All receptacles supplied by the emergency system must have a distinctive color or marking. [517.30 (E)]
The life safety branch is permitted to supply only illumination of egress means, exit signs, alarm and
alerting systems, communications systems used during emergency conditions, task illumination at
the generator set location, elevator cab lighting, control, communications, and signal systems, and
automatic doors. [517.32]
The critical branch is permitted to supply task illumination and selected receptacles in critical care areas,
isolated power systems in special environments, task illumination and selected receptacles for selected
patient care areas, general care beds, selected labs, additional patient care task illumination and
receptacles as needed, nurse call systems, blood, bone, and tissue banks, telephone equipment rooms
and closets, etc. (complete list given in the NEC text). [517.33]
The critical branch may be subdivided into two or more branches. [517.33 (B)]
Delayed automatic connection to the equipment system must be provided for central suction systems,
sump pumps, compressed air systems, smoke control and stair pressurizing systems, kitchen hood
supply or exhaust systems, and supply, return, and exhaust ventilating systems for selected locations
(complete list given in NEC text). [517.34 (A)]
Delayed automatic or manual connection to the equipment system must be provided for selected heating
equipment, selected elevators, hyperbaric and hypobaric facilities, automatically operated doors, selected
electrically-heated autoclaving equipment, and other selected equipment. [complete list given in NEC text
and NFPA 99:4.2.2.2.3.5(9)] [517.34 (B)]
Generator accessories, such as the transfer fuel pump, electrically operated louvers, and other
accessories essential for generator operation, must be arranged for non-delayed automatic connection to
the alternate power source via the equipment system.
A minimum of two sources of power are required, one normal, one alternate. The alternate source may be
generator(s) on the premises, an external utility service if the normal service is a generator(s) on the
premises, or a battery system.
N Nursing homes and limited care facilities (Article 517.40 517.44)

G
Applicability depends upon the type of care given at the facility. Specific exceptions are listed for certain
types of facilities (see NEC text for details). If a nursing home provides inpatient hospital care, it must
conform to the requirements for hospitals. Nursing homes and limited care facilities that are contiguous or
located on the same site with a hospital are permitted to have their essential electrical systems supplied
by the hospital. [517.40]
The essential electrical system must be comprised of two separate branches: The life safety branch and
the critical branch. [517.41 (A)]
Requirements for transfer switches are essentially the same as for hospitals. [517.41 (C)]
The life safety branch must be kept entirely independent of all other wiring and equipment and must not
enter the same raceways, boxes, or cabinets with other wiring. Exceptions apply where transfer or load
equipment must have wiring from two sources. [517.41 (D)]
Requirements for receptacle identification are essentially the same as for hospitals. [517.41 (E)]
The life safety branch must be automatically restored via the alternate power source within 10 seconds
after interruption of the normal source. [517.42]
The life safety branch must supply only illumination of means of egress, exit signs, alarm and alerting
systems, communications systems used during emergency conditions, dining and recreation areas,
task illumination at the generator set location, and elevator cab lighting, control, communications,
and signal systems.
Delayed automatic connection to the critical branch must be provided for task illumination and selected
receptacles in selected patient care areas, sump pumps and other equipment required to operate for the
safety of major apparatus and associated control systems and alarms, smoke control and stair
pressurization systems, kitchen hood supply and/or exhaust systems if required to operate during a fire in
or under the hood, and supply, return, and exhaust ventilating systems for airborne infections isolation
rooms (complete list given in NEC text). [517.43(A)]
13
Delayed automatic connection to the critical branch must be provided for heating equipment to provide
heating for patient rooms (exceptions apply), elevator service and additional illumination, receptacles,
and equipment. [517.43 (B)]
The alternate source of power must be a generator(s) located on the premises unless the normal source
is a generator(s) on the premises, in which case the alternate source may be either another generator set
or external utility service. In certain cases a battery system may be used (see NEC text). [517.44 (B)]
N Other health care facilities (Article 517.45)

G
The essential electrical system must be a battery or generator system, if required per NFPA 99.
Where electrical life support equipment is required or critical care areas are present, the requirements
for hospitals apply.
The requirements of Article 700 apply to battery systems. Generator systems must be as described
for hospitals.
References
14
[1]
IEEE Recommended Practice for Emergency and Standby Power Systems for Industrial and Commercial
Applications, IEEE Std. 446-1995, December 1995.
[2]
[3]
Standard for Emergency and Standby Power Systems, NFPA 110, The National Fire Protection Association,
2005 Edition.
[4]
Life Safety Code, NFPA 101, The National Fire Protection Association, 2003 Edition.
[5]
Standard for Health Care Facilities, NFPA 99, The National Fire Protection Association, 2005 Edition.
Section 11:
Power Quality Considerations
Introduction
The term power quality may take on any one of several definitions. The strict definition of power quality is the
concept of powering and grounding electronic equipment in a manner that is suitable to the operation of that
equipment and compatible with the premises wiring system and other connected equipment [1]. In practice,
however, the term power quality is often used to denote the proximity of the system voltage to its sinusoidal form
at the nominal voltage level. Deviation from this sinusoidal norm therefore denotes a power quality issue. Strictly
speaking, this deviation is actually a power disturbance, defined as any deviation from the nominal value (or
from some selected thresholds based upon tolerance) of the AC input power characteristics [1]. The most
common power disturbances are, as defined by [1]:
Overvoltage: An RMS increase in the AC voltage, at the power frequency, for a period of time greater than 1 min.
Typical values are 110%-120% of nominal.
Undervoltage: An RMS decrease in the AC voltage, at the power frequency, for a period of time greater than
1 min. Typical values are 80-90% of nominal.
Swell: An increase in RMS voltage or current at the power frequency for durations from .5 cycle-1 min. Typical
values are 110%-180% of nominal.
Sag: An RMS reduction in the AC voltage, at the power frequency, for durations from _ cycle to a few seconds.
Interruption: The complete loss of voltage. A momentary interruption is a voltage loss (<10% of nominal) for a
time period between .5 cycles and 3 seconds). A temporary interruption is a voltage loss (<10% of nominal) for a
time period between 3 seconds and 1 min. A sustained interruption is the complete loss of voltage for a time
period greater than 1 min.
Notch: A switching (or other) disturbance of the normal power system voltage waveform, lasting less than _ cycle;
which is initially of opposite polarity to the waveform, and is thus subtractive from the normal waveform in terms of
the peak value of the disturbance voltage. This includes a complete loss of voltage for up to _ cycle.
Transient: A subcycle disturbance in the AC waveform that is evidenced by a sharp discontinuity of the
waveform. May be of either polarity and may be additive to, or subtractive from, the nominal waveform.
Flicker: A variation in input voltage, either magnitude or frequency, sufficient in duration to allow visual
observation of a change in electric light source intensity.
Harmonic Distortion: The mathematical representation of distortion of the pure sine waveform. This refers to
the distortion of the voltage and/or current waveform, due to the flow of non-sinusoidal currents.
Electrical Noise: Unwanted electrical signals that produce undesirable effects in the circuits of the control
systems in which they occur. Noise may be further categorized as transverse-mode noise, which is measurable
between phase conductors but not phase-to-ground, and common-mode noise, which is measurable phase-toground but not between phase conductors. This noise may be conducted or radiated. Also referred to as RFI
(radio-frequency interference) or EMI (electro-magnetic interference).
The causes of the common power disturbances listed can vary greatly. Common causes are listed in Table 11-1:
Table 11-1: Common power disturbance causes

Disturbance
Common causes
Overvoltage
Voltage regulator malfunction

Improperly set transformer taps
Improperly-applied power factor correction capacitors
Undervoltage
Voltage regulator malfunction

Improperly set transformer taps
Large source impedance (weak system)
Voltage Swell
Recovery of system voltage following a fault

Remote switching (capacitors, etc.)
Voltage Sag
Remote fault
Cold-load pickup (motor starting, transformer energization, etc.)
Large step loads
Transient
(Typically voltage surges)
Lightning strikes
Close-in switching (capacitors, etc.)
Complex circuit phenomena such as current chopping, restrikes, system resonance, etc.
Flicker
Arcing loads such as arc furnaces

Also same sources that cause voltage sags and swells
Notches and
Harmonic Distortion
Power electronic converter equipment such as rectifiers, inverters, drives, etc., which produce nonsinusoidal load current and commutation notches
Interruptions
Faults causing overcurrent protective device operation

Utility maintenance activities
Electrical Noise
Power electronic converter equipment such as drives

Conductors and power equipment which carry large amounts of current
Arcing in overcurrent protective devices
Power disturbances can greatly affect utilization equipment. For example, sensitive electronic medical
equipment can malfunction, adjustable speed motor drives may trip off-line, etc. Interruptions can cause
microprocessor-based equipment such as computers to lose data. In extreme conditions, such as for voltage
surges caused by direct lightning strikes, both power equipment and utilization equipment may be subject to
failure. With the high reliability requirements imposed upon power systems, it is imperative that power system
disturbances, or potential disturbances, be mitigated to avoid down-time, equipment failure, and risk to human life.
Power quality metrics

There are various methods for categorizing the severity of power disturbances. The most typical indices for
measuring power quality disturbances are:
Distortion Factor: The ratio of the root square value of the harmonic content to the root square value of the
fundamental quantity, expressed as a percentage of the fundamental, also known as total harmonic distortion [1].
(11-1)
where
Vh
V1
is the RMS harmonic voltage (or current) value at a frequency of n times the fundamental frequency
is the RMS fundamental-frequency voltage or current
Alternate forms for the distortion factor are given in [2] as percentages of the nominal voltage or demand load
current for the system under consideration, for use in evaluation of the harmonic content of the system voltage or
current. These are referred to as Total Harmonic Distortion (THDVn) and Total Demand Distortion (TDD), defined
as follows:
(11-2)
(11-3)
where
Vh
Vn
Ih
IL
is the RMS value of the nth harmonic component of the voltage

is the RMS nominal fundamental voltage value
is the RMS value of the nth harmonic component of the current
is the maximum demand load current, typically the average maximum monthly demand over a 12month period
Crest Factor: The ratio of the peak value of a periodic function to the RMS value, i.e.:
(11-4)
where
ypeak
yrms
is the peak value of a periodic function

is the RMS value of the function
Because power system voltages and currents are nominally sinusoidal, the nominal crest factor for these would
be 2, which is 1.414 (see Electric Power Fundamentals section (section 2) for details).
Notch Area: A notch in the power system voltage (or current) is illustrated in figure 11-1 [2]:
Figure 11-1: Voltage (or current) notch illustration
The notch area for the notch as illustrated in figure 11-1 is defined as:
(11-5)
where
An
t
d
is the notch area in volt-microseconds

is the notch time duration in microseconds
is the notch depth in volts
Recovery time: This is the time needed for the output voltage or current to return to a value within the regulation
specification after a step load or line change.
Displacement Power Factor: The ratio of the active power of the fundamental wave, in watts, to the apparent
power of the fundamental wave, in volt-amperes. This is the traditional definition of power factor.
Total Power Factor: The ratio of the total input power, in watts, to the total volt-ampere input. This includes the
effects of harmonics.
K Factor: A measure of a transformers ability to serve non-sinusoidal loads. The K factor is defined as:
(11-6)
where
Ih
h
hmax
is the harmonic component at h times the fundamental frequency

is the harmonic order of Ih in multiples of the fundamental frequency
is maximum harmonic order present
Voltage surges
The causes of voltage surges may be split into two major categories: Power system switching and environmental
[1]. Both exhibit decaying oscillatory transients. Capacitor switching close to the point under consideration is the
most common cause of switching surges, while lightning is the most common cause of environmentally-induced
voltage surges. Both can cause severe damage to unprotected power system components, with the potential for
lightning damage being the most severe; in the worst case, lightning damage can be catastrophic.
Surge arrestors, as described in Section 7, are typically used to protect against voltage surges. On low voltage
systems transient voltage surge suppressors (TVSS), also described in Section 7 are also used. For motors,
surge capacitors are an option. In severe cases, custom-designed R-C snubber circuits may be required as well.
Voltage sags, swells and interruptions

Voltage sags, swells and interruptions have many causes. Remote switching or lightning strikes can cause voltage
swells, as can the recovery of the system voltage after a fault. Voltage sags can be caused due to transformer or
motor inrush or large step loads, especially on systems without large amounts of available fault current. Voltage
interruptions are generally caused due to protective device operation.
Protection of sensitive equipment against voltage sags and swells can be difficult. Fast-acting voltage regulators
offer the one means of defense against these phenomena, although any voltage regulator must be properly
applied to avoid worsening the problem. Fast-acting voltage regulators can generally be classified as tapswitching, buck-boost, or ferroresonant (also known as CVT constant voltage transformer) types [1]. New solidstate tap switching technologies for voltage regulators provide faster response than older, electromechanical
switching technologies. Other devices, such as power line conditioners which combine some TVSS functions
with voltage regulation and noise reduction, and motor-generator sets, are also used [1].
Protection of sensitive loads against voltage interruptions is best performed with an uninterruptible power supply
or UPS. This device is available in several different topologies and is crucial where microprocessor-based devices
are to be powered. UPSs are discussed in more detail in a later section of this guide.
Harmonic distortion
Harmonic distortion is a subject of great interest in modern power systems. Harmonic distortion results from
non-sinusoidal load currents. These currents are the result of non-linear loads, such as drives, which employ
power electronic devices to rectify the AC waveform. These devices draw non-sinusoidal currents which, in turn,
cause non-linear voltages to be developed in the system.
IEEE Standard 519-1992 [2] gives recommended limits for current distortion due to consumer loads and voltage
distortion in the utility supply voltage. Both are referenced at the point on the utility system where multiple
customers can be served, referred to as the Point of Common Coupling (PCC). The requirements from [2] for
current distortion limits on general distribution systems 120 V - 69 kV are given in table 11-2. Table 11-3 shows
the corresponding utility voltage distortion limits.
Note that the current limits are given both as limits on the individual harmonic levels and a limit on the TDD, and
that as the ratio Isc/IL increases the limits also increase. The reason for this is that the current distortion limits are
designed to limit the voltage distortion at the PCC, and the voltage distortion for a given current distortion worsens
with a larger source impedance (V - I Z).
Table 11-2: IEEE 519-1992 Harmonic current distortion limits for general distribution
systems 120 V through 69 kV (essentially same as [2] table 10-3)
Maximum harmonic current distortion in percent of IL
Individual harmonic order (Odd harmonics)
Isc/IL
<11
11<h<17
17h<23
23h<35
35h
TDD
<20*
4.0
2.0
1.5
0.6
0.3
5.0
20<50
7.0
3.5
2.5
1.0
0.5
8.0
50<100
10.0
4.5
4.0
1.5
0.7
12.0
100<1000
12.0
5.5
5.0
2.0
1.0
15.0
>1000
15.0
7.0
6.0
2.5
1.4
20.0
Even harmonics are limited to 25% of the odd harmonic limits above.
Current distortions that result in a DC offset, e.g. half-wave converters, are not allowed.
*All power generation equipment is limited to these values of current distortion, regardless of actual ISC/IL
where
ISC
IL
= maximum short-circuit current at PCC

= maximum demand load current (fundamental frequency component) at PCC
Table 11-3: IEEE 519-1992 Harmonic voltage distortion limits (essentially same as [2]
table 11-1)
Individual voltage distortion (%)
THDVn (%)
69 kV and below
3.0
5.0
69.001 kV through 161 kV
1.5
2.5
161.001 kV and above
1.0
1.5
Bus voltage at PCC
Note: High voltage systems can have up to 2.0% THD where the cause is an HVDC terminal that will attenuate by the time
it is tapped for a user
Mitigation of harmonic distortion is generally accomplished by one of the following means:

I
Passive tuned filters
Use of phase multiplication on power conversion equipment
Active filters
Passive tuned filters are simple series L-C filters. A single tuned passive filter can effectively mitigate one
harmonic frequency. They are generally tuned to a value below the harmonic frequency to be attenuated to avoid
a resonance condition at that frequency. These are custom-engineered solutions that must be designed
specifically for the circuit in question. Passive filters are also used for power factor correction. However, there is
5
a limit to their effectiveness and if higher-order harmonics must be attenuated their use is generally not
cost-effective. Care must be taken in all cases to balance the harmonic and power factor correction
considerations.
Phase multiplication operates on the principle that if m six-pulse rectifiers are shifted 60/m degrees from each
other, are controlled by the same delay angle, and are loaded equally, the only harmonics present will be:
(11-7)
where
h
q
k
is a harmonic order present

= 6m and is known as the pulse number of the circuit
is any integer
Thus, for standard 6-pulse rectifiers the harmonic orders present will be 5, 7, 11, 13,, etc. 18-pulse rectifiers are
the current state-of-the-art; for an 18-pulse rectifier (m=3), the harmonic orders present are 17, 19, 35, 37, , etc.
For an 18-pulse converter, the lower-order harmonics are thus eliminated. For systems with large numbers of
phase-multiplied converters the harmonic current limits in table 10-3 are increased by the factor (q/6)1/2,
where q is the pulse-number of the predominate non-linear load on the system. In this case the limits for the
harmonic orders that do not fit equation (11-7) for the q of the predominate non-linear load are multiplied by a
factor of 0.25. Phase-shifting transformer connections are used to achieve the 60/m degree phase shift between
6-pulse rectifier units.
Active filtering technology is a still-evolving art. Current state-of-the-art designs measure the current, filter out the
fundamental frequency of the measured current, and inject current that is the negative of the result into the
system to cancel the harmonics up to a given harmonic order. These systems are generally used in existing
installations that have existing 6-pulse drives where replacing the drives is not a cost-effective solution, or where
multiple smaller 6-pulse drives are utilized since phase multiplication for a drive below 100hp is generally not costeffective. State-of-the-art units can also dynamically correct the power factor, and are advantageous vs. passive
filters both in their effectiveness and their flexibility in power factor correction.
Power quality monitoring

Power quality monitoring is vital when sensitive equipment is to be powered, and also for the over-all reliability of
the system. Microprocessor-based technology allows the most common power-quality instrumentation to be
combined into a single monitoring device which incorporates wave-form capture, measures of the power quality
metric values per the above discussion, and conventional current, voltage, power, and energy measurements, with
min/max logging capabilities. These devices are typically true RMS-reading instruments, with measurements up to
a given harmonic (typically 31st harmonic or higher).
The inclusion of power monitoring equipment in the initial power system design will make diagnosis of any
subsequent power quality issues, should they arise, much easier and more efficient. Reference [3] contains much
information on power quality monitoring and should be consulted for further reference.
References
[1]
IEEE Recommended Practice for Powering and Grounding Electronic Equipment, IEEE Std. 1100-1999,
March 1999.
[2]
IEEE Recommended Practices and Requirements for Harmonic Control in Electrical Power Systems,
IEEE Std. 519-1992, June 1992.
[3]
IEEE Recommended Practice for Monitoring Electric Power Quality, IEEE Std. 1159-1995.
Section 12: Arc
Flash Hazard Considerations
Introduction
The consideration of arc flash hazards is a relatively new concern for power system design. However, it is a
concern that is rapidly gaining momentum due to increasingly strict worker safety standards and system reliability
requirements that demand work on live electrical equipment.
Background
Electrical arcs form when a medium that is normally an insulator, such as air, is subjected to an electric field
strong enough to cause it to become ionized. This ionization causes the medium to become a conductor which
can carry current. The phenomenon of electrical arcing is as old as the world itself. Lightning is a natural form of
electrical arc. Man-made electrical arcs exist in devices such as arc furnaces. However, utilization of electrical
energy invariably requires equipment where unintentional arcing between conductors becomes a possibility.
Electric arcs in equipment liberate large amounts of uncontrolled energy in the form of intense heat and light.
Unintentional arcing in power equipment can impose several different types of hazards:
I
Heat from arc can cause severe flash burns many feet away (temperatures can reach 20,000 K, four times the
temperature at the surface of the sun!).
Byproducts from the arc, such as molten metal spatter, can cause severe injury.
Pressure wave effects caused by the rapid expansion of air and vaporization of metal can distort enclosures and
cause doors and cover panels to be ejected with severe force, injuring personnel.
Sound levels can damage hearing.
Figure 12-1 gives an indication of the amount of uncontrolled energy an arc can contain, as seen by the amount
of damage to the equipment shown.
Electrical safety has traditionally been concerned only with electric shock hazards. The recognition of arc flash
hazards began formally in 1981 with a paper The Other Electrical Hazard: Arc Blast Burns [5] by Ralph Lee,
presented at the 1981 IEEE IAS Annual Meeting. This paper established theoretical modeling for the heat energy
incident upon a surface a given distance from the arc. Subsequent developments followed over the next 20 years,
including testing to develop more accurate empirical calculation methods and to evaluate protective clothing.
Figure 12-1: Example of arcing damage to equipment
At the time of publication, there are two basic standards which establish requirements for arc flash hazards. The
first is NFPA 70E, Standard for Electrical Safety in the Workplace [1], which defines the basic practices to be
followed for electrical safety, including protective clothing levels which must be worn for given levels of arc flash
incident energy and what steps must be taken prior to live work on electrical equipment. The second is the IEEE
Guide for Performing Arc-Flash Hazard Calculations, IEEE 1584-2002 [2] which gives the engineer the methods
for calculating the severity of arc flash incident energy levels. The NEC [3] requires only that certain equipment
(switchboards, panelboards, industrial control panels, meter socket enclosures, and motor control centers in other
than dwelling occupancies and likely to require examination, adjustment, servicing, or maintenance while live) be
field marked to warn qualified persons of potential electric arc flash hazards.
NFPA 70E requirements for flash hazards

NFPA 70E [1] is divided into four chapters: Safety Related Work Practices (Chapter 1), Safety Related
Maintenance Requirements (Chapter 2), Safety Requirements for Special Equipment (Chapter 3), and Installation
Safety Requirements (Chapter 4). The discussion here is centered upon Chapter 1.
Several terms are of particular importance when discussing arc flash hazards [1]:
Flash Hazard: A dangerous condition associated with the release of energy caused by an electric arc.
Incident Energy: The amount of energy impressed on a surface, a certain distance from the source, generated
during an electrical arc event. One of the units used to measure incident is calories per square centimeter
(cal/cm2).
Flash Hazard Analysis: A study investigating a workers potential exposure to arc-flash energy, conducted for
the purpose of injury prevention and the determination of safe work practices and appropriate levels of PPE.
Live Parts: Energized conductive components.
Exposed (as applied to live parts): Capable of being inadvertently touched or approached nearer than a safe
distance by a person. It is applied to parts that are not suitably guarded, isolated, or insulated.
Shock Hazard: A dangerous condition associated with the possible release of energy caused by contact or
approach to live parts.
Flash Protection Boundary: An approach limit at a distance from exposed live parts within which a person could
receive a second degree burn if an electrical arc flash were to occur.
Limited Approach Boundary: An approach limit at a distance from an exposed live part within which a shock
hazard exists.
Restricted Approach Boundary: An approach limit at a distance from an exposed live part within which there is
an increased risk of shock, due to electrical arc over combined with inadvertent movement, for personnel working
in close proximity to the live part.
Prohibited Approach Boundary: An approach limit at a distance from an exposed live part within which work is
considered the same as making contact with the live part.
Qualified Person: One who has skills and knowledge related to the construction and operation of the electrical
equipment and installations and has received safety training on the hazards involved.
Working On (live parts): Coming in contact with live parts with the hands, feet, or other body parts, with tools,
probes, or with test equipment, regardless of the personal protective equipment a person is wearing.
Working Near (live parts): Any activity inside the Limited Approach Boundary.
Electrically Safe Work Condition: A state in which the conductor or circuit part to be worked on or near has
been disconnected from energized parts, locked/tagged in accordance with established standards, tested to
ensure the absence of voltage, and grounded if determined necessary.
NFPA 70E [1] chapter 1 covers personnel responsibilities (both the employer and the worker have specific
responsibilities for safety), training requirements, the establishment of an electrical safety program, and the
establishment of an electrically safe working condition. These will not be discussed in detail here, but the reader is
strongly encouraged to refer to the NFPA 70E [1] to become more familiar with them as they are important topics.
For arc flash hazard considerations, the focus is on Article 130, Working On or Near Live Parts. The basic
requirement is that live parts over 50 V to ground to which an employee might be exposed should be put into an
electrically safe work condition prior to working on or near them, unless the employer can demonstrate that deenergizing introduces additional or increased hazards or is infeasible due to equipment design or operational
limitations. In this case live work requires an Energized Electrical Work Permit, for which the requirements are
given in Article 130.1 (A) (2). Some exemptions are given to the requirement for an electrical work permit, such as
testing, troubleshooting, etc., performed by qualified persons.
The approach boundaries to live parts are defined above, and are illustrated in figure 12-2. These form a series of
boundaries from an exposed, energized electrical conductor(s) or circuit part(s). The requirements for crossing
these become increasingly restrictive as the worker moves closer to the exposed live part(s). The limited,
restricted, and prohibited approach boundaries are shock protection boundaries and are defined in NFPA 70E
table 130.2 (C) [1]. Qualified persons can approach live parts 50V or higher up to the restricted approach
boundary, and can only cross this boundary if they are insulated or guarded and no uninsulated part of the body
crosses the prohibited approach boundary, if the person is insulated from any other conductive object, or if the live
part is insulated from the person and from any other conductive objects at a different potential. Unqualified
persons must stay outside the limited approach boundary unless they are escorted by a qualified person.
Unqualified persons cannot cross the restricted approach boundary.
A flash hazard analysis must be performed in order to protect personnel from the possibility of injury due to arc
flash. This analysis must set the flash protection boundary, which for voltages below 600 V is equal to 4 ft. based
upon a clearing time of 0.1 second and a bolted fault current of 50 kA (5000 Ampere-seconds) or, where the
clearing time x bolted fault is greater than 5000 ampere seconds or under engineering supervision, may be
calculated with the equations given in the NFPA 70E text. For voltages over 600V, the flash protection boundary is
defined as the distance from the potential arc which has an incident energy of 1.2 cal/cm2, or 1.5 cal/cm2 if the
clearing time is 0.1 second or faster. The means of calculating the arc flash protection boundary for voltages 600V
or less is based upon the theoretical Lee method developed in [5]. The method for calculating the arc flash
incident energy for a given working distance from live parts is not specified in NFPA 70E code text itself; several
methods are given in Annex D of NFPA 70E. The preferred methods for performing these calculations are given in
IEEE 1584 [2], as detailed below. The option is also given to use pre-prepared tables given in NFPA 70E based
upon given levels of fault current and protective device clearing time to select personal protective equipment in
lieu of a formal arc flash study.
Figure 12-2: Approach boundaries, from [1]
The classifications for personal protective equipment (PPE)for arc flash protection are given in NFPA table 130.7
(C)(11), reproduced below as table 12-1. PPE for arc flash protection is given an Arc Rating in cal/cm2, which
must be compared to the arc flash incident energy for the location in question to select the proper clothing.
Employees working within the flash protection boundary must wear nonconductive head protection wherever there
is a danger of head injury from electric shock or burns or from flying objects resulting from electrical explosion.
Face, neck, chin and eye protection must be worn wherever there is a danger of injury from electric arcs or
flashes or from flying objects resulting from electrical explosion. Body protection, in the form of flame-retardant
(FR) clothing as defined in table 11-1, must be worn where there is possible exposure to arc flash incident energy
levels above 1.2 cal/cm2; an exception allows Category 0 clothing to be worn for exposures 2 cal/cm2 or lower. An
example of a full flash suit is shown in figure 12-3.
Table 12-1: Protective clothing characteristics (essentially the same as [1]

table 130.7 (c) (11))
Hazard/risk category
Clothing description
Required minimum arc rating of PPE

(cal/cm2)
Non-melting, flammable materials (i.e.,

untreated cotton, wool, rayon, or silk, or
blends of these materials) with a fabric weight
of at least 4.5 oz/yd2
N/A
FR shirt and FR pants or FR coverall
Cotton underwear conventional short sleeve

and brief/shorts, plus FR shirt and FR pants
Cotton underwear plus FR shirt and FR pants

plus FR coverall, or cotton underwear plus two
FR coveralls
25
Cotton underwear plus FR shirt and FR pants

plus multilayer flash suit
40
Figure 12-3: Example of a full flash suit
IEEE 1584
IEEE 1584 [2] is the guide for determining arc flash incident energy levels and protection boundaries. It contains
an empirical calculation method based upon extensive test results using a Design-of-Experiments (DOE) method,
resulting in a 95% confidence level that the arcing fault current will be higher than calculated. In situations where
the empirical method does not apply, the Lee method from [5] is recommended, and is described in IEEE 1584.
IEEE 1584 only takes into account the heat of an arc, and not the secondary effects such as molten metal spatter
and pressure-wave effects.
A.) IEEE 1584 empirical method

This method is valid for the following systems with the following characteristics:
Voltages in the range of 208 V-15 kV, three phase
Frequencies of 50 Hz or 60 Hz
Bolted fault current in the range of 700 A-106 kA
Grounding of all types and ungrounded
Equipment enclosures of commonly available sizes
Gaps between conductors of 13mm-152mm
Faults involving three phases in applying the empirical method it is assumed that a phase-to-ground fault will
escalate into a phase-to-phase fault
The first step in this method is to determine the predicted arcing fault current using the following equation for
system voltages less than 1000 V [2]:
(12-1)
For system voltages 1000 V or greater, the following equation is used [2]:
(12-2)
where
Ia
K
Ibf
V
G
is the arcing fault current in kA

= -0.153 for open configurations and -0.097 for box configurations
is the bolted fault current for three-phase faults in kA
is the system voltage in kV
is the gap between conductors in mm
The arcing fault current will typically be 40-60% of the bolted fault current for systems 1000 V or less, and 90-95%
of the bolted fault current for systems greater than 1000 V.
The arcing fault current is then used to find the clearing time for the overcurrent protective device which clears the
fault. Care must be taken to identify which device actually clears the fault. The clearing time then becomes the
arcing time for the purpose of finding the incident energy.
The normalized incident energy, referenced to a working distance of 610mm and an arcing time of 0.2 seconds,
is then calculated using the following equation [2]:
(12-3)
where
En
K1
K2
G
is the normalized incident energy

= -0.792 for open configurations and -0.555 for box configurations
= 0 for ungrounded and high-resistance grounded systems and -0.113 for grounded systems.
is the gap between conductors in mm
Now, using the actual working distance and arcing time, the incident energy is calculated as [2]:
(12-4)
where
Cf
En
t
D
x
= 1.0 for voltages above 1 kV, and 1.5 for voltages below 1 kV
is the incident energy in cal/cm2
is the arcing time per above
is the working distance in mm
is a distance exponent from [2] table 4
5
Table 4 in [2] gives the distance exponents, along with typical gaps between conductors, for different voltage
levels and equipment types.
The flash protection boundary may be found using the following equation [2]:
(12-5)
where
DB
EB
is the boundary distance in mm

is the incident energy level at the boundary, in cal/cm2.
From [1] EB must be 1.2 cal/cm2 unless the voltage is above 600 V and the clearing time is 0.1 s or faster, in
which case it may be increased to 1.5 cal/cm2. However, equation (12-5) may be used to calculate the boundary
for any incident energy level, for example, to calculate the boundaries where different categories of PPE per table
11-1 may be worn. Note that the larger of the boundaries as calculated from IEEE 1584 or NFPA 70E should be
used in order to satisfy the NFPA 70E requirements.
Note that the incident energy is proportional to the arcing time, which is set by the overcurrent protective device
time-current characteristic and the arcing current level. Because overcurrent protective device tripping times are
lower for larger currents due to inverse time-current characteristics, this is an important point. Larger bolted fault
currents lead to larger predicted arcing fault currents, which lead to generally lower values of arc flash incident
energy. Lower bolted fault currents lead smaller predicted arcing fault currents, which lead to generally higher
values of incident energy.
For conservatism, a second predicted arcing fault current is calculated at 85% of the value per equation (11-1) or
(11-2), and the result is used to calculate a second value for the incident energy and flash protection boundary.
The larger of the incident energy/protection boundary values are used as the final result. If the overcurrent
protective device time-current characteristic is horizontal, such as for the instantaneous characteristic of an
electronic-trip circuit breaker, the two values will be equal since the arcing time will not change.
B.) Lee method

Where the IEEE 1584 empirical method cannot be used due to being outside the limits of applicability as defined
above, the theoretically-derived Lee method per [4] may be used. This is based upon maximum power transfer
and is very conservative above 15 kV. To calculate the incident energy with this method, the following equations
are used [2]:
(12-6)
(12-7)
C.) Simplified device equations

Further testing was performed for circuit breakers and current-limiting fuses, and simplified equations of the form
( A+Blog Ibf ) were developed. These are given in [2]. The equations for fuses are applicable within the bolted fault
current ranges given in [2]. The equations for circuit breakers will yield conservative results and should only be
used when they are within the ranges of applicability given in [2] and where nothing else about a particular circuit
breaker is known.
Manufacturers also publish device-specific equations for certain devices, such as fuses and some
high-performance circuit breakers. These are preferred vs. the IEEE 1584 Empirical Method since they will more
accurately model the arc-flash performance of a given device.
Application guidelines
A.) Arc flash calculations
The following guidelines are helpful when performing arc flash calculations [3]:
I
When choosing a calculation method, be sure the system conditions fall into the calculation methods range
of applicability.
Use the newest methods given in IEEE 1584-2002. Older methods given in previously-published papers are
superseded by this standard.
If the manufacturer publishes device-specific equations, use them.
Use realistic fault current values. The actual minimum available fault current, rather than the worst-case values
typically used for short-circuit analysis, give more conservative (and realistic) results.
Consider the effects of arc fault propagation to the line side of the main overcurrent device when determining
which device should be used to calculate the arcing time. For example, for the electrical panel in figure 12-4,
device A would be used rather than device B for calculating the arcing time for a fault on the panelboard bus,
since the fault can propagate to the line side of device B. Similar considerations should be made for
switchboards, MCCs, etc.
Figure 12-4: Example electrical panel
Quantify the variables. The working distance, bus gap, equipment configuration, and system grounding are all
dependent upon the particular installation and must be accurately determined.
Be aware of motor contribution. Motor contribution can both increase and decrease the arc flash incident energy,
depending upon where in the system the arcing fault occurs.
Use a computer for analysis. This is the most efficient way to accurately calculate the incident energies and flash
protection boundaries where multiple sources, such as generation and motor contribution, must be taken into
account. Several commercial software packages are available for arc flash hazard analysis. Be aware, though,
what the user-configurable options for the software are and be sure they are set correctly for accurate results.
B.) System design

Arc flash hazard analysis is typically performed after the system design process, including the time-current
coordination study, is complete. This can result in the need for tweaking of overcurrent protective device
settings to obtain acceptable arc flash results or, in the worst case, system re-design with additional equipment.
The following guidelines, if observed during the system design phase, can serve to minimize the need for
such activities:
Use a dedicated main overcurrent device at transformer secondaries. The secondary of a transformer is one of
the most difficult places to achieve acceptable arc flash hazard levels. If multiple mains are used for transformer
secondaries, the arc flash hazard level downstream from the main but ahead of the feeders must be calculated
using the transformer primary device timing characteristics, significantly increasing the incident energy. If the
secondary main and feeders are in the same switchboard or panel, this will usually not be applicable due to arc
fault propagation to the line side of the main device as described above. For ANSI low voltage switchgear per
ANSI C37.20.1, however, this can be of real benefit, as well as in cases where the secondary overcurrent device
is remote from the feeders.
Closely coordinate devices where possible. The lower the clearing time for the predicted arcing current, the
lower the arc flash incident energy.
Use high-performance devices, such as low-arc-flash circuit breakers, where possible. These will significantly
reduce the arc flash incident energy.
Use bus differential protection and/or zone selective interlocking where possible. This is high-speed protection
that can significantly lower the arc flash incident energy.
References
[1]
Standard for Electrical Safety in the Workplace, NFPA 70E, The National Fire Protection Association,
2004 Edition.
[2]
IEEE Guide for Performing Arc Flash Hazard Calculations, IEEE 1584-2002, September 2002.
[3]
[4]
A. C. Parsons, Arc Flash Application Guide Arc Flash Energy Calculations for Circuit Breakers and Fuses,
Square D/Schneider Electric Engineering Services, August 2004.
[5]
Lee, R., The Other Electrical Hazard: Electrical Arc Blast Burns, IEEE Transactions on Industry Applications,
vol. 1A-18, no. 3, May/June 1982.
Section 13:
Utility Interface Considerations
Introduction
The vast majority of industrial and commercial facilities are served from public utilities. However, the utility
interface is often the most neglected aspect of system design. This is especially true at the medium voltage level.
Often, the service equipment manufacturer is expected to resolve issues that severely impact the design of the
system. This can result in unexpected costs and project delays. These issues should be addressed during the
system design stage, where the impacts to system reliability and cost can be adequately managed; only by
knowing the utilitys requirements is this possible.
The utilitys jurisdiction

Because utilities must serve multiple consumers, they must take the steps they consider necessary to ensure
reliable service over their entire system. Because of this, most utilities impose requirements on the design of the
systems to which they supply power.
Those elements of the system design over which the utility has jurisdiction vary from utility to utility. The utility
always dictates which service voltages are available for a given size of service. The utility usually has some
jurisdiction over the service disconnect and service overcurrent protection. Certainly, the utility has jurisdiction over
(and usually the only access to) their revenue meters and metering instrument transformers. However, in some
cases the utility will require jurisdiction over the entire service equipment, and can impose requirements upon
system protection, equipment control power, and other parts of the system design. In some cases, the over-all
arrangement of the system itself, including emergency/standby power systems, may be dictated by the utility.
Because in most cases the utility is the sole service provider for a given region, negotiating these requirements is
usually not feasible. Therefore, knowledge of the utilitys requirements is vital to successful, on-time, on-budget
system design and construction.
Utility service requirements standards

Each utility typically maintains its own series of standards for individual consumer service requirements.
Such requirements are often published in the form of a service requirements handbook or similarly-titled
publication. The format of the standards, and the standards themselves, vary from utility to utility. This can be
challenging to those engineers who design industrial and commercial facilities in different areas, and to
equipment manufacturers.
In recognition of this issue, EUSERC (Electric Utility Service Equipment Requirements Committee) was formed
in 1983, combining southern-California-based PUSERC and northern-California-based WUESSC, which were
older organizations formed in 1947 and 1950, respectively. The purposes of EUSERC are to promote uniform
electric service requirements among its member utilities, to publish existing utility service requirements for electric
service equipment, and to provide direction for development of future metering technology. EUSERC publishes a
manual [1] which delineates requirements for electric service equipment through 34.5 kV. At the time of
publication, 80 utilities from 12 states are involved with EUSERC. While EUSERC does not eliminate the need
for individual utility requirements, it does help a great deal in making electrical service equipment more
standardized and less costly.
System topology snf protection

Requirements for the system topology are designed to increase both the reliability of the over-all utility system and
with the reliability of service to the installation in question. These requirements typically take the following forms:
I
Restrictions on the size of services
Restrictions on, or requirements for, normal and alternate services and transfer equipment between the two
Restrictions or requirements for the configuration of emergency and standby power systems
Restrictions on the types of service disconnecting devices allowed
Restrictions on the types of service overcurrent protection allowed
Requirements for service cable compartments in service equipment
Requirements or restrictions on the number and types of protective relaying
Requirements for the service switchgear as a whole
The most common requirement, which is applied to virtually every utility installation, is that the service
overcurrent device must coordinate with the upstream utility overcurrent device, typically a recloser or utility
substation circuit breaker. If there is standby power on the premises, the utility will typically require that
paralleling the alternate power source with the utility source not be possible unless stipulated in the rate
agreement for the service in question.
Requirements for restricted access to service cable termination and service disconnect compartments in the
service switchgear are another common. In some cases these must be in a dedicated switchgear or switchboard
section, increasing the service equipment footprint. In many cases grounding means must be provided with the
equipment to allow the utilitys preferred safety grounding equipment to be installed. In some cases, requirements
may be imposed on the entire service switchgear, such as electrical racking for circuit breakers or barriers that are
not standard for the equipment type used.
In some cases the control power for the service switchgear, such as a battery, must be designed to the
utilitys specifications.
Additional protective relaying may be required to prevent abnormal conditions which, although not harmful to the
system being served, affect the reliability of the utility system. In some cases the makes and models of protective
relays for the service overcurrent protection are restricted to those the utility has approved.
Revenue metering requirements

Often the utilitys revenue metering requirements can have an effect the over-all system topology. There are two
basic utility revenue metering arrangements:
Hot-Sequence Metering: The metering instrument transformers are placed ahead of the service disconnect.
Cold-Sequence Metering: The metering instrument transformers are placed on the load side of the
service disconnect.
With hot-sequence metering, the instrument transformers and meters may be placed on the last distribution pole
for overhead services, or in a dedicated utility-supplied metering compartment outside the facility to be metered
for underground services. In these cases, the effect of the utilitys instrument transformers and meters on the
over-all design for the facility power system and equipment is usually minimal. However, in many cases the
end-user, at their expense, must supply a utility instrument transformer compartment which houses the
instrument transformers. The design requirements for these compartments are often detailed, and are present to
insure that no tampering occurs with the instrument transformers or meters. These compartments typically take an
entire section, or part of a section, of the service switchgear or a switchboard, increasing the footprint of this
equipment. In some cases, the service equipment must provide housing for the meters as well, along with
convenient access for the utilitys personnel. The utility typically provides and installs the instrument transformers
and meters, although a few utilities require the end-user or equipment manufacturer to install these. In extreme
cases the end-user must supply the instrument transformers and send them to the utility for testing. Identifying the
requirements early in the design process helps to insure that all parties are aware of the costs involved.
Utility revenue metering instrument transformers for services up to 600 V typically consist of two or three current
transformers depending upon the system configuration, unless the service is small enough to be directly metered.
In some cases voltage transformers may be required as well. Both the current and voltage transformers are
designed for metering, with the current transformers typically being bar or wound-primary type. For services over
600 V, both voltage and current transformers are required, either two or three of each depending upon the system
configuration. In some cases the utility will not allow the voltage transformers to be fused.
Additional regulatory requirements

In some cases there may be additional state regulatory requirements which apply. These are typically concern
distributed generation and may severely restrict or otherwise impact the system design. These requirements must
be fully understood before the system design is begun to avoid expensive changes later in the process. The
Public Service Commission or similar governmental regulatory agency for the region in question typically controls
these requirements.
Utility information required for system design

In designing the power system for any commercial or industrial facility the following information is crucial to
adequate system design:
I
Nominal service voltage.
Maximum available fault current and associated X/R ratio.
Minimum available fault current.
Data on the utilitys nearest upstream protective device (device type and ratings, relay type and
settings if applicable).
Latest edition of the utilitys service handbook or similar publication.
Latest edition of additional state regulatory requirements, if applicable.
Contact information for utilitys system engineer or equivalent for the region in question.
Utility rate agreement, if available.
All of these, except items 6 and 8, should be available from the serving utility. Item 6 should be available from the
regional Public Service Commission or similar governmental regulatory agency. Item 8 may not be available at the
outset, but should be taken into consideration as soon as it becomes available.
References
[1]
EUSERC Manual, Electric Utility Service Equipment Requirements Committee, 2005 Edition
Section 14:
Electrical Energy Management
Introduction
Electricity is a powerful form of energy that is essential to the operation of virtually every facility in the world.
It is also an expensive form of energy that can represent a significant portion of a manufacturing facilitys cost
of production.
This energy management primer is intended to introduce some electricity billing fundamentals, especially focusing
on the two major aspects of the electric bill, demand and energy. This section also highlights key aspects of
identifying energy-saving opportunities among major industrial processes and equipment.
Electricity billing basics

Most electric utilities serve a designated geographic territory, largely without other competitors having access
to their customers. As such, utility prices have often been set by local, state, or federal regulators, entities
that review electric utility costs, revenues, investment decisions, fuel prices, and other factors to arrive at a
target rate of return. This approved rate of return, coupled with the utilitys cost structure, determine prices
customers will pay.
These prices are established in electric utility tariffs, or rate schedules. Rate tariffs are usually established
for different classes or sizes of customers. Common class types may include industrial, commercial, residential,
municipal, and agricultural. Each customer class may have one or more rate schedules available, and it is
common for the electric utility to allow a facility to choose the rate schedule within its class that offers the
lowest price.
I
Electricity metering: Electric utilities meter both the real and reactive power consumption of a facility. The real
power consumption, and its integral energy, usually comprise the largest portion of the electric bill. Reactive
power requirements, usually expressed in power factor, can also be a significant cost and will be discussed later.
Demand: Real power consumption, typically expressed in kilowatts or megawatts, varies instantaneously over
the course of a day as facility loads change. While instantaneous power fluctuations can be significant, electric
utilities have found that average power consumption over a time interval of 15, 30, or 60 minutes is a better
indicator of the demand on electrical distribution equipment.
Transformers, for example, can be selected based on average power requirements of the load. Short-duration
fluctuations in load current may cause corresponding drops in load voltage, but these drops are within the
normal operating tolerances of typical machines and within the design parameters of the transformer.
The demand rate, in $/kW, may also be referred to as a capacity charge, since it has historically been related to
the necessary construction of new generating stations, transmission lines, and other utility capital projects.
Demand charges often represent 40% or more of an industrial customers monthly bill.
INSTANTANEOUS
POWER
POWER (KILOWATTS)
Area under curve = ENERGY (KW h)

Area under line = ENERGY (KW h)
(For the interval defined)
DEMAND
DEMAND
DEMAND
INTERVAL
DEMAND
INTERVAL
DEMAND
INTERVAL
DEMAND
INTERVAL
TIME
Demand for each interval = Average Power over that interval
Demand is the average instantaneous power consumption over a set time interval,
usually 15, 30, or 60 minutes.
I
Energy: The other major component of an electric bill is energy. The same metering equipment that measures
power demand also records customer energy consumption. Energy consumption is reported in kilowatt-hours or
megawatt-hours. Unlike power demand with its capacity relationship, customer energy consumption is
sometimes related to fuel requirements in electric utility generating stations. The cost per kilowatt-hour in a
given electric utility rate structure, therefore, is often influenced by the mix of generating plant types in the
utility system. Coal, fuel oil, natural gas, hydroelectric, and nuclear are typical fuel sources on which power
generation is based.
Load factor Demand/energy relationship: One useful parameter to calculate each month is the ratio of
the average demand to the peak demand. This unit-less number is a useful parameter that tracks the
effectiveness of demand management techniques. A load factor of 100% means that the facility operated at
the same demand the entire month, a so-called flat profile. This type of usage results in the lowest unit cost
of electricity.
Few facilities operate at a load factor of 100%, and that is not likely to represent an economical goal for most
facilities. But a facility can calculate its historical load factor, and seek to improve it by reducing usage at peak
times, moving batch processes to times of lower demand, and so forth. Load factor can be calculated from
values reported on practically every electric bill:
LF = kWh/(kW * days * 24);
Where LF is Load Factor, kWh is the total energy consumption for the billing period, kW is the peak demand set
during the billing period, and days is the number of billing days in the month (typically 28-32). 24, of course is
the number of hours in a day.
Time-of-Use customers may prefer to track load factor only during on-peak time periods. In that case, the kWh,
kW, days, and hours/day in the formula are changed to reflect the parameters established only during the onpeak periods.
Typical load factor for an industrial facility depends to a great degree on the number of shifts the plant operates.
One shift, five-day operations typical record a load factor of 20-30%, while two-shifts yield 40-50%, and three
shift, 24/7 facilities may reach load factors of 70-90%.
T hree S hifts
One S hift
Demand, kW
Demand, kW
Equal Energy
Unequal Demand
Graphical comparison of facilities with dramatically different load factors. The three shift facility produces an average demand that is nearly equal to its peak demand, while the average and peak
demand for the one shift facility is much less than one.
30%
50%
Peak Demand, kW
Load Factor:
1142
685
70%
489
Energy Usage, kWh
250,000
250,000
250,000
Demand Cost
$11,420
$6,850
$4,890
Energy Cost
$10,000
$10,000
$10,000
Total Monthly Bill
$21,420
$16,850
$14,890
Average Cost/kWh
8.57
6.74
5.96
Demand Cost As
Percent of Total
53%
41%
33%
Power factor: The relationship of real, reactive, and total power has been introduced previously, and described
as the power triangle. For effective electricity cost reduction, it is important to understand how the customers
electric utility recoups its costs associated with reactive power requirements of its system. Many utilities include
power factor billing provisions in rate schedules, either directly in the form of penalties, or indirectly in the form of
real-power billing demand that is higher than the actual measured peak.
Even if a utility does not charge directly for poor power factor, there are at least three other reasons that a
customer may find it economical to install equipment to improve power factor within its facility, thereby reducing
the reactive power requirements of the utility. PowerLogic Solutions, volume 1, issue 4 (www.powerlogic.com)
describes each of these cost-reduction opportunities in considerable detail.
N Reduce power factor penalties
N
Release capacity of an existing circuit
Reduce heating losses associated with power distribution (often called I2R losses)
Improve voltage regulation
Typical energy auditing process:

N Evaluate the current rate schedule
N
Determine if other rate schedules are available
Complete the Facility Energy Profile
Assess no-cost/low-cost energy saving options
Complete feasibility analysis of energy management project options
Recommend Energy Action Plan
Facility energy profile Wheres the energy going?

An important initial step in evaluating energy saving opportunities is to estimate both:
I
The contribution to peak billing demand, and
The amount of energy consumption
Of each major load or process within the facility being evaluated.

This Facility Energy Profile helps to focus the energy optimization efforts on those processes or loads that have
the most savings potential. This Profile also may identify batch processes or discretionary loads that may be
scheduled at times of low demand for the rest of the facility, or during times of off-peak utility prices.
Lighting Systems
8%
Production
Equipment
9%
Compressed Air
8%
Packaging Lines
8%
Utility Systems
3%
HVAC
10%
Miscellaneous
6%
Cooling Tower
Fans
3%
Chilled Water
Pumps
9%
Condenser Water
Pumps
3%
Chillers
33%
The Facility Energy Profile identifies the major energy consuming

processes and equipment in the facility.
The FEP is best developed using actual power measurements from existing facility-wide monitoring systems.
Some types of loads, lighting, for instance, may comprise part of the usage of every major circuit in the facility.
This fact would suggest that the meter measuring the power consumption of a feeder serving the buildings
centrifugal water chillers.
Circuit Monitors
Actual power monitoring data from existing circuit monitors measuring the power consumption
of individual feeders is the best basis for establishing the Facility Energy Profile.
Demand analysis techniques

Demand analysis is the methodology used to determine if there are opportunities for a given facility to reduce
peak demand charges. Demand analysis involves manipulation of historical demand interval data to determine
which major processes or loads are operating at times of highest demand; how steep or flat the facilitys load
profile appears; and what times of day these peaks are occurring. Armed with this information, the energy auditor
can better evaluate the potential for a variety of demand reduction techniques.
The demand sort is produced by rearranging individual integrated demand readings for a given
billing period. Meters record demand readings chronologically, 3000 or so readings for a 30-day
billing period at 15-minute demand intervals; the demand sort utilizes a software tool to
distribute the readings from highest to lowest, so that times and values of peak usage are
easily analyzed.
The demand sort table facilitates demand analysis by depicting the number of intervals
(or hours) during which the plants peak electrical demand exceeded certain levels.
Using the demand sort table, the engineer is able to determine that a reduction in peak demand
to 2200 kW at this example facility would have required a demand reduction of 122 kW for 25
15-minute intervals, or 6.25 hours, in August of the sample year.
17000
16000
OffPeak
15000
Demand, kW
14000
13000
12000
Off-Peak
11000
On-Peak
2330
2245
2200
2115
2030
1945
1900
1645
1600
1515
1430
1345
1300
1215
1130
915
1045
1000
830
745
700
615
530
445
400
315
230
145
15
100
8000
ShoulderPeak
Shoulder-Peak
1815
9000
1730
10000
Interval Ending Time (Pacific Standard Time)
Peak-Day load profiles from actual power monitoring data can show consistency, or, as in
this case, a single-day aberration in peak demand that set the demand minimum billing level
(ratchet) for the remainder of the year.
Demand control
Demand controls systems are available that perform these basic functions:
I
Measure power consumption (demand) in real time
Predict demand level based on rate of instantaneous usage
Compare predicted value to target setpoint
Transmit signals to pre-determined equipment to turn off or curtail power usage if demand is predicted to
exceed target kW
These demand controls systems are intended to reduce peak demand for a facility to some predetermined level.
The design engineers foremost demand control system challenge is to identify loads in the facility that
can be controlled effectively. Ideal load candidates includes those machines or processes that are (1)
currently contributing to the facilitys load at peak times, and (2) whose function can be delayed or curtailed
at times of peak.
Most facilities lack equipment or processes that fit this ideal description, despite the numerous machines and
processes that may be operating at peak times. In fact, successful demand control is usually the exception
rather than the rule.
One common candidate for the demand control system is the air conditioning system. Buildings equipped with
multiple packaged direct-expansion air conditioning systems are typical targets of demand control sales efforts.
Unfortunately, demand control of air conditioning compressors usually leads to loss of temperature or humidity
control within the conditioned space, or lack of demand savings.
The reason for this paradox is twofold. One, natural diversity among multiple air conditioning compressors
ensures that all compressors are not operating at full load at the same time. Strangely, this fact is often
highlighted in the demand control system sales pitch: Not all compressors are running at the same time, so you
should turn some off for short periods of time.
Secondly, basic thermodynamic principles of moist air and vapor-compression refrigeration systems require
compressor power consumption to reduce air temperature and condense moisture. This process is controlled by
thermostats and humidistats within the facility. When cooling or dehumidification is removed or reduced at times
when these devices are calling for them, temperature and humidity will rise in the conditioned space.
So, if not air conditioning equipment, what loads have been successful demand control candidates?
An electrolysis process providing chemicals for a paper mill was able to reduce peak demand and flatten the
demand profile for the overall facility. A battery-charging system for forklift vehicles in an automotive facility was
capable of producing real demand savings during peak times. Finally, a large induction furnace melting scrap
metal proved to be an effective candidate for the rolling mill at a steel plant.
Chilled water supply and return temperatures increase over the course of a day due to
demand control of inlet guide vanes on a centrifugal water chiller. Space conditions
could not be maintained as a result of the demand control.
Peak shaving with onsite generators

How, the engineer might ask, can a facility save money by burning fossil fuel in an onsite generator at a unit cost
of 12 /kWh, when the average unit cost of utility purchased power is 8 /kWh? Very carefully, is the expected
and accurate response.
The key to economical peak shaving is to understand and optimize the demand savings associated with generator
operation. That is, the onsite generator must be operated the absolute minimum time necessary to reduce peak
demand the maximum amount. Because the overall average unit price of electricity is not necessarily equivalent
to the effective price of electricity at the plants peak.
For example, the facility that pays an overall average unit price of 8 /kWh probably pays only about 3-4 /kWh
for actual energy consumption, yet an additional $10-$20/kW for demand. At the end of the month, the total billing
amount divided by the total kWh usage might yield 8 /kWh average, but the actual cost of power at its peak
when demand charges are included may equate to an effective unit price of 20 /kWh or higher. For the facility
with a sharp demand peak, when the peak for the month is set in a few hours or less and the remainder of the
time demand is low, peak-shaving at 12 /kWh can be preferable to paying 20 /kWh.
I
Costs of generated power: Onsite generators typically utilize natural gas, wood, fuel oil, or steam derived
from a fossil fuel or as a part of a production process. Unit fuel costs for fossil fuels are usually calculated based
on the fuels heating value, an estimated efficiency of the generator system, and the fuel cost.
Cost/kWh = fuel price/gal * 3413/HV/efficiency,
In this equation, HV is the heating value of fuel oil in BTU/gal, and 3413 is the conversion from BTU to kWh.
Internal combustion diesel generators typically range in efficiency from 25-30%.
For a typical example, #2 fuel oil may be burned in an IC engine. For a fuel-oil price of $2.00/gal, and a
generator efficiency of 25%, the fuel cost/kWh is:
Cost/kWh = $2.00 * 3413/108,000 BTU/gal/0.25
Cost/kWh = 25 /kWh.
Obviously, peak-shaving is much less attractive at a fuel cost of $2.00/gal, unless required generator operation
can be predicted accurately and electricity charges are comparably high as well.
Utility rates affecting peak-shaving generation: Electric utility rates must be analyzed carefully prior to
implementing peak shaving or cogeneration opportunities. Some utilities have special interconnection and
protective relaying requirements to ensure that onsite generation does not pose a safety hazard for utility
workers. In addition, many utility rate schedules impose standby charges for onsite generation.
These charges are intended to recoup the utilitys investment in transformers and other equipment necessary to
serve the facilitys entire load when the onsite generation equipment is not operating. Without this standby
equipment, utilities often reserve the right to replace service equipment with smaller facilities, at risk to the
facility of overloading the smaller equipment when onsite generation is not operating.
Plant Total Power Requirement
Plant Demand, kW
On-Peak
Period
Purchased Power
Facilities with onsite generation may be able to operate this equipment to reduce purchased power
requirements during periods of high demand, or high utility prices.
4%
2%
0%
6000
5800
5600
5400
5200
5000
4800
-2%
-4%
-6%
$0.60/gal
$0.80/gal
$1.00/gal
-8%
-10%
-12%
-14%
-16%
-18%
Generator Setpoint, kW
Savings or losses associated with operation of peak-shaving generators is dependent on fuel

prices, on-peak electricity prices, the amount of time the generator has to operate for a given
peak-reduction target, and, most importantly, the accuracy with which plant personnel can
predict these variables.
Process
Steam
Turbines
Electricity
Boilers
Condenser
Condensate
Return
Electricity generation and peak shaving can also be accomplished with steam cogeneration
systems typical of paper mills, refineries, and other large industrial processes.
Lighting control
Lighting systems in industrial facilities can represent an attractive savings opportunity, especially if lighting
systems have not been upgraded or maintained in the past five years. The most cost-effective approach for
lighting energy savings is to address the following three issues, in order:
I
Turn off lights during times when they are not needed
Reduce light levels to match the requirements for the tasks being performed in the area
Replace less efficient lamps, ballasts, or fixtures with more efficient sources
The second priority in lighting conservation involves light level reductions. The Illuminating Engineering Society of
North America (www.iesna.org) has established recommended light levels for different types of work tasks and
area usage types. In addition, it offers design guidance in laying out lighting systems, estimating light levels by
zonal cavity and point-by-point lighting design methodologies.
These light level recommendations are typically described as ranges of footcandles, the footcandle being a
quantity of light measured at a horizontal or vertical surface. Light output of a fixture is usually published in
lumens. Many manufacturers of lamps and lighting systems offer software tools to aid in designing new systems,
or in evaluating changes to existing systems.
I
Some lighting essentials

N Lighting controls work better than people
While turn-off-the-light programs have been widely utilized in all types of facilities, sophisticated lighting
control systems have proven to be much more cost-effective. Certainly, its cheaper to have a worker turn off
a light, but workers forget, workers may not have access to circuit breakers controlling large banks of
industrial lighting fixtures, those same circuit breakers are not designed for daily operation as light switches,
and so on.
Lighting system controls that utilize microprocessors and specially-designed remote-operated circuit
breakers are much more effective. These devices can be programmed to accommodate complicated shift
configurations, including nights, weekends, and holidays. They also include simple over-ride features for
temporary or unusual work schedules. In addition, these systems can be monitored and controlled remotely
using standard web-browser software packages, and they can interface with other control devices such as
motion sensors or photocells.
Power Supply
Microprocessor
Remote-Operated
Circuit Breakers
Control Bus Strips
Square D PowerLink lighting control panelboard utilizes patented technology

to control lighting circuits, and offers Transparent Ready web-based monitoring
and control. See www.powerlogic.com.
N
Light levels decline with age of the lighting system.

Several factors contribute to this decline. Lamps, including fluorescent and high-intensity discharge sources
like high-pressure sodium and metal halide, experience Lamp Lumen Depreciation, or LLD. The LLD is
typically less than 1.0, indicating that average lamp light output at some point in the future is less than light
output of a new lamp.
Light levels are also adversely affect by dirt and the accumulation of dust on the light fixture. Luminaire Dirt
Depreciation, or LDD, also a factor less than 1.0, is a function of the type of light fixture as well as the
environment in which the fixture operates.
Ballast Factor, or BF, is yet another commonly used factor. BF is also a published value that is a function of
the type of ballast used to control the arc characteristics of fluorescent and HID lighting systems.
The designer usually applies these factors to the rated light level output of a lighting system, in order to
estimate the number of fixtures required to provide the desired light level not at initial installation, rather at
some designated point in the future. For example,
# fixtures = total required lumens/initial lumens/fixture/(LLD * LDD * BF).
N
Lighting designers need to know the facilitys lamp replacement practices

Manufacturers publish the rated life expectancy of a given lamp. This value, usually given in thousands of
hours, is not a guarantee that every lamp will extinguish at the same rated-life time. In fact, the rated life is
a statistical value indicating the point at which half of the lamps of a representative sample will burn out.
Some lamps will fail well shy of the rated life; others may last beyond the rated life.
The facilitys lamp replacement practices usually fall into one of two categories:
1. Replace individual lamps as they fail (spot replacement)
2. Replace all lamps at a predetermined point in time, even though many of those lamps are still
burning (group replacement)
Group replacement runs counter to common sense for most people if it aint broke, dont fix it. Thats why
spot replacement is the most common practice by far. There is, however, a sound reason for considering the
group-replacement strategy: Economics.
If the lighting designer knows, for example, that a facility will adopt the practice of group replacement, the
designer can utilize fewer light fixtures at the outset. Thats because the lamps replaced before their end of
life produce considerably more lumens than those allowed to burn to failure. The designer can use a higher
LLD in the initial light fixture calculations to achieve the same target footcandle level.
Fewer light fixtures means lower energy costs attributable to lighting, and less heat for the buildings air
conditioning system. Labor costs have also been shown to be lower for group replacement as compared to
spot replacement. Group replacement can be scheduled to occur during unoccupied times; set up and take
down costs are reduced; the cost per lamp itself can be lower with large-quantity purchases.
Electric motors
Three-phase squirrel-cage induction motors comprise a considerable percentage of the electrical load in the
United States. Design, operation, and maintenance of these machines is well described in other references; this
document focuses on their energy efficiency aspects.
Induction motors typically range in full load efficiency from about 87% to 94%. This efficiency is very difficult to
measure accurately in the field, requiring a dynamometer and other specialized equipment. Fortunately, energy
saving projects associated with electric motors do not require actual efficiency of a given motor to be established.
One of the foremost opportunities for energy savings is to implement a program of replacing rather than
rewinding induction motors at failure. Rewinding a damaged induction motor is a common practice in industry,
but studies have proven that rewinding an induction motor drops its efficiency by a couple percentage points.
Multiple rewinds can further reduce the efficiency of the rewound motor.
While a drop in efficiency from 89% to 88% seems insignificant, a quick estimate reveals that this reduction can
be costly. A standard efficiency 20 hp motor operating 8000 hours annually, for example, costs about $7000 per
10
year to operate at an average electricity rate of 7 /kWh. Once this motor fails, the least-cost option for returning it
to service is typically rewinding.
The incremental cost of replacing this failed motor with an energy-efficient motor, however, is only $430. This
amount assumes considers the rewound cost, and the labor necessary to perform the motor change-out,
as sunk costs.
The annual energy savings associated with replacing the failed motor with an energy-efficient model, at a
new efficiency of 92.9%, is approximately $510. The simple payback for the replacement, therefore, is less
than one year.
Energy-efficient motor programs are applicable to any AC motor installations utilizing NEMA Design B induction
motors. Since the programs are based on replacement at failure, the full savings potential is realized after three
years or more.
HP
Rewound
Efficiency
Standard
Efficiency
Energy
Efficient
Efficiency
1
2
3
5
8
10
15
20
25
30
40
50
60
75
100
125
150
69.7%
79.5%
79.4%
81.4%
83.1%
85.1%
85.5%
87.3%
88.0%
88.1%
88.7%
90.0%
89.9%
90.4%
90.4%
90.6%
91.5%
70.7%
80.5%
80.4%
82.4%
84.1%
86.1%
86.5%
88.3%
89.0%
89.1%
89.7%
91.0%
90.9%
91.4%
91.4%
91.6%
92.5%
82.6%
83.4%
86.6%
88.3%
90.0%
91.1%
92.0%
92.9%
93.5%
93.7%
94.2%
94.4%
94.7%
94.9%
95.4%
95.3%
95.7%
Published efficiencies of typical rewound, standard,

and energy-efficient three-phase induction motors.
Electric motors are efficient machines, even at partial load.
But power factor drops off sharply at half load.
11
Variable-speed drives
There are many devices used to provide AC motor control starting, stopping, changing speed, varying torque,
providing protection from voltage and current anomalies. This section will focus, however, on variable-frequency
control devices designed to reduce energy consumption and improve operation of three-phase AC induction
motors. See www.squared.com for technical publications that describe these devices in greater detail.
AC motor loads are typically grouped in four major categories:
Type of Load
Typical Examples
Variable torque
Centrifugal pumps and fans
Constant torque
Reciprocating pumps, conveyors, hoists
Constant horsepower
Grinders
Impact
Punch press
Energy-saving opportunities commonly focus on the variable-torque category, because the energy saving potential
is large even with small changes in pump or fan speed control.
This opportunity is driven by the power and speed characteristics of the variable-torque load. The capacity
of a pump or fan is directly proportional to the speed. A change in speed of 10% yields a change in pump gpm
or fan CFM of 10%.
Brake horsepower, however, is proportionally to the cube of the speed, meaning that a 10% reduction in pump
or fan speed can yield a 27% reduction in power consumption.
In addition, pumps and fans are often controlled by mechanical devices in the fluid flow stream, such as dampers,
control valves, and guide vanes. These devices are typically much less efficient means of varying pump volume or
fan delivery than changing the speed of the pump or fan.
Since most pumps and fans are driven by fixed-speed electric motors, where speed of the driven load is
determined by the number of motor poles, AC frequency, and motor slip, varying the speed of a motor requires an
external device. This external device is commonly referred to as an adjustable-speed drive, variable-frequency
drive, inverter, vector drive, or adjustable-frequency controller.
Variable-torque loads, such as centrifugal pumps and fans, exhibit a cubic relationship
between brake horsepower and speed.
12
Compressed air
Compressed air systems can consume a significant amount of electric energy in an industrial facility. Many textile,
automotive, chemical, and petroleum facilities operate large, multi-stage air compressors driven by electric motors
representing hundreds, thousands, or even ten-thousands of horsepower in capacity. One chemical plant
providing raw materials for synthetic textile manufacturing operated one 22,000 hp, and two 8,000 hp
compressors in a portion of its process.
While the 22,000 hp compressor is rare, significant energy reduction opportunities associated with compressed
are available.
Power consumption of a typical air compressor is a function of the air volume required (V),
the inlet air temperature (Tin), and the required pressure rise (Pout/Pin).
These opportunities may include:

I
Reduce outlet pressure compressor discharge pressure in some facilities is set too high. Since the pressure
rise across a compressor is a key factor in its power consumption, reducing outlet pressure can offer significant
savings. Some reasons for excessive pressure may be straightforward; eg, production equipment with lower
requirements has replaced older machines without a corresponding reduction in compressed air setpoint. Other
reasons may be more complex; eg, piping system losses or leaks may force higher setpoints at the compressors
in order to provide adequate air pressure at production equipment.
Reduce air volume (CFM) requirements compressed air leaks usually offer the most attractive opportunity for
reducing compressed air volume. Some facilities have ignored leaks to the point that one compressor is
effectively operating 24/7 simply to serve air leaks.
Reduce inlet temperature warm air is less dense than cold air. As the compressed air work equation above
indicates, reducing inlet air temperature can reduce the work associated with a compressor. The usual method
of reducing air temperature is to provide outside air intakes for the compressor, rather than allowing the
compressor to utilize air from a hot equipment room.
Increase inlet pressure Its common to assume that inlet pressure to a compressor is fixed at atmospheric
pressure, but this is a misconception. Air compressor inlet systems, especially air filters, need to be kept clean
and free of obstructions. Pressure drop across dirty or blocked intakes serves to reduce the pressure at the
compressor and increase power consumption.
Multi-stage air compressors, equipped with inter- and after-coolers to optimize

efficiency and provide heat recovery, are common in industrial facilities.
13
Centrifugal water chillers

Centrifugal water chillers comprise a significant portion of industrial and large commercial electrical load.
These machines are efficient, typically producing a cooling effect two-to-three times greater than the required
energy input. Centrifugal water systems were the focus of cholorfluorocarbon (CFC) legislation in the 1980 that
drove the replacement or reconditioning of many of these machines. Opportunities still exist, however, for chiller
optimization.
One opportunity is to change the operational strategy of multiple chillers operating on a common chilled water
header. Typically, these machines are staged so that none are loaded beyond about 80% of their rated
capacity. This strategy developed as a result of the published part-load efficiencies of the machines, which
tended to produce a U-shaped efficiency curve. The curve indicated that optimal efficiency was obtained at
60%-80% of full load.
Actual measurements in industrial facilities, however, suggest that the laboratory-based efficiency curve is not
representative of plant conditions. Cooling load on a typical industrial water chiller systems is often influenced by
process changes that do not correspond to a linear change in condenser water temperature. The chiller efficiency,
therefore, increases with increasing cooling load so that it reaches its optimum point at about full rated capacity.
Dem and per Ton (kW/ton)
0.95
0.9
Chiller Efficiency
0.85
0.8
0.75
0.7
0.65
0.6
0.55
0.5
20%
30%
40%
50%
60%
70%
80%
90%
100%
Part Load (%)
140.0%
120.0%
%loaded CHILLER1_KWD
100.0%
80.0%
60.0%
40.0%
3/30/98 20:01
3/28/98 22:16
3/27/98 0:31
3/25/98 2:46
3/23/98 5:01
3/21/98 7:16
3/19/98 23:19
3/13/98 9:57
3/16/98 16:52
3/8/98 5:31
0.0%
3/10/98 15:17
20.0%
3/6/98 7:58
Percent of Full Load kW (%)
Typical chiller efficiency curve is shown as a u-shape, with maximum efficiency

at 60%-80% of full load. The actual measured efficiency of chillers serving
industrial loads, however, is highest at full load.
Operating three chillers at partial loads is less efficient that operating two chillers
at or near their rated capacity.
Other successful strategies for chiller optimization include:
14
Chilled water reset this strategy involves increasing the chilled water supply temperature setpoint to match
the requirements of the cooling load. Reset is often performed as part of the control routines in an automatic
chiller controller. Chilled water reset can reduce compressor power consumption by 1.5%-2% per degree.
Reduce condenser water temperature similar to raising the chilled water setpoint, reducing the condenser
water temperature serves to reduce the compressor power requirements. Condenser water temperature
reduction of one degree can reduce compressor power consumption by 0.5%-1%.
Monitor and maintain chiller approach temperatures chiller condensers and evaporators are shell-and-tube
heat exchangers that require periodic maintenance to maintain optimum heat transfer characteristics. Since
water travels through the condenser and evaporator tubes, solids have a tendency to accumulate on internal
tube surfaces, requiring annual rodding to remove the scale and restore heat transfer coefficients.
Annual average readings of condenser approach temperature (difference in temperature

between condenser water and refrigerant in shell-and-tube heat exchanger) gradually
crept up from the initial design value of 6 F to nearly 15 F over three years.
Effect of Scale on Compressor Horsepower
Source: Carrier System Design Manual, Carrier Air Conditioning Co, 1963.
145%
Relative HP per Ton
140%
137% Additional Power Required!
135%
130%
125%
120%
115%
110%
Est'd FF = 0.0034
105%
100%
Clean
0.001
0.002
0.003
0.004
Condenser Fouling Factor
Increase in condenser tube fouling can have a significant adverse effect on

compressor power consumption.
Heating, ventilating, and air conditioning systems

HVAC systems should be the focus of a targeted energy study, with similar objectives as the lighting analysis:
I
I
I
Turn off unnecessary HVAC equipment during unoccupied times

Match HVAC operation, including temperature and humidity, to minimum occupancy requirements
Replace inefficient HVAC systems and equipment with energy-saving alternatives
WAGES
WAGES is the acronym for the complete power and energy monitoring system in a typical industrial facility.
Industrials are concerned about the costs of Water, Air (compressed), Gas (natural gas), Electricity, and Steam.
These systems are often interrelated to the degree that reductions in one utility can increase usage in another.
The power monitoring system used by industrials has to have the capability of monitoring each of these
parameters accurately, and of posting this information in a common, preferably web-based, format for use by the
local site and by remote engineers and managers.
Web-based power monitoring systems allows energy managers to monitor the results
of their demand and energy reduction techniques through the internet, and facilitate
identification of new opportunities.
15
Energy survey checklist

Lighting
1. Lighting operating more hours than needed?
N Reduce operating hours with lighting control system.
2. Areas over lit for task performed?
N Reduce light levels by disconnecting or replacing lamps or fixtures.
3. Incandescent or quartz lamps operating more than 2,000 hours per year?
N Convert to fluorescent or other energy efficient source.
4. Mercury vapor lamps.
N Convert to energy saving fluorescent, metal halide, or high-pressure sodium.
5. Standard fluorescent lamps operating one shift.
N Convert to energy saving fluorescent lamps and ballasts.
6. Standard fluorescent lamps operating two or three shifts.
N Convert to energy saving fluorescent lamps and ballasts.
7. Fluorescent at 18-feet or higher mounting heights.
N Convert to high pressure sodium.
8. VHO fluorescent fixtures.
N Convert to energy saving fluorescent, metal halide, or high pressure sodium.
9. Standard fluorescent ballasts.
N Replace with energy savings electronic ballasts at failure.
Induction motors
1. Motors operating 75%+ full load, more than 6,000 hours per year.
N Replace with energy efficient motors at failure.
2. Standard V-belts on pumps or fans.
N Convert to cog V-belts.
3. Fans or pumps that are throttled with dampers or control valves.
N Consider variable speed drives.
Demand management
1. Sharp demand peaks of short duration (low load factor)?
N Identify loads to shed or reschedule to off-peak.
2. Batch processes?
N Shift to off-peak.
3. Consider Time-of-Use savings opportunities.
Exhaust, ventilation, and pneumatic conveying

1. Transport velocities or exhaust flows higher than minimum required?
N Consider changing belts and sheaves to reduce air velocity.
2. Consider variable speed or inlet vane control.
16
3. Consider exhaust air heat recovery.

4. Make-up air properly provided for all exhaust?
5. Fume hoods designed to minimize exhaust?
6. Properly designed stack heads (no Chinese hats or caps on outlets)
Fan-coil unit air handling units

1. Consider air side economizers.
2. Considered chilled water reset.
3. Consider water side economizer.
Centrifugal water chillers

1. Multiple chillers operating on a common header.
N Fully load one chiller before starting another.
2. Consider chilled water reset.
3. Consider water side economizer.
4. Consider variable speed chiller control (long hours at light loads).
5. Excessive approach temperatures Check trends or design data.
N Clean condenser and evaporator tubes.
6. Adding cooling load or chillers?
N Consider thermal energy storage.
Cooling towers
1. Consider variable speed drives for fan motors.
2. Consider PVC fill to replace wood fill material.
3. Consider velocity recovery stacks.
Boilers
1. Stack gas temperature > 400 F? (Ideal temperature: 100 degrees plus saturation temperature of the steam)
N Consider economizer to preheat feedwater or combustion air.
2. Manual or intermittent blowdown?
N Consider automatic blowdown system.
3. Continuous blowdown?
N Consider blowdown heat recovery system.
4. Excess air high or unburned combustibles?
N Consider boiling tuning.
5. Large amounts of high pressure condensate?
N Consider high pressure condensate receiver.
6. Increase amount of condensate returned.
7. Improve boiler chemical treatment.
8. Maintain steam traps.
17
Heat recovery
1. Waste water streams > 100 F?
N Consider heat exchanger and/or heat pump.
2. Waste air or gas stream > 300 F?
N Consider heat exchanger.
Cogeneration
1. Boiler rated pressure 100 psi greater than pressure required by process?
2. Concurrent steam and electrical demands?
N Consider back-pressure turbine.
Refrigeration
1. Consider hot gas heat recovery.
2. Consider thermal storage.
Compressed air
1. Provide additional small air compressor for loads.
2. Provide outside air intake.
3.
18
Eliminate air leaks.
Section 15:
Project Coordination
Introduction
With all of the technical details that must be considered, project coordination is often given a low priority or,
worse, left to chance. However, this often-overlooked aspect of power system design is vital to insure the
success of any project.
During the initial design phase

The following aspects are often overlooked during the initial design phase of a project, and can cause
considerable time and money to be expended later in the process:
I
Coordination with the Serving Utility: Coordination with the serving electric utility is vital if a clear understanding
is to be achieved between both parties. Often, additional requirements are uncovered that affect the design of
the project and its cost.
Coordination with the Local Planning/Regulatory/Codes Authorities: This is vital to the success of the
project. Additional requirements can be uncovered that affect the project, saving time and money vs.
identifying them later.
Coordination with equipment manufacturers: If this is possible prior to bidding, it can make the project run
smoother later in the process, especially for difficult equipment application situations, since a clear
understanding can be gained regarding the characteristics of the equipment in question and the best
alternatives can be evaluated.
Coordination with the installing contractor, if the actual construction is under your purview, can save time and
frustrating delays by making your installation requirements clear.
Evaluating equipment bids

In evaluating equipment bids, any clarifications or exceptions to the project specifications that the equipment
manufacturers have submitted must be taken into account. This is the time to request re-quotes based upon
rejection of the equipment manufacturers clarifications, and to evaluate any alternates that have been submitted.
It is a good idea to allow extra time for this process.
Post-bid/approval process
Once the bid process is complete, further details must be coordinated with the equipment manufacturers.
This is vital for two reasons:
1. To understand the details of the equipment proposed, and
2. To insure that the manufacturer understands the requirements of the equipment, including delivery
requirements. The time taken at this stage will save time and money later in the process.
Once shop drawings are received, it is important to review them in a timely manner, with any changes marked
clearly. Blanket statements to adhere to the specifications, without details, can lead to frustrating project delays.
This is also the time to submit the manufacturers shop drawings to the serving utility and/or local
planning/codes/regulatory authorities, if required. The equipment manufacturer will typically submit to the serving
utility if required, but this should be double-checked to avoid confusion. Also, it is good practice to obtain the
equipment submittal markups from the serving utility in order to be aware of any changes they request.
Equipment inspections
If you require an inspection of the equipment, be sure that the manufacturer clearly understands your
expectations. Make sure the manufacturer contacts you well in advance of the equipment availability date to allow
for adequate trip planning time.
1
Commissioning
When the equipment begins to arrive on site, it is a good idea to coordinate frequently with the installing
contractor. Arrange for the local sales representatives for the major equipment to be on-site periodically during
construction so that problems can be quickly resolved. If the manufacturers service technicians are responsible
for commissioning, make sure your expectations for the scope of their work are clear.
If the serving utility requires witness testing of any equipment or system, make sure they are notified at least two
weeks in advance to allow for proper planning.
Final acceptance
Once system commissioning is complete, arrange a walk-through with the client to show the completed
installation. Also, obtain all equipment operation and maintenance manuals, including field and factory test
reports, and store them in a secure area for future use.
Document Number
0100DB0603
Guide to Power System Selective Coordination

600V and Below
1. Introduction
With the inclusion of new language in the 2005 National Electrical

Code (NEC), the requirements for selective coordination of electrical
power systems are, at present, more stringent than ever before.
This paper describes the nature of selective coordination, the NEC
requirements pertaining to selective coordination, and approaches for
obtaining selective coordination in commonly-encountered scenarios
for systems 600V and below.
2. Background
The term selective coordination refers to the selection and setting of

protective devices in an electric power system in such a manner as to
cause the smallest possible portion of the system to be de-energized
due to an abnormal condition. The most commonly encountered
abnormal condition is an overcurrent condition, defined by the NEC
as any current in excess of the rated current of equipment, or the
ampacity of a conductor [1]. The NEC uses a more restricted
definition of selective coordination as follows: Localization of an
overcurrent condition to restrict outages to the circuit or equipment
affected, accomplished by the choice of overcurrent protective
devices and their ratings or settings [1]. This is the definition
used herein.
2.1. What is Selective

Coordination?
The concept of selective coordination is best illustrated by example.

In the example system of Fig. 1, all of the devices shown are
overcurrent protective devices, in this case circuit breakers. Five
system locations, labeled A-E, have been identified. If selective
coordination exists, an overcurrent condition at location E will only
cause the lighting panel circuit breaker CB B1 to trip. Similarly, an
overcurrent fault at location D should only cause lighting panel circuit
breaker CB PM1 to trip. Table I shows the protective device that
should operate for a fault in each labeled location in Fig. 1, assuming
selective coordination exists.
Document Number
0100DB0603
Guide to Power System Selective Coordination 600V and Below
Fig. 1: Example System

UTILITY SERVICE
MAIN SWITCHBOARD
A
CB M1
CB F1
C
LIGHTING PANEL
"LP1"
CB PM1
CB B1
E
Table I: Protective Device Operation for System of Fig. 1
2.2. The Nature of Overcurrents
Fault location
Device that should operate for

selective coordination
A
B
C
D
E
Utility protective device

CB M1
CB F1
CB PM1
CB B1
Overcurrent conditions may be divided into two types. An overload is

defined by the NEC as operation of equipment in excess of normal,
full-load rating, or of a conductor in excess of rated ampacity that,
when it persists for a sufficient length of time, would cause damage or
dangerous overheating [1]. Similarly, a fault is defined as an
unintentional connection of a power system conductor, resulting in an
abnormally high flow of current. Faults typically produce higher
overcurrents than do overloads, depending upon the fault impedance.
A fault with no impedance in the unintentional connection is referred
to as a short circuit or bolted fault.
Faults may also be classified as to their geometry. A three-phase fault
involves all three phases. A line-to-line fault involves only two phases.
A short circuit involving a ground path is referred to as a ground fault,
and may be a three-phase-to-ground fault, two-line-to-ground fault,
or single-line-to-ground fault (note: the typical usage of the term
ground-fault usually means a single-line to-ground fault).
2006 Schneider Electric. All rights reserved.
Document Number
0100DB0603
Statistically, ~95% of all system faults are single-line-to-ground faults.

A very low percentage of faults are bolted faults. Thus, the frequency
of occurrence of high-magnitude bolted faults is much lower than that
of lower-magnitude faults, such as arcing ground faults. These
statistics should be kept in mind when considering the requirements
for selective coordination, for reasons that are outlined herein.
2.3. The Protective Zone Concept
To further visualize the system coordination, the system of Fig. 1 can

be divided into protective zones. A fault in a given protective zone
causes a given protective device to operate. The ideal primary
protective zones for the system of Fig. 1 are shown in Fig. 2. CB B1
should be the only device to operate for a fault in its primary
protective zone, CB PM1 should be the only device to operate for an
overcurrent condition in its protective zone, etc. Note that the ideal
primary protective zone for a given protective device includes the next
level of downstream protective devices, since a protective device
cannot be assumed to trip for an internal fault in the device itself. In
other words, the ideal protective zone boundaries cannot be arbitrarily
established, but must take into account which overcurrent conditions
each protective device is able to sense and interrupt.
Note that the closer a protective zone is to the source of power, in
this case a utility service, the more of the system is de-energized for
an overcurrent condition that zone. In fact, in a radial system with
only one source of power an overcurrent condition within a protective
zone will cause all protective zones downstream from that zone to
be affected due to the trip of the overcurrent protective device for
that zone.
Fig. 2: Ideal Primary Protective Zones for the
Example System of Fig. 1
UTILITY SERVICE
CB M1
CB M1 PRIMARY
PROTECTIVE
ZONE
CB F1
CB F1 PRIMARY
PROTECTIVE
ZONE
CB PM1
CB B1
CB B1 PRIMARY PROTECTIVE ZONE
CB PM1 PRIMARY PROTECTIVE ZONE
Document Number
0100DB0603
Note also that, for an overcurrent condition in CB B1s primary

protective zone, if CB B1 fails to operate CB PM1 should operate
as a backup. Thus, CB B1s protective zone may be said to be in the
backup protective zone for CB PM1. This same relationship follows
to upstream devices as well. Each backup protective zone is limited
by the lowest level overcurrent condition the protective device can
sense. This limit is referred to as the reach of the device and is
dependent upon the size and characteristics of the device, its
settings (if applicable), and the available fault currents at various
points downstream from the device. In practice, however, the backup
protective zones should at least overlap the primary protective zone
for the next downstream device, to allow each portion of the
system to have backup protection should its primary protective
device fail to operate.
Typical backup protective zones for the system of Fig. 1 are shown in
Fig. 3. (based upon the time-current characteristics and available fault
currents for this system). Note that although the backup protective
zones overlap in a way determined by the reach of the protective
devices, the next upstream device should operate upon failure of the
primary protective device. For example, for a fault on the branch
circuit supplied by CB B1, CB PM1 should operate if CB B1 fails to
operate. For a fault on this circuit close to CB B1, the backup
protective zones for CB M1, and CB F1 overlap, as dictated by the
reach of these circuit breakers. However, if CB PM1, CB F1, and CB
M1 are selectively coordinated, even in the region where the backup
protective zones overlap CB PM1 will trip should CB B1 fail to
operate. If CB PM1 fails to operate, CB F1 will operate so long as the
fault is within its backup protective zone. Should CB F1 fail to operate,
then CB M1 will operate, again so long as the fault is within its backup
protective zone. In this case a fault on the CB B1 branch circuit, even
close to CB B1, is beyond the reach of the utility protective device, so
CB M1 is the last line of defense to clear a fault on this circuit close
to CB B1. Only CB PM1, however, provides backup protection for the
entire circuit, since its backup protective zone is the only one which
extends around the entire circuit.
Document Number
0100DB0603
Fig. 3: Backup Protective Zones for the Example

System of Fig. 1
UTILITY SERVICE
CB M1
UTILITY PROTECTION
BACKUP PROTECTIVE ZONE
CB M1 BACKUP
PROTECTIVE
ZONE
CB F1
CB PM1
CB B1
CB PM1 BACKUP
PROTECTIVE ZONE
CB PM1 BACKUP
PROTECTIVE ZONE
CB F1 BACKUP PROTECTIVE ZONE
A more specific definition of selective coordination between two

devices in series may now be stated: Selective coordination exists
between two overcurrent protective devices in series if and only if
each device is the only device which operates for faults within its ideal
primary protective zone, where the ideal primary protective zone
begins at the load terminals of that device and ends at the load
terminals of the next level of downstream devices. Operation of a
protective device in its backup zone of protection may indicate a lack
of coordination or may indicate that a protective device has failed.
Using this definition, the term system selective coordination may
be applied to an entire electric power system as follows: System
selective coordination for an electric power system exists if and
only if any outage due to an overcurrent condition is restricted to the
smallest possible number of loads, as defined by the overcurrent
device placement and the ideal protective zone for each device.
While not an official industry term, system selective coordination is
an important concept as it is the ideal condition for protective device
coordination in the context of the over-all system.
Document Number
0100DB0603
2.4. How is selective coordination

achieved?
In most cases selective coordination is achieved via the timing

characteristics of the devices to be coordinated. For example, each of
the circuit breakers for the system of Fig. 1 has its own time-current
characteristic; by coordinating these, selective coordination may be
achieved. This is usually accomplished by comparing the device
time-current characteristics graphically. An example is shown in
Fig. 4, which illustrates the time-current coordination between circuit
breakers CB M1 and CB F1 from Fig. 1. Note that a log-log scale is
used to display the device time-current characteristics. The curves for
both devices end at the available fault current for their respective
busses, in this case 30kA. Because there is no overlap in the
time-current characteristics up to 30kA, selective coordination exists
between these two devices. For example, for the 30kA available fault,
CB F1 will operate in 0.01 0.02s and CB-M1 will operate in
0.22 0.31s. CB F1 will therefore operate more quickly than
CB M1 for a fault (up to the 30kA available fault current) sensed by
both devices.
Using this graphical method, it may be stated that to achieve selective
coordination between two devices, they must have no time-current
curve overlap up to the available fault current where their ideal
primary protective zones meet. This concept is illustrated in Fig. 5.
The fact that CB M1 and CB F1 will both sense an overcurrent
condition at the primary protective zone boundary, along with the
time-current coordination between the two, establishes the actual
primary protective zone boundary at the location shown, which in this
case coincides with the ideal boundary location.
Fig. 4: Typical Time-Current Coordination Plot
100K
10K
1K
100
CURRENT IN AMPERES
10
1000
1000
CB M1
100
100
CB F1
10
0.10
TIME IN SECONDS
10
0.10
100K
10K
1K
0.01
10
100
0.01
30kA Available Fault
Document Number
0100DB0603
The fact that time is used to coordinate the operation of protective

devices in series has an important, and unfortunate, drawback: The
closer to the source of power, the slower the protective device must
be to coordinate with downstream devices. This means that for faults
close to the source of power, fault clearing will be slower than it could
be if coordination were not a consideration. This has important
implications for equipment damage and arc-flash hazards, both of
which must be taken in to consideration in an over-all system design.
It also has important implications for the backup protection described
above, since fault clearing will be slower if the closest upstream
device fails to operate or clear the fault. Techniques to mitigate these
problems, such as Zone Selective Interlocking (ZSI), are available.
Fig. 5: Primary Protective Zones for the System of Fig. 1,
Showing the Available Fault Current Referenced in Fig. 4
UTILITY SERVICE
Overcurrent
30kA
Available
Fault
CB M1
CB M1 PRIMARY
PROTECTIVE
ZONE
CB F1
CB F1 PRIMARY
PROTECTIVE
ZONE
CB PM1
CB B1
CB B1 PRIMARY PROTECTIVE ZONE
To illustrate how miscoordination of devices affects the protective

zones, consider the coordination between CB F1 and CB PM1 per
Fig. 6. CB F1 and CB PM1 have been deliberately selected so as to
miscoordinate for purposes of illustration. Note that coordination
between CB F1 and CB PM1 exists up to 21.6kA. There is, however,
25kA available fault current at the line terminals of CB PM1 (because
protective devices generally do not present significant impedance in
the circuit, the available fault current at either the line or load
terminals of a protective device is the same. The line side of the
circuit breaker is referenced by convention, although the ideal
protective zone boundaries meet at the load terminals). This has the
effect of causing the primary protective zones for CB F1 and CB PM1
to overlap to the point in the system where the available fault current
is 21.6kA. This is illustrated in Fig. 7. Similarly, the primary protective
zones for CB PM1 and CB B1 overlap to the point in the system
Document Number
0100DB0603
where the available fault current is 2kA. It can be readily seen that the
primary protective zones in Fig. 7 are not the ideal primary protective
zones per Fig. 2.
Fig. 6: Time-Current Plot showing lack of selective
coordination between CB F1 and CB PM1
100K
10K
1K
100
CURRENT IN AMPERES
10
1000
1000
CB M1
CB F1
100
100
CB PM1
10
CB B1
10
TIME IN SECONDS
0.10
0.10
100K
10K
1K
0.01
10
100
0.01
CB PM1, CB B1 Coordinate through 2kA

CB F1, CB PM1 Coordinate through 21.6kA
Fig. 7: Protective Zones for Time-Current Plot of Fig. 6

UTILITY SERVICE
CB M1
CB M1 PRIMARY
PROTECTIVE
ZONE
CB F1
30kA
Available
Fault
CB F1 PRIMARY
PROTECTIVE
ZONE
21.6 kA Available Fault

2kA Available Fault
25kA
Available
Fault
CB PM1
CB B1
CB B1 PRIMARY
PROTECTIVE ZONE
Document Number
0100DB0603
From the discussion above that it is apparent that it becomes more

difficult to coordinate two overcurrent protective devices as the fault
current increases. This is an important concept in light of the statistics
presented earlier: The frequency of occurrence of high-magnitude
bolted faults is much less than that of lower-magnitude faults, such as
arcing ground faults.
2.5. What about Equipment

Protection?
Equipment protection is an important part of the coordination process.

Time-current curves such as those shown above may be used to
show protection for cables, transformers and other equipment.
Essentially, the damage curve for the equipment in question is
superimposed upon the time-current characteristic curve(s) for the
device(s) that protect it. Equipment damage curves which fall to the
right and above the protective device curves with sufficient margin are
considered to be protected by the device(s). Equipment damage
curves which fall on top of or to the left and below the protective
device curves are considered not to be protected by the device(s).
Because this paper focuses on protective device coordination, device
protection is only addressed where it helps illustrate why a particular
protective device is set at a given level. However, it should be
understood that device protection is important. Reference [2] is an
excellent reference both for equipment protection and protective
device coordination.
2.6. NEC Requirements for

Selective Coordination
The 2005 NEC requirements for selective coordination are, at

present, more stringent that ever before (like all code requirements,
however, they are subject to interpretation). These requirements are
as follows. Code text is in italics [1]:
2.6.1. Coordinated Short-Circuit

Protection/Overload Indication
Permitted When Orderly Shutdown
is Required (NEC 240.12)
240.12 Electrical System Coordination. Where an orderly shutdown

is required to minimize the hazard(s) to personnel and equipment, a
system of coordination based upon the following two conditions shall
be permitted:
(1) Coordinated short-circuit protection.
(2) Overload indication based on monitoring systems or devices.
Where an orderly shutdown is required, short-circuit protection must
be present, but overload protection can be indicating only. This is in
lieu of full coordinated overload protection and is intended to minimize
the risk of unintentionally shutting down part of a system automatically
due to an overload condition where a lack of coordination can cause
hazards to personnel and equipment. An overload condition can
generally be tolerated for a longer period of time than a fault; the
overload indication must be acted upon by operating personnel, but
the time can be taken for an orderly, rather than an abrupt, shut-down
of the affected equipment.
2.6.2. Elevators, Dumbwaiters,

Escalators, Moving Walks,
Wheelchair Lifts, and Stairway
Chair Lifts (NEC 620.62)
Document Number
0100DB0603
620.62 Selective Coordination. Where more than one driving

machine disconnecting means is supplied by a single feeder, the
overcurrent protective devices in each disconnecting means shall be
selectively coordinated with any other supply side overcurrent
protective devices.
This requirement has been in the NEC for some time and is intended
to prevent an overcurrent condition in one elevator, escalator, etc.,
motor from de-energizing the entire feeder which supplies other
elevator(s), escalator(s), etc., which is important for fire fighter access
during a fire.
2.6.3. Emergency Systems (NEC 700.27)
700.27 Coordination. Emergency system(s) overcurrent devices

shall be selectively coordinated with all supply side overcurrent
protective devices.
The definition of an emergency system is a system legally required
and classed as emergency by municipal, state, federal, or other
codes, or by any governmental agency having jurisdiction. These
systems are intended to automatically supply illumination, power, or
both, to designated areas and equipment in the event of failure of the
normal supply or in the event of accident to elements of a system
intended to supply, distribute, and control power and illumination
essential for safety to human life. The requirement for emergency
system protective device selective coordination is new to the 2005 NEC.
Health Care facilities in Florida have long been subject to the active
oversight of the Florida Agency for Health Care Administration
(Florida AHCA). Depending upon the jurisdiction, Florida AHCA in
the past has required coordination only down to the 0.1s level
(i.e., ignoring short-circuit coordination). The advent of the new
language above and in NEC 701.18 below will undoubtedly have an
effect on this, however as of the time of writing the disposition of this
issue with Florida AHCA is unknown.
Note that selective coordination is referenced in terms of devices
rather than as system selective coordination as discussed herein.
This can have important consequences for engineers trying to
meet the requirements of this Code section, as discussed in
further detail below.
2.6.4. Legally Required Standby

Systems (NEC 701.18)
701.18 Coordination. Legally required standby system(s) overcurrent

devices shall be selectively coordinated with all supply side
overcurrent protective devices.
The definition of a legally required standby system is a system
consisting of circuits and equipment intended to supply, distribute,
and control electricity to required facilities for illumination or power, or
both, when the normal electrical supply or system is interrupted. The
requirement for legally required standby system selective coordination
is new to the 2005 NEC (see comments above).
10
Document Number
0100DB0603
2.6.5. Service Ground-Fault

Protection for Equipment
(NEC 230.95)
230.95 Ground-Fault Protection of Equipment. Ground-fault

protection of equipment shall be provided for solidly grounded wye
electrical services of more than 150 volts to ground but not exceeding
600 volts phase-to-phase for each service disconnect rated 1000
amperes or more. The grounded conductor shall be connected
directly to ground without inserting any resistor or impedance device.
The rating of the service disconnect shall be considered to be the
rating of the largest fuse that can be installed or the highest
continuous current trip setting for which the actual overcurrent device
installed in a circuit breaker is rated or can be adjusted.
Exception No. 1: The ground-fault protection provisions of this section
shall not apply to a service disconnect for a continuous industrial
process where a nonorderly shutdown will introduce additional or
increased hazards.
Exception No. 2: The ground-fault protection provisions of this section
shall not apply to fire pumps.
(A) Setting. The ground-fault protection system shall operate to
cause the service disconnecting means to open all ungrounded
conductors of the faulted circuit. The maximum setting of the groundfault protection shall be 1200 amperes, and the maximum time delay
shall be one second for ground-fault currents equal to or greater than
3000 amperes.
(B) Fuses. If a switch and fuse combination is used, the fuses
employed shall be capable of interrupting any current higher than the
interrupting capacity of the switch during a time that the ground-fault
protective system will not cause the switch to open.
(C) Performance Testing. The ground-fault protection system shall
be performance tested when first installed on site. The test shall be
conducted in accordance with instructions that shall be provided with
the equipment. A written record of this test shall be made and shall be
available to the authority having jurisdiction.
Electrical services of 1000A or greater, with over 150V to ground and
600V or less phase-to-phase (such as 480Y/277V systems), require
ground-fault protection at the service. This protection must be set to
pick up at no more than 1200A and with a maximum time delay of 1
second at 3000A or greater. Exceptions apply to continuous industrial
processes and fire pumps. This has a direct bearing on coordination
with downstream devices, as explained below.
2.6.6. Feeder Ground-Fault

Protection for Equipment
(NEC 215.10)
215.10 Ground-Fault Protection of Equipment. Each feeder

disconnect rated 1000 amperes or more and installed on solidly
grounded wye electrical systems of more than 150 volts to ground,
but not exceeding 600V phase-to-phase, shall be provided with
ground-fault protection of equipment in accordance with the
provisions of 230.95
11
Document Number
0100DB0603
Exception No. 1: The provisions of this section shall not apply to

a disconnecting means for a continuous industrial process or where
a nonorderly shutdown will introduce additional or increased hazards.
Exception No. 2: The provisions of this section shall not apply
to fire pumps.
Exception No. 3: The provisions of this section shall not apply if
ground-fault protection of equipment is provided on the supply
side of the feeder.
Feeders disconnects rated 1000A or more on systems with more than
150V to ground and 600V or less phase-to-phase require ground-fault
protection with the same requirements for services as stated in NEC
230.95. Exceptions apply to continuous industrial processes and fire
pumps, just as for NEC 230.95. In addition, if ground-fault protection
is provided on the supply side of the feeder (such as a feeder
supplied from a service with ground-fault protection) the ground-fault
protection is not required.
2.6.7. Ground-Fault Protection
in Heath Care Facilities
(NEC 517.17)
517.17 (B) Feeders. Where ground-fault protection is provided for

operation of the service disconnecting means or feeder disconnecting
means as specified by 230.95 or 215.10, an additional step of
ground-fault protection shall be provided in all next level feeder
disconnecting means downstream toward the load. Such protection
shall consist of overcurrent devices and current transformers or other
equivalent protective equipment that shall cause the feeder
disconnecting means to open.
The additional levels of ground-fault protection shall not be
installed as follows:
(1) On the load side of an essential electrical system transfer switch
(2) Between the on-site generating unit(s) described in 517.35(B) and
the essential electrical system transfer switch(es)
(3) On electrical systems that are not solidly-grounded wye systems
with greater than 150 volts to ground but not exceeding 600 volts
phase-to-phase
517.17 (C) Selectivity. Ground-fault protection for operation of the
service and feeder disconnecting means shall be fully selective such
that the feeder device, but not the service device, shall open on
ground faults on the load side of the feeder device. A six-cycle
minimum separation between the service and feeder ground-fault
tripping bands shall be provided. Operating time of the disconnecting
devices shall be considered in selecting the time spread between
these two bands to achieve 100 percent selectivity.
12
Document Number
0100DB0603
Note that NEC 517.17 applies to hospitals and other buildings with
critical care areas or utilizing electrical life support equipment, and
buildings that provide the required essential utilities or services for the
operation of critical care areas or electrical life support equipment.
NEC 517.17 (B) requires an additional level of ground-fault protection
for health care facilities where a service or feeder disconnecting
means is equipped with ground-fault protection. This additional level
of ground-fault protection must be at the next level of protective
devices downstream from the service or feeder. In NEC 517.17 (C),
not only is it stated that selectivity must be achieved, but the amount
of selectivity (6 cycles) is specified.
Note that NEC 517.17(B) effectively prohibits the use of ground-fault
protection on the essential electrical system. The result is a conflict
between NEC 517.17(B) and NEC 700.27 and NEC 701.18. This will
be discussed in further detail below.
2.7. The Coordination Study
The only true method for achieving selective coordination and

equipment protection, and documenting with certainty the fact
that these have been achieved, is via a coordination study. The
coordination study, also known as a time-current coordination study,
compares the timing characteristics of the protective devices
used with each other and with the damage characteristics of
equipment to be protected. For electronic-trip circuit breakers, the
appropriate settings for the breaker trip units are developed in the
coordination study.
Because the short-circuit currents available at different points in the
system is a concern, a coordination study is usually performed in
conjunction with a short circuit study. The short-circuit study evaluates
the short-circuit currents available in the system.
Note that the new, stringent 2005 NEC requirements mentioned
above for emergency and standby power systems do not in any way
exempt the power system engineer from performing a coordination
study. In fact, in order to fit in with the competitive bidding process
for equipment the timing of the study may need to be performed
sooner in the project timeline than previously, in order to avoid
costly mistakes in protective device selection. This is discussed in
more detail below.
13
Document Number
0100DB0603
3. Protective Device
Characteristics
Overcurrent coordination is influenced heavily by the characteristics

of the overcurrent protective devices themselves. For systems 600V
and under, the two primary types overcurrent protective devices are
circuit breakers and fuses. The characteristics of each, as they apply
to overcurrent coordination, are discussed below.
3.1. Fuses
Fuses are the simplest of all overcurrent protective devices. As such,

they offer the least amount of adjustability of any overcurrent
protective device. A fuse consists of a melting element which melts
with a pre-determined time-current characteristic for overcurrents.
Low-voltage fuses are divided into classes based upon their
characteristics. Some fuses are classified current-limiting. By strict
definition, a current-limiting fuse will interrupt currents in its currentlimiting range within cycle or less, limiting the current to a value
less than that which would be available if the fuse were replaced by a
conductor of the same impedance.
Fuse timing response to a given level of overcurrent may be
separated into melting time, which is the time required to melt the
current-responsive element, and arcing time, which is the time
elapsed from the melting of the current-responsive element to the
final interruption of the circuit. The arcing time is dependent upon the
circuit characteristics, such as the voltage and impedance of the
circuit. The total clearing time is the sum of the melting time and the
arcing time, as shown in Fig. 8.
Fig. 8: Fuse Timing Illustration
Overcurrent condition initiated.
Fuse element begins to melt
Fuse element is melted.

Arcing begins.
Melting Time
Overcurrent is cleared.
Arcing Time
Total Clearing Time
For all low-voltage fuse classes, the basic timing characteristics can
be classified in the same manner. Fuses are typically assigned a
minimum melting characteristic and a total clearing characteristic by
their manufacturer. These define the boundaries of the fuse timecurrent characteristic band. For currents with time durations below
and to the left of the time current characteristic band, the fuse will not
blow or be damaged. For currents with time durations within the timecurrent characteristic band, the fuse may or may not blow or be
damaged. For currents with time durations above and to the right of
the time-current characteristic band, the fuse will blow with a minimum
melting time given by the minimum melting time characteristic and a
total clearing time given by the total-clearing time characteristic.
Alternatively, the fuse may be assigned an average melting time
14
Document Number
0100DB0603
characteristic; in this case the total clearing characteristic is

considered to be the average melting time characteristic shifted in
time by +15% , and the minimum melting characteristic is considered
to be the average melting time characteristic shifted in time by -15%.
A typical fuse time-current characteristic band is shown in Fig. 9.
Fig. 9: Typical Low-Voltage Fuse Time-Current Characteristic Band
100K
10K
100
1K
CURRENT IN AMPERES
1000
1000
100
100
Total Clearing
Characteristic
10
10
TIME IN SECONDS
Minimum
Melting
Characteristic
0.10
0.10
100K
10K
0.01
100
1K
0.01
Note that in Fig. 9 the time-current characteristic is only shown down

to 0.01 seconds. Below this level the arcing time may be equal to or
greater than the maximum melting time [2]. The I2t energy let-through
characteristics are used in this case to determine coordination; the
minimum melting energy of the upstream fuse must be less than the
total clearing energy of the downstream fuse for two fuses to
coordinate. Fuse manufacturers publish selectivity ratio tables to
document the performance of fuses under these circumstances.
Consider, then, two fuses in series, as shown in the one-line diagram/
time current plot of Fig. 10. It is possible to establish, by means of the
time-current plot alone, that fuses FU1 and FU2 coordinate up to
8200A. Above 8200A FU1 operates in 0.01s or less and FU2 may
operate in 0.01s or less, and coordination must be established via the
fuse selectivity ratio tables.
15
Document Number
0100DB0603
Fig. 10: Fuse Coordination Example

100K
10K
10
1K
100
CURRENT IN AMPERES
1000
1000
FU 1
100
100
FU 2
10
UTILITY BUS
1
TIME IN SECONDS
10
FU 1
FU 2
0.10
0.10
FU1 & FU2 Coordinate to 8200A
3.2. Circuit Breakers
100K
10K
1K
10
100
0.01
0.01
Above 8200A Coordination must

be established by fuse ratio tables
Circuit breakers offer many advantages over fuses for the

protection of low-voltage power systems. These advantages will not
be elaborated upon here, however it should be noted that for this
reason circuit breakers are the prevalent form of overcurrent
protection for low-voltage power systems. Successful selective
coordination with circuit breakers is therefore a vital topic for
successful power system design.
Circuit breakers may be subdivided into two basic categories:
Molded-case and low-voltage power circuit breakers. Molded-case
circuit breakers may be generally divided into thermal-magnetic and
electronic tripping types. Molded-case electronic-trip circuit breakers
may be generally be further divided into two categories: those with
two-step stored energy mechanisms, often referred to as insulated
case circuit breakers (not a UL term, but does appear in the IEEE
Blue Book [5]) and those without.
From a coordination standpoint, of particular importance is the rated
short-time withstand current. This is defined as follows [5]:
Rated Short-Time Withstand Current: (A) The maximum RMS
total current that a circuit breaker can carry momentarily without
electrical, thermal, or mechanical damage or permanent deformation.
The current shall be the RMS value, including the DC component, at
16
Document Number
0100DB0603
the major peak of the maximum cycle as determined from the

envelope of the current wave during a given test time interval. (IEEE
C37.100-1992) (B) That value of current assigned by the
manufacturer that the device can carry without damage to itself, under
prescribed conditions. (NEMA AB1 1993) Syn: withstand rating;
short-time rating
All circuit breakers which have inherent time-delay characteristics
(which is essentially every circuit breaker that is not an instantaneousonly circuit breaker) have a short-time withstand capability. This
capability may or may not be published as a short-time withstand
rating, however it will manifest itself in the time-current characteristics
for the circuit breaker since a circuit breaker must be designed so that
it will not be damaged for fault currents up to its interrupting rating.
Table II gives a summary of the various low-voltage circuit breaker
types with respect to typical levels of short-time withstand capability.
Because the information given in Table II is general in nature, specific
manufacturers data must be consulted for a given circuit breaker.
Table II: Low-Voltage Circuit Breaker Types1
Circuit
Standard
Breaker Type
Tripping Type
Short-time Withstand
Capability2
Thermal-magnetic Typically much lower

than interrupting rating
Molded-Case
Low-Voltage
Power
3.2.1. Thermal-Magnetic Molded-Case

Circuit Breakers
UL 489
ANSI C37.13
UL 1066
Electronic
Typically lower than

interrupting rating
Electronic
(insulated case)3
Often comparable to
interrupting rating
Electronic
Typically comparable
to interrupting rating
Other circuit breaker types, such as molded-case circuit breakers with instantaneous-only trip
units, are available for specific applications, such as short-circuit protection of motor circuits
Short-time current is defined by ANSI C37.13 as the designated limit of available

(prospective) current at which the circuit breaker is required to perform a duty cycle consisting
of two -second periods of current flow separated by a 15s interval of zero current. For UL
489-rated circuit breakers short-time withstand is not defined and the duty cycle may vary.
Insulated-case circuit breakers exceed the UL 489 standard. The term insulated case is
not a UL term.
The typical time-current characteristic band of a thermal-magnetic

molded-case circuit breaker is shown in Fig. 11. The time band is, by
necessity, quite large; for example, the UL 489 standard allows the
instantaneous trip characteristic for a circuit breaker with an
adjustable instantaneous characteristic to vary from -20% to +30% of
the marked instantaneous trip current setting. The long-time portion of
the trip characteristic is established by a thermal element and is used
for overload and low-level fault protection. The instantaneous
characteristic is often adjustable, as shown in Fig. 12, and is used for
short circuit protection.
17
Document Number
0100DB0603
Fig. 11: Typical Thermal-Magnetic Molded-Case Circuit Breaker

Time-Current Characteristic Band
100K
10K
1K
10
100
CURRENT IN AMPERES
1000
1000
100
100
10
10
Thermal
(long-time)
Characteristic
0.10
TIME IN SECONDS
Magnetic
(instantaneous)
Characteristic
0.10
100K
10K
1K
0.01
10
100
0.01
Fig. 12: Thermal-Magnetic Circuit Breaker Time-Current

Characteristic showing adjustable instantaneous
characteristic
100K
10K
1000
1000
100
100
10
10
0.10
0.10
18
Magnetic
(instantaneous)
Characteristic
HI setting
100K
10K
1K
0.01
10
100
0.01
TIME IN SECONDS
Magnetic
(instantaneous)
Characteristic
LO setting
1K
10
100
CURRENT IN AMPERES
Document Number
0100DB0603
3.2.2. Electronic-Trip Circuit Breakers
Electronic-trip circuit breakers typically are equipped with trip units

which give the circuit breakers the general characteristics per Fig. 13.
The adjustable long-time pickup sets the trip rating of the circuit
breaker. The adjustable long-time delay, short-time pickup, short-time
delay, and instantaneous pickup allow the circuit breakers tripping
characteristics to be customized to the application. The trip unit
represented by Fig. 13 is referred to as an LSI trip unit, since it is
equipped with long-time, short-time, and instantaneous trip
characteristics. Trip units without a short-time setting are referred to
as LI trip units, and units without an instantaneous characteristic are
referred to as LS trip units. In most cases, the instantaneous
characteristic on an LSI trip unit can be turned off if necessary.
A trip unit which includes ground fault protection is denoted with
a G, i.e., LSIG.
Of particular importance to the tripping characteristic is the
instantaneous selective override level. For currents above this
override level, even if the instantaneous characteristic is turned off
the circuit breaker will trip instantaneously. The override level is
factory-set to protect the circuit breaker according to its short-time
withstand capability. Therefore, the higher the withstand level, the
higher the override is set. This is an extremely important concept and
often determines whether two circuit breakers in series selectively
coordinate. Note also that the tripping times for the instantaneous
characteristic and for currents above the override level are nonadjustable. Further, as is the case for the circuit breaker represented
in Fig. 13, there can be a difference in tripping time when the circuit
breaker is operating in the instantaneous region below the override
level vs. above the override level.
19
Document Number
0100DB0603
Fig. 13: Typical Time-current characteristics for electronic-trip

circuit breaker (molded-case circuit breaker with low
short-time withstand shown)
10K
1K
10
100K
1000
100
CURRENT IN AMPERES
1000
Adjustable Long-Time Pickup

100
100
Adjustable Long-Time Delay
10
10
TIME IN SECONDS
Adjustable
Short-Time Pickup
Adjustable
Short-Time Delay
Adjustable
Instantaneous Pickup 0.10
0.10
Instantaneous Timing
is Non-Adjustable
Instantaneous Override
Timing is Non-Adjustable
100K
10K
1K
0.01
10
100
0.01
Instantaneous Override Value (24kA -10%)
3.3. Current-Limiting
Circuit Breakers
Like fuses, circuit breakers can be designed to limit the flow of

prospective short-circuit current. Similar to a current-limiting fuse,
a current-limiting circuit breaker limits the let-through I2t to a value
that is less than its prospective value. Circuit breakers which are
current-limiting are typically shown with instantaneous characteristics
in which the tripping time decreases with current, as shown in Fig. 14.
It is worthy of note that, in some cases, even though the circuit
breaker is not officially classified as current-limiting, a degree of
current-limitation may exist [3]. This results in the circuit breaker
exhibiting time-current characteristics similar to those shown in
Fig. 14, although the instantaneous characteristic is shown as a
horizontal band.
20
Document Number
0100DB0603
Fig. 14: Typical Time-Current Characteristic for

Current-Limiting Circuit Breaker
100K
10K
10
1K
100
CURRENT IN AMPERES
1000
1000
100
10
10
0.10
0.10
100K
10K
10
1K
0.01
1
100
0.01
0.5
TIME IN SECONDS
100
3.4. Circuit Breakers in Series: The

Dynamic Impedance Concept
A relatively new, but important, concept in the coordination of

low-voltage circuit breakers is the concept of dynamic impedance.
Simply stated, a circuit breaker, when it begins to open, serves to
limit the prospective flow of current, even if it is not UL listed as a
current-limiting circuit breaker [3]. The impedance presented to the
circuit by the circuit breaker during opening changes with time as the
circuit breaker opens, hence the term dynamic. This impedance can
increase the level coordination between two circuit breakers in series
by limiting the current that the upstream circuit breaker sees for a
fault downstream of both circuit breakers when the downstream
breaker is opening.
3.4.1. Short-Circuit Coordination Tables
Taking the dynamic impedance characteristics of circuit breakers into

account for selective coordination leads to an important new tool for
the coordination of circuit breakers: Short-Circuit Coordination Tables.
Similar to fuse ratio tables, these show the level of coordination
between two circuit breakers in series, as determined by test.
Because of the dynamic impedance effects of ordinary circuit
breakers, often the level of coordination between two circuit breakers
in series is greater than their time-current characteristic bands
would indicate. As an example, in Fig. 6 the coordination level
21
Document Number
0100DB0603
between CB F1 and CB PM1 was established graphically via the

time-current bands as 21.6kA. However, testing shows that these two
circuit breakers in series, as manufactured by one specific
manufacturer, coordinate up to 35kA! So, even though the timecurrent bands do not reflect this, CB F1 and CB PM1 do coordinate
up to the available fault current of 25kA, as illustrated in Fig. 15. This
level of extra time-current coordination can often make a large
difference, as in this case.
Fig. 15: Time-Current Curve of Fig. 6, Showing Effects of
Dynamic Impedance and Current-Limiting on Level of
Selective Coordination Between CB F1 and CB PM1
100K
10K
1K
10
100
CURRENT IN AMPERES
1000
1000
CB M1
100
CB F1
100
CB PM1
10
10
0.10
TIME IN SECONDS
CB F1 and CB PM1
coordinate up to the
available fault
current of 25kA,
despite what
time-current bands
show, due to
dynamic impedance
effects (for one
specific
manufacturers
circuit breakers)
CB B1
0.10
CB PM1, CB B1
Coordinate through 2kA
100K
10K
1K
0.01
10
100
0.01
21.6kA
As with fuse ratio tables, these tables must be developed by the

manufacturer. It is extremely important that the levels of short-circuit
coordination in the short-circuit coordination tables, if different from
the levels determined from the time-current bands, be determined by
test. The present state of the art does not lend confidence to
calculated values.
3.5. Ground-Fault Protection

of Equipment
22
Ground-fault protection of equipment is designed to provide sensitive

protection for ground-faults, typically set below the level of phase
overcurrent protection. Typically, ground-fault protection is built into
the trip unit of an electronic-trip circuit breaker or, in the case of a
Document Number
0100DB0603
thermal-magnetic circuit breaker or fuses, can be supplied via a

separate ground relay. Note that if fuses are used a separate
disconnecting means with shunt-trip capability is required. A typical
time-current characteristic is given in Fig. 16.
The current-sensing arrangement for ground-fault protection may
consist of a simple residual connection of current sensors/CTs,
a single zero-sequence sensor/CT, or may be a complex affair
with differential connections of the sensors/CTs, known as a
modified-differential ground-fault arrangement. The application of
the sensors/CTs is beyond the scope of this paper but the engineer
responsible for coordination should be cognizant of the requirements
and potential application issues.
Fig. 16: Typical Ground-Fault Protection Characteristic
100K
10K
10
1K
100
CURRENT IN AMPERES
1000
1000
100
100
GF CHAR (TYP)
10
10
0.10
0.10
100K
10K
10
1K
0.01
0.5 1
100
0.01
4. Tying it All Together

Design Philosophies
and Guidelines
TIME IN SECONDS
From a system performance standpoint, it is easy to take the position

that selective coordination between overcurrent protective devices is
always beneficial, regardless of the circumstances However, on a
practical basis full selective coordination may not always be
achievable or desirable. Various industry standards recognize this
fact. Compromises may be required between selectivity and
equipment protection to achieve the desired results. Further,
economic trade-offs are often frequently encountered, as well as
code issues. Some examples of wording from various industry
standards regarding selective coordination are given in Table III.
23
Document Number
0100DB0603
Table III: Selective coordination requirements/comments per various industry standards

Standard
Requirement/Comment
NFPA 110 Standard for

Emergency and Standby
Power Systems [6]
6.5 Protection
6.5.1* General. The overcurrent protective devices in the EPSS shall be coordinated to optimize
selective tripping of the circuit overcurrent protective devices when a short circuit occurs.
Annex A
A.6.5.1 It is important that the various overcurrent devices be coordinated, as far as practicable, to
isolate faulted circuits and to protect against cascading operation on short circuit faults. In many
systems, however, full coordination is not practicable without using equipment that could be
prohibitively costly or undesirable for other reasons. Primary consideration also should be given to
prevent overloading of equipment by limiting the possibilities of large current inrushes due to
instantaneous reestablishment of connections to heavy loads.
IEEE Std. 141 IEEE

Recommended Practice
for Electric Power
Distribution for Industrial
Plants (Red Book) [4]
Chapter 5 Application and Coordination of Protective Devices

5.1.3 Importance of Responsible Planning
Protection in an electric system is a form of insurance. It pays nothing so long as there is no fault or
other emergency, but when a fault occurs it can be credited with reducing the extent and duration
of the interruption, the hazards of property damage, and personnel injury. Economically, the
premium paid for this insurance should be balanced against the cost of repairs and lost production.
Protection, well integrated with the class of service desired, may reduce capital investment by
eliminating the need for equipment reserves in the industrial plant or utility supply system.
5.2 Analysis of System Behavior and Protection Needs
5.2.1 Nature of the Problem
Operating records show that the majority of electric circuit faults begin as phase-to-ground
failures
IEEE Std. 241 IEEE

Recommended Practice for
Electric Power Systems in
Commercial Buildings
(Gray Book) [9]
Chapter 9 System Protection and Coordination

9.7 Selective Coordination
9.7.1 Coordination of Protective Devices
On all power systems, the protective device should be selected and set to open before the
thermal and mechanical limitations of the protected components are exceeded.
9.7.3 Mechanics of Achieving Coordination
Quite often, the coordination study will not demonstrate complete selective coordination because
a compromise has to be made between the competing objectives of maximum protection and
maximum service continuity.
IEEE Std. 242 IEEE

Protection and Coordination
of Industrial and Commercial
Power Systems
(Buff Book) [2]
Chapter 1 First Principles

1.1.2.2 Equipment damage versus service continuity
Whether minimizing the risk of equipment damage or preserving service continuity is the more
important objective depends upon the operating philosophy of the particular industrial plant or
commercial business. Some operations can avoid to limited service interruptions to minimize the
possibility of equipment repair or replacement costs, while others would regard such an expense
as small compared with even a brief interruption of service.
In most cases, electrical protection should be designed for the best compromise between
equipment damage and service continuity
Chapter 15 Overcurrent coordination
15.1 General discussion
In applying protective devices, it is occasionally necessary to compromise between protection and
selectivity. While experience may suggest one alternative over the other, the preferred approach is
to favor protection over selectivity. Which choice is made, however, is depended upon the
equipment damage and the affect on the process.
IEEE Std. 446 1995 IEEE

Emergency and Standby
Power Systems for
Industrial and Commercial
Applications
(Orange Book) [7]
24
Chapter 6 Protection
6.2 Short Circuit Considerations
Careful planning is necessary to design a system that assures optimum selectivity and
coordination with both power sources
Document Number
0100DB0603
4.1. Consider Selective

Coordination Early in
the Design Process
The earlier in the design process selective coordination is considered,

the less painful achieving selective coordination will be. The need
for a coordination study, even a preliminary study, early in the design
process is increasingly becoming recognized as a need if selective
coordination is to be achieved without costly re-designs.
Working with overcurrent protective device manufacturers early in the
design process generally makes the effort to achieve selective
coordination go much more smoothly. In some cases this will require
changes to the way projects are contracted and managed, since
working with a particular manufacturer generally means staying with
that manufacturer for the protective devices considered.
Good data is essential to the selective coordination effort. The utility
available fault current, impedance data for the generator units to be
used, motor fault current contribution, and good estimates of cable
run lengths are all crucial. The earlier this information is obtained, the
easier the coordination effort will generally be. When obtaining the
utility available fault current, avoid infinite bus calculations, even on
the primary side of a service transformer. Real world fault current
values will be lower than those which rely on infinite-bus assumptions.
While infinite bus assumptions have long been recognized as being
conservative for short-circuit and coordination studies, coordination
per the 2005 NEC requirements and arc-flash concerns both
necessitate obtaining actual fault current values from the utility.
Typically, obtaining both a maximum available fault current value for
use with the short-circuit and coordination studies and a minimum
available fault current value for use with arc-flash studies is preferred
(and is an acknowledgement of the electric utility industrys assertion
that available fault current values can change over time due to system
changes), although this is typically a challenge due to the industrys
reliance on infinite bus calculations.
4.2. Recognize the Conflicts and

Issues with the 2005 NEC
4.2.1. Selective Coordination What is it?
The wording of 2005 NEC 700.27 and NEC 701.18 leaves an open
issue. Although selective coordination is defined in NEC 100 as
localization of an overcurrent condition to restrict outages to the
circuit or equipment affected, accomplished by the choice of
overcurrent protective devices and their ratings or settings, NEC
700.27 and NEC 701.18 contain the wording shall be selectively
coordinated with all supply side overcurrent protective devices. What
about scenarios where two devices that are effectively in series
protect a given piece of equipment?
Such a scenario is given in Fig. 17. The transformer shown is
protected for short-circuits by the primary circuit breaker, and for
overloads by the secondary circuit breaker. For a fault where the
protective zones overlap, does it matter whether the primary or
secondary circuit breaker trips? The answer is, of course, no.
However, because of the wording of NEC 700.27 and NEC 701.18 the
two circuit breakers would need to be selectively coordinated with
25
Document Number
0100DB0603
each other, even though it has no bearing on the performance of the

system. So long as there are no connections to other devices
between the two circuit breakers, the system may be selectively
coordinated even though these two circuit breakers themselves do
not coordinate. This is a crucial difference between selective
coordination of devices and system selective coordination as
described in section 2.3 above.
Fig. 17: Typical Low-Voltage Transformer Protection Scenario
PRIMARY CB
PRIMARY CB
PROTECTIVE ZONE
TRANSFORMER
SECONDARY CB
SECONDARY CB
PROTECTIVE ZONE
Note that for transformers, such as the transformer shown in Fig. 17,
removal of the secondary overcurrent protective device may not be
possible due to restrictions in NEC 450. Removal of this device may
also hinder transformer protection. For these and other scenarios in
which two overcurrent protective devices in series must be utilized,
the local Authority Having Jurisdiction should be consulted to provide
a waiver.
Other possible scenarios for this issue are given in Fig. 18. In both
cases, selective coordination of CB 1 and CB 2 is not required for
over-all system coordination, since there are no additional devices
between the two. Both devices could be the same size device with the
same settings.
26
Document Number
0100DB0603
Fig. 18: Other Examples Where Selective Coordination of

Devices Is Not Required for Selective Coordination
of the System:
a.) Engine-Generator Set with Circuit Breaker
Feeding Switchboard with Main Circuit Breaker
b.) One Panelboard Feeding Another Panelboard
with a Main Circuit Breaker
ENGINE-GENERATOR SET
PANEL 1
CB 1
CB 1
PANEL 2
CB 2
SWITCHBOARD
CB 2
a.)
b.)
What can be done about this issue? For the short-term, the solution is
to minimize occurrences of overcurrent protective devices in series,
as discussed below. Long-term actions may include the submission of
change proposals for consideration in a future code cycle. The more
proposals that are made on this issue, the more likely the issue is to
be recognized and corrected.
4.2.2. Ground-Fault Protection in
Health-Care Facilities
From the information in the preceding sections, a conflict in the 2005

NEC with respect to health care facilities can be recognized. To do
this, consider the typical hospital electrical system per Fig. 19. Per
NEC 700.27, all emergency system devices must be selectively
coordinated with all supply-side devices. Taken literally, this forces
coordination of emergency system protective devices up to the
alternate power source and to the utility service. For services meeting
the criteria of NEC 230.95 (such as a 480Y/277V utility service 1000A
or greater), ground-fault protection is required at the service. This
ground-fault protection must be set at no greater than 1200A pickup
and a time delay of no more than 1s at 3000A or greater. NEC
517.17(B) requires an additional level of ground-fault protection in
health-care facilities, and NEC 517.17(C) requires the two levels of
ground-fault protection to coordinate with no less than a six-cycle
(0.1s) margin between the two.
NEC 517.26 requires the essential electrical system to meet the
requirements of NEC 700, which includes NEC 700.27. NEC 700.27
does not specifically require coordination of ground-fault protection.
27
Document Number
0100DB0603
To achieve selective coordination for ground-fault protection, the

lowest level of ground-fault protection would have to coordinate with
the phase time-current characteristics of the next lower downstream
device. Most often, this will require additional levels of ground-fault
protection to supplement the two required levels per NEC 517.17(B).
However, there is a problem: NEC 517.17(B) also effectively prohibits
the use of additional levels of ground-fault protection in the essential
electrical system! Therefore, selective coordination of ground-fault
protection when the essential electrical system is supplied by the
normal (utility) source cannot be achieved, in most instances,
without violating NEC 517.17(B). As stated above, ~95% of all
system faults are ground faults, therefore this is an issue with
important consequences: A ground fault on a given branch of the
essential electrical system, when it is supplied from the normal
source, can cause that branch to be taken off-line, forcing a transfer
to the alternate (generator) source. The response of the generator(s)
would be a function of the ground fault current magnitude.
All of this can transpire even if the system complies with the
wording of NEC 700.27!
Fig. 19: Typical Health-Care Facility Electrical System
(Source: NEC 2005 FPN Figure 517.30)
Normal
source
Alternate
power source
Normal
system
Nonessential
loads
Automatic
switching
equipment
Delayed
automatic
switching
equipment
Equipment
system
Life safety
branch
Critical
branch
Emergency system
Essential electrical system
What can be done about this issue? For the short-term, bringing the
issue up to the local Authority Having Jurisdiction for resolution is the
only recourse. Long-term actions may include the submission of
change proposals for consideration in a future code cycle. The more
proposals that are made on this issue, the more likely the issue is to
be recognized and corrected.
28
Document Number
0100DB0603
4.2.3. Is Coordination up to the

Available Fault current Justified
on a Practical Basis?
As mentioned in 2.2 above, the frequency of occurrence of highmagnitude bolted faults is much lower than that of lower-magnitude
faults, such as arcing ground faults. Also, the higher the current level
to which two overcurrent protective devices are coordinated, the more
difficult the coordination effort becomes. The impact of this fact upon
system protection and selective coordination are twofold, namely:
1.) It diminishes the practical need for selective coordination up to the
available fault current in favor of practicable coordination to a
lower level of fault current.
2.) It reinforces the need for coordinated ground-fault protection.
The wording of the 2005 NEC ignores the statistical evidence of the
frequency of occurrence of high-level bolted faults. In reality, these
faults are most common during the commissioning phase of the
electrical system in a facility, when damage to cable insulation and
other application and installation issues are corrected. During the
normal lifetime of the system, these types of short-circuits are rare
indeed, especially at lower levels in the system. One practical way to
address selectivity in emergency and standby systems might be to set
an established limit of 50% of the bolted fault current as the level of
coordination for overcurrent devices below a given level (for example,
400A or below); this is an approximate worst-case for the calculated
value of the arcing fault current for a 480V system when calculated
per the empirical equations in IEEE-1584 IEEE Guide for Performing
Arc-Flash Hazard Calculations [8]. Selective coordination up to such a
limit would be justifiable on a practical basis. However, no code or
standard presently sets this limit.
Arc-flash performance of the system is also a factor. In some cases,
arc-flash performance, particularly at the lower levels of the system,
may be impaired by forcing selectivity up to the available bolted fault
current. The reason for this is that the arc-flash incident energy level
is directly proportional to the time duration of an arcing fault, which
is the clearing time for the overcurrent protective device which
clears the fault.
Also, as described above the NEC effectively prohibits coordinated
ground-fault protection in health care facility essential electrical
systems, even though ~95% of all system faults are ground faults.
4.2.4. Avoid Placing Protective Devices

in Series with No Equipment
Between Them
From the foregoing discussion in 4.2.1, in many cases it is possible

to meet the wording of NEC 700.27 and NEC 701.18 by avoiding
the use of overcurrent protective devices in series with no equipment
in between. Two examples of this are shown in Fig. 18 above.
Fig. 20 shows the examples re-designed to eliminate redundant
protective devices.
29
Document Number
0100DB0603
Fig. 20: Examples of Fig. 18 re-designed to eliminate

redundant devices
ENGINE-GENERATOR SET
PANEL 1
CB 1
PANEL 2
SWITCHBOARD
CB 1
a.)
b.)
Care must be taken to insure that another NEC section is not violated
when this is done, and that adequate protection of system
components is maintained. For example, the panelboard PANEL 2 of
Fig. 20 b.) may be a main-lugs only panel because there is no NEC
requirement for a panelboard to have a local main disconnect, only
overcurrent protection; this applies in all cases, even when the
supplying panel is on a different floor. Overcurrent protection for the
feeder cables between PANEL 1 and PANEL 2, and for PANEL 2, is
provided by CB 1 in PANEL 1. For the generator of Fig. 20 a.),
however, the removal of the circuit breaker at the generator should be
verified with the local Authority Having Jurisdiction due to possible
conflicts in interpretation of NEC 445.18, which requires a generator
to be equipped with a disconnect by which the generator can be
disconnected from the circuits it supplies. From a protection
standpoint, the cables between the generator and CB 1 can typically
withstand more short-circuit current than the generator can provide,
and, further, the generator voltage regulators control system may
have inherent features to shut down the generator if the generator
supplies a fault for an extended period of time; this must, of course,
be double-checked before making the decision to remove the circuit
breaker at the generator. Overload protection for the generator and
generator load cables is provided by CB 1.
4.3. Recognize the Pitfalls

of Generator Protection
30
Selective coordination of devices is often difficult or impossible while

maintaining adequate generator protection. Consider the system of
Fig. 21. It can be shown that adequate short-circuit protection of the
generators and coordination of CB 1 and CB 2 with CB 3, CB 4 and
CB 5 are usually mutually exclusive, especially if only one generator
is running and when CB 3, CB 4, and CB 5 short-time settings have to
be maximized to achieve coordination lower in the system (it is
assumed that CB 1 CB 5 are electronic-trip circuit breakers with
high short-time withstand ratings, such as ANSI power circuit
breakers or insulated-case circuit breakers). This would be the case
regardless of the requirements of the NEC for selective coordination
Document Number
0100DB0603
or the selectivity of downstream devices. As an illustration of the

effects of this lack of selectivity, consider the system of Fig. 22, which
is the same system from Fig. 21 expanded to show the primary
protective zones of the overcurrent protective devices. Note that
although CB 3 and CB 6 selectively coordinate, the required settings
of CB 1 and CB 2 for generator protection cause their primary
protective zones to completely overlap the CB 3 protective zone and
extend into the CB 6 protective zone. One method to prevent this is to
design the system with a larger number of smaller-size generators, as
shown in Fig. 23. This is a gross simplification, but it does illustrate
the concept. In reality, reliability concerns will, in many cases, force
additional generators to be added for redundancy; this is much more
economically feasible for the system of Fig. 22 than for the system of
Fig. 23. The addition of 51V or 51C voltage restrained/controlled
relays can often improve the generator protection, but will not
improve coordination.
Fig. 21: Application with Paralleled Generators
G
CB 1
CB 3
AUTOXFER
E
SW
TO NORMAL SOURCE
CB 2
CB 4
CB 5
AUTOXFER
N
E
SW
AUTOXFER
SW
Fig. 22: System of Fig. 21 Expanded to Show Primary

Protective Zones
G
CB 1
CB 2
CB 1 PROTECTIVE ZONE
CB 3
PROTECTIVE
ZONE
AUTOXFER
E
SW
CB 3
TO NORMAL SOURCE
CB 4
CB 5
AUTOXFER
SW
AUTOXFER
SW
CB 6
31
Document Number
0100DB0603
Fig. 23: System of Fig. 22 Re-designed for Selective Coordination

G
TO NORMAL SOURCE
CB 1
AUTOXFER E
SW
AUTOXFER
SW
AUTOXFER
SW
CB 6
Another approach is to raise the settings of the generator circuit

breakers so that they coordinate with the next level downstream.
In Fig. 22, this means that CB 1 and CB 2 would coordinate with CB
3, CB 4, and CB 5. But, CB 1 and CB 2 would no longer protect the
generators adequately for short circuits. However, CB 3, CB 4, and
CB 5 can typically be set to protect the generators for short circuits.
Therefore, only for a fault on the paralleling switchgear bus between
CB 1/CB 2 and CB 3/CB 4/CB 5 are the generators unprotected.
This can be remedied by adding a bus differential relay for this bus,
as shown in Fig. 24:
Fig. 24: System of Fig. 22 with Higher Settings for CB 1 and
Differential Relaying Added
G
TO NORMAL SOURCE
87 B
87B PROTECTIVE ZONE

CB 1
CB 2
CB 3
CB 3
PROTECTIVE
ZONE
AUTOXFER E
SW
CB 4
CB 5
AUTOXFER
SW
AUTOXFER
SW
CB 6
32
Document Number
0100DB0603
In Fig. 24, the differential relay 87B would typically be of the highimpedance type, and would trip CB 1, CB 2, CB 3, CB 4, and CB 5.
A fault between CB 1/CB 2 and CB 3/CB 4/CB 5 will cause this relay
to trip, and, if it is set appropriately, it will operate faster than the trip
unit settings of CB 1 or CB 2, providing short-circuit protection for the
generators in this protective zone as well as providing short-circuit
protection for the paralleling switchgear bus. Generator overload
protection would still be provided by CB 1 and CB 2. Note that
generator differential protection is not shown; it could be provided to
provide additional protection for the generator, but would not be an
aid to selectivity. Generator differential relays, if used, should be of
the percentage-differential type rather than impedance type. Note
also that lockout relays, while recommended, are not shown. The
circuit breakers which must be tripped by the differential relays must
be suitable for external relay tripping (suitable insulated case circuit
breakers or ANSI power circuit breakers are recommended, but are
typically used in this application anyway). Economic concerns (cost
of differential relays and CTs and the extra wiring required) must, of
course, be taken into account when considering this approach.
A more in-depth treatment of generator protection for emergency and
standby power systems is given in a separate paper, Protection of
Low-Voltage Generators Considerations for Emergency and
Standby Power Systems.
4.4. Utilize Circuit Breakers

with High Short-Time
Withstand Capabilities
Circuit breakers are the de-facto standard for low-voltage

overcurrent protection, for various reasons. As discussed above,
circuit breakers need not be ANSI power circuit breakers to have
a short-time withstand capability. Contrary to popular belief, circuit
breakers also need not be electronic trip breakers to have a
short-time withstand capability. When specifying circuit breakers,
remember, however, that the UL 489 standard to which molded-case
circuit breakers are designed and tested does not require a short-time
withstand capability.
The net effect of a high short-time withstand capability for a circuit
breaker is in its tripping performance in the short-circuit region. This
can be seen by evaluating the time-current characteristics for a given
circuit breaker, although short-circuit coordination tables must be
used to gain the full advantage from such circuit breakers due to the
dynamic impedance and current-limiting effects described above. In
many cases it will be necessary to increase the frame size of the
upstream circuit breaker in order obtain short-time withstand levels
high enough to achieve total selective coordination.
For the service switchgear/switchboards, ANSI power circuit breakers
or insulated case circuit breakers are essential, especially for
medium- to large systems.
33
Document Number
0100DB0603
A fairly popular misconception is that when using electronic circuit

breakers with the instantaneous function turned off, ANSI C37.20.1
low-voltage power switchgear is required. The reason behind this
misconception is that UL 891 switchboard through-bus withstand tests
are only required to be conducted for 3 cycles, whereas ANSI lowvoltage switchgear is required to have a short-time withstand rating of
30 cycles. The exception, of course, would be where a manufacturer
tests a switchboard configuration to the full 30-cycle withstand rating.
In reality, the need for a short-time withstand rating for the
switchboard bussing is only a concern where ANSI low-voltage power
circuit breakers or insulated-case circuit breakers with high (or no)
instantaneous override level is provided when the instantaneous
function is turned off. In most cases the circuit breakers provided with
switchboards have instantaneous overrides that cause the circuit
breaker to trip instantaneously above a given level even if the
instantaneous function is turned off, and these are tested with the
switchboard to insure compatibility.
4.5. Avoid Multiple Levels

of Protective Devices
Where Possible
It must be stressed that the fewer the number of levels of

overcurrent protective devices, the easier coordination becomes.
Fig. 25 illustrates this point. In Fig. 25 a.), three panels are
arranged so that three levels of selective coordination are required
(CB 1 CB 2 CB 3). In Fig. 25 b.) the same number of panels has
been re-arranged so that only two levels of selective coordination are
required (CB 1 CB 2 and CB 1 CB 3). Often such an
arrangement can be realized in a very economically feasible manner.
Fig. 25: Illustration Showing Multiple Levels of Selectivity:
a.) Three Levels
b.) Same Number of Panels Re-Arranged with Two Levels
PANEL 1
PANEL 1
CB 1
CB 1
PANEL 2
PANEL 2
CB 2
CB 2
PANEL 3
PANEL 3
CB 3
a.)
34
CB 3
b.)
Document Number
0100DB0603
4.6. Utilize Step-Down

Transformers to Lower
Fault Current
Remember that transformer impedance will lower the available fault

current, and the smaller the kVA size of the transformer, the more
drastic the reduction. Where coordination at the 480V level, for
example, is not possible, coordination from 480V to 208V through a
step-down transformer may be. If loads can be converted to utilize the
lower voltage, this can be a way to achieve selectivity.
4.7. Increase Transformer Sizes

Where Necessary
Although the smaller the transformer, the lower the available fault
current at the secondary, there may be cases where transformers
must be up-sized in order to achieve selective coordination. This is
usually due to the frame size of the primary circuit breaker required to
coordinate with devices at the next level below the transformer
secondary main. A careful balance between the required frame size of
the primary circuit breaker and the available fault current at the
transformer secondary is usually required.
4.8. Zone-Selective Interlocking

The Facts and the
Misconceptions
A popular misconception is that zone-selective interlocking (ZSI)

between electronic-trip circuit breakers can force otherwise
miscoordinated systems to coordinate. While it is true that ZSI can
reduce the amount of energy let-through during a fault, it cannot be
used to force selective coordination. The reason for this is that ZSI
typically uses the short-time or ground-fault pickup (or both) on a
downstream circuit breaker to identify that the circuit breaker detects
a fault; the downstream breaker then sends a signal to restrain the
next level upstream circuit breaker from tripping instantaneously while
at the same time itself tripping instantaneously to clear the fault.
However, the upstream circuit breaker will still continue to time out on
its time-current band, ultimately tripping if the downstream circuit
breaker fails to clear the fault in time. If the two circuit breakers are
miscoordinated, the upstream circuit breaker may trip before the
downstream circuit breaker, even with ZSI in place.
Used for the right reasons, however, ZSI is a powerful tool for
reducing equipment damage and arc-flash incident energy since, on a
coordinated system, it forces the device closest to a given fault to
open in the minimum amount of time. Typically this time is somewhat
longer than the instantaneous characteristic of the circuit breaker, due
to the inherent time delay required for the ZSI logic operation.
4.9. Dont Forget On-Site

Adjustment Requirements
Despite the most careful planning, selective coordination efforts can

quickly come to nothing if the overcurrent protective devices are not
properly set on-site. Most manufacturers factory-set all but the
ampere rating switch for electronic-trip circuit breakers in their lowest
positions, for example. The coordination study should include
tabulated settings for each overcurrent protective device which
requires adjustment, such as electronic-trip circuit breakers, thermalmagnetic circuit breakers with adjustable instantaneous
characteristics, ground-fault relays, etc.
35
5. References
Document Number
0100DB0603
[1] The National Electrical Code, NFPA 70, The National Fire
Protection Association, Inc., 2005 Edition.
[2] IEEE Recommended Practice for Protection and Coordination of
Industrial Power Systems, IEEE Std. 242-2001, December 2001.
[3] Short Circuit Selective Coordination for Low Voltage Circuit
Breakers, Square D Data Bulletin 0100DB0501, October 2005.
[4] IEEE Recommended Practice for Electric Power Distribution for
Industrial Plants, IEEE Std. 141-1993, December 1993.
[5] IEEE Recommended Practice for Applying Low-Voltage Circuit
Breakers Used in Industrial and Commercial Power Systems,
IEEE Std.1015-1997, October 1997.
[6] Standard for Emergency and Standby Power Systems, NFPA 110,
The National Fire Protection Association, Inc., 2005 Edition.
[7] IEEE Recommended Practice for Emergency and Standby Power
Systems for Industrial and Commercial Applications, IEEE Std.
446-1995, July 1996.
[8] IEEE Guide for Performing Arc-Flash Hazard Calculations, IEEE
Std. 1584-2002, September 2002.
Schneider Electric - North American Operating Division

1415 S. Roselle Road
Palatine, IL 60067
Tel: 847-397-2600
Fax: 847-925-7500
tk
[9] IEEE Recommended Practice for Electric Power Systems in

Commercial Buildings, IEEE Std. 241-1990, December 1990.
Document Number
0100DB0604
Selectivity Guidelines For Square D Panelboards

The natural advantages of circuit breakers make them the logical choice for
overcurrent protection. New requirements in the National Electrical Code for
emergency and legally required standby systems make it advantageous to
consider selective coordination at the beginning of the design process. This
guide is intended to facilitate the design of selectively coordinated systems
when using Square D I-Line, NF and NQOD panelboards.
Introduction
In this guide, the specific application of circuit breakers in Square D I-Line, NF,
and NQOD panelboards at the 480V and 208V levels are considered.
Information from Data Bulletin 0100DB0501 (Short Circuit Selective
Coordination for Low Voltage Circuit Breakers) is utilized, along with TCC
comparisons where necessary. The result is a set of tables which allow for
easy and efficient selection of Square D panelboards and their overcurrent
devices. Two specifications for selective coordination are considered:
coordination from 0.1 1000s and coordination from 0.01 - 1000s. The
specification that is used will depend upon the NEC and other code
requirements of the installation and the interpretation of these requirements
by the authority having jurisdiction.
The tables herein may be used to select feeder and branch circuit breakers
that will be selectively coordinated when NF and NQOD panelboards are used
in a configuration as illustrated below:
I-LINE (UPSTREAM)
PANELBOARD - 480Y/277V
or 208Y/120VV
FEEDER
CIRCUIT
BREAKER
BRANCH
CIRCUIT
BREAKER
FEEDER
NF Available Ampacities: 125A,

250A, 400A, 600A, 800A
NQOD Available Ampacities:
100A, 225A, 400A, 600A
Available Ampacities: 100A,

225A, 400A, 600A, 800A,
1200A
NF (480Y/277V
208Y/120V) or
NQOD (208Y/120V)
MLO
(DOWNSTREAM)
PANELBOARD
Alternatively, the upstream panelboard may be an I-Line section incorporated

into a QED-2 switchboard.
Document Number
0100DB0604
Selectivity Guidelines for Square D Panelboards
Listing Of Tables
Upstream Panelboard Type: I-Line
Downstream Panelboard Type

208Y/120V
480Y/277V
0.1 - 1000s
0.01 - 1000s
0.1 - 1000s
0.01 - 1000s
NF
Table IIA
Table IIB
Table IA
Table IB
NQOD
Table IIIA
Table IIIB
N/A
N/A
To find the table which applies to your application: Select a downstream

panelboard type in the left-hand column. Read across the row to find the table
which is listed under the appropriate voltage level and selectivity specification.
For example, if the downstream panelboard is an NF panelboard, the
appropriate table for selectivity from 0.01 1000s at 480Y/277V is Table IB.
Assumptions
All thermal-magnetic circuit breakers with adjustable instantaneous trip

settings are assumed to have their instantaneous settings at maximum
Electronic-trip circuit breakers are assumed to have the smallest
sensor/rating plug size which meets or exceeds the ampacity requirements
of the given circuit. The long-time trip/delay must be set to the appropriate
level to give the breaker trip setting shown. The instantaneous function is
assumed to be turned off, if possible for the breakers under consideration,
or otherwise set to maximum. Electronic-trip circuit breakers are assumed to
have a short-time function and the short-time pickup and delay settings are
assumed to be set at maximum
All circuit breakers are shown with their maximum available ampacity
ranges. For most circuit breakers, these apply for 2- or 3-pole
configurations, although this is not always the case. The availability of a
given circuit breaker ampacity for a given model and configuration must be
double-checked
How To Use The Tables
If feeder size is known:

1. Locate Feeder Size/Upstream Panelboard Circuit Breaker Size in
leftmost column
2. Required Downstream Panelboard Ampacity is in next column to right
3. Follow row to right and select the closest Maximum Available Fault
Current at Upstream Panelboard which is greater than or equal to the
available fault current at upstream panelboard (adjust available fault current
value if necessary due to system X/R ratio - see table explanatory notes)
4. Follow row to right and select an Upstream Panelboard Feeder Circuit
Breaker Type
5. Follow row to right to obtain the Downstream Panelboard Branch Circuit
Breaker Type and the Largest Possible Branch Circuit Breaker. For
total coordination tables, the Maximum Available Fault Current at
Downstream Panelboard is given also. As long as the circuit breaker type
and maximum size are adhered to (and the available fault current at the
downstream panelboard is less than or equal to the value shown for total
coordination tables), selective coordination will be achieved as per the
coordination parameters for the table
Document Number
0100DB0604
6. If results do not yield a branch circuit breaker size which is large enough,
repeat steps 4 and 5 using a different Upstream Panelboard Feeder
Circuit Breaker Type
7. If results do not yield a branch circuit breaker size which is large enough
(or an acceptable level of fault current at the downstream panelboard),
a larger feeder will be required. Go to the next larger Feeder
Size/Upstream Panelboard Circuit Breaker Size and repeat steps 1
through 6
8. Repeat steps 1 through 7 until the desired branch circuit breaker size
is obtained
If branch circuit size/branch circuit breaker size is known:

1. Starting at top of table, scan Largest Possible Branch Circuit Breaker
sizes in rightmost column. Select the first one that is greater than or equal
to the desired branch circuit size. For total coordination tables, Maximum
Available Fault Current at Downstream Panelboard must be greater
than or equal to the actual fault current at the downstream panelboard
2. When the desired branch circuit breaker is found, follow row to left. Make
sure that the actual available fault current at the upstream panelboard is
less than or equal to the Maximum Available Fault Current at Upstream
Panelboard (adjust available fault current value if necessary due to system
X/R ratio - see table explanatory notes)
3. The required Downstream Panelboard Ampacity and Feeder
Size/Upstream Panelboard Circuit Breaker Size are as shown. This is
the smallest feeder circuit breaker that will satisfy the coordination criteria
for the table
4. Scan the rightmost column for other instances of the required branch circuit
breaker size and follow steps 1 through 3 again. The feeder circuit/I-Line
feeder circuit breaker size may be larger, but the I-Line circuit breaker may
be less expensive
Document Number
0100DB0604
Table IA
I-Line/NF Panelboard Selective Coordination At 480Y/277V
0.1s 1000s
FEEDER SIZE /
UPSTREAM (I-LINE)
PANELBOARD CIRCUIT
BREAKER SIZE
(A)
REQUIRED
DOWNSTREAM (NF)
PANELBOARD
AMPACITY
(A)
MAXIMUM AVAILABLE
FAULT CURRENT AT
UPSTREAM (I-LINE)
PANELBOARD
(kA RMS Sym.)1
18
25
100
125
35
65
110
125
125
125
18
35
65
18
30
35
65
18
30
150
250
35
65
18
175
250
30
35
65
18
200
250
30
35
65
18
30
225
250
35
65
18
250
250
30
35
65
UPSTREAM (I-LINE)
PANELBOARD
FEEDER CIRCUIT
BREAKER TYPE
FA, HD, LX
PG
FH
HG
LX
PG
HJ, LX
PJ
HD
HG
HJ
HD, LA, LX
PG
LA
HG, LH, LX
PG
HJ, LX
PJ
HD
JD, LA
LX
PG
LA
HG
JG, LH
LX
PG
HJ
JJ
LX
PJ
JD, LA
LX, PG
LA
JG, LH
LX, PG
JJ
LX, PJ
JD, LA, LX,
PG
LA-MC
LA
LA-MC
JG, LH, LX
PG
LH-MC
JJ, LX
PJ
JD, LA, LX
PG
LA-MC
LA
LA-MC
JG, LH, LX
PG
LH-MC
JJ, LX
PJ
JD, LA
LA-MC
LX, PG
LA
LA-MC
JG, LH
LH-MC
LX, PG
JJ
LX, PJ
DOWNSTREAM (NF)
PANELBOARD BRANCH
CIRCUIT BREAKER
TYPE
LARGEST POSSIBLE
BRANCH CIRCUIT
BREAKER
(A)
ED
ED
EG
EG
EG
EG
EJ
EJ
ED
EG
EJ
ED
ED
EG
EG
EG
EJ
EJ
ED
ED
ED
ED
EG
EG
EG
EG
EG
EJ
EJ
EJ
EJ
ED
ED
EG
EG
EG
EJ
EJ
ED
ED
ED
EG
EG
EG
EG
EG
EJ
EJ
ED
ED
ED
EG
EG
EG
EG
EG
EJ
EJ
ED
ED
ED
EG
EG
EG
EG
EG
EJ
EJ
30
50
30
30
30
50
30
50
30
30
30
30
50
30
30
50
30
50
30
35
40
70
35
30
35
40
70
30
35
40
70
40
70
40
40
70
40
70
70
80
125
70
125
70
80
125
70
80
70
110
125
70
125
70
110
125
70
110
70
125
125
70
125
70
125
125
70
125
Document Number
0100DB0604
FEEDER SIZE /
UPSTREAM (I-LINE)
PANELBOARD CIRCUIT
BREAKER SIZE
(A)
REQUIRED
DOWNSTREAM (NF)
PANELBOARD
AMPACITY
(A)
300
400
350
400
400
400
450
600
500
600
600
600
MAXIMUM AVAILABLE
FAULT CURRENT AT
UPSTREAM (I-LINE)
PANELBOARD
(kA RMS Sym.)1
18
30
35
65
18
30
35
65
18
800
800
800
DOWNSTREAM (NF)
PANELBOARD BRANCH
CIRCUIT BREAKER
TYPE
LARGEST POSSIBLE
BRANCH CIRCUIT
BREAKER
LA, MG, LX, PG3

LA
LH, MG, LX, PG3
LC, MJ, LX, PJ3
LA, MG, LX
LA
LH, MG, LX
LC, MJ, LX
LA, LA-MC, MG, LX, PG
ED
EG
EG
EJ
ED
EG
EG
EJ
ED
125
125
125
125
125
125
125
125
125
EG
125
30
LA, LA-MC
35
LH, LH-MC, MG, LX, PG
EG
125
65
18
35
65
18
35
65
18
LC, MJ, LX, PJ

LC, MG, LX, PG3
LC, MG, LX, PG3
LC, MJ, LX, PJ3
LC, MG, LX, PG3
LC, MG, LX, PG3
LC, MJ, LX, PJ3
LC, MG, LX, PG
(ET1.0)2, PG2
LC, MG, LX, PG
(ET1.0)2, PG2
LC, MJ, LX, PJ (ET1.0)2,
PJ2
MG, PG2,3
MG, PG2,3
MJ, PJ2,3
MG, PG (ET1.0)2, PG2
MJ, PJ (ET1.0)2, PJ2
EJ
ED
EG
EJ
ED
EG
EJ
125
125
125
125
125
125
125
ED
125
EG
125
EJ
ED
EG
EJ
ED
EG
EJ
125
125
125
125
125
125
125
35
65
700
UPSTREAM (I-LINE)
PANELBOARD
FEEDER CIRCUIT
BREAKER TYPE
18
35
65
18
35
65
1 Available fault currents are based upon system X/R ratios less than or equal to the circuit breaker test X/R ratio. See
the explanatory note below for additional information
2 The P-Frame Powerpact circuit breaker is available with ET1.0 or Micrologic 5.0/6.0 trip units in this size range
PG (ET1.0) = ET1.0 trip unit
PG = Micrologic 5.0/6.0 trip unit
3 Requires larger sensor size if standard rating plug is used (300A: 600A w/ LTPU=0.5, 450A: 1000A w/LTPU=0.45,
800A: 1000A w/LTPU=0.625, 700A: 1000A w/LTPU= 0.7)
X/R Ratio Adjustment:

All available fault currents are given in RMS symmetrical amperes. For a
system X/R ratio larger than the test X/R ratio of the circuit breaker in
question, the available fault current equivalent RMS symmetrical duty for
comparison with the values in the tables must be adjusted by a multiplying
factor. See IEEE Std. 242-2001 (Buff Book), IEEE Std. 1015-1997 (Blue Book)
or NEMA AB 3-2001 for details.
Molded Case Circuit Breaker Interrupting Rating
Greater than 20kA
10kA - 20kA
Less than 10kA
Test X/R
4.9
3.2
1.7
Note that this is a consideration for breaker fault duty rather than for
selective coordination.
Document Number
0100DB0604
Table IB
0.01s 1000s
FEEDER SIZE /
UPSTREAM (I-LINE)
PANELBOARD CIRCUIT
BREAKER SIZE
(A)
REQUIRED
DOWNSTREAM (NF)
PANELBOARD
AMPACITY
(A)
100
125
125
125
150
250
175
250
MAXIMUM AVAILABLE
FAULT CURRENT AT
UPSTREAM (I-LINE)
PANELBOARD
(kA RMS Sym.)1
18
35
65
18
35
65
18
35
65
18
35
65
18
200
250
30
35
65
18
225
250
30
35
65
18
250
250
30
35
300
300
65
18
35
65
18
400
400
30
35
450
600
500
600
600
600
65
18
35
65
18
35
65
18
UPSTREAM (I-LINE)
PANELBOARD
FEEDER CIRCUIT
BREAKER TYPE
PG
PG
PJ
PG
PG
PJ
PG
PG
PJ
PG
PG
PJ
PG
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
PG
LH-MC
LH-MC
LH-MC
PJ
PG
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
PG
LH-MC
LH-MC
LH-MC
LH-MC
PJ
LA-MC
LA-MC
LA-MC
PG
LA-MC
LA-MC
LA-MC
LH-MC
LH-MC
LH-MC
PG
PJ
PG4
PG4
PJ4
LA-MC
LA-MC
PG
LA-MC
LA-MC
LH-MC
LH-MC
PG
PJ
PG4
PG4
PJ4
PG4
PG4
PJ4
PG (ET1.0)2, PG2
DOWNSTREAM (NF)
PANELBOARD BRANCH
CIRCUIT BREAKER
TYPE
ED
EG
EJ
ED
EG
EJ
ED
EG
EJ
ED
EG
EJ
ED
ED
ED
ED
EG
EG
EG
EG
EG
EG
EG
EJ
ED
ED
ED
ED
ED
EG
EG
EG
EG
EG
EG
EG
EG
EG
EJ
ED
ED
ED
ED
EG
EG
EG
EG
EG
EG
EG
EJ
ED
EG
EJ
ED
ED
ED
EG
EG
EG
EG
EG
EJ
ED
EG
EJ
ED
EG
EJ
ED
MAXIMUM AVAILABLE
FAULT CURRENT AT
DOWNSTREAM (NF)
PANELBOARD
(kA RMS Sym.)1,3
18
35
65
18
35
65
18
35
65
18
35
65
18
18
10
6
18
10
6
35
18
10
6
65
18
18
14
8
7
18
14
8
7
35
18
14
8
7
65
18
10
8
18
18
10
8
18
10
8
35
65
18
35
65
18
6
18
18
6
18
6
35
65
18
21.6
9
18
21.6
9
18
LARGEST POSSIBLE
BRANCH CIRCUIT
BREAKER
50
50
50
50
50
50
70
70
70
70
70
70
80
15
20
100
15
20
100
80
15
20
100
80
110
15
20
30
100
15
20
30
100
110
15
20
30
100
110
30
40
100
125
30
40
100
30
40
100
125
125
125
125
125
100
125
125
100
125
100
125
125
125
125
125
125
125
125
125
125
Document Number
0100DB0604
FEEDER SIZE /
UPSTREAM (I-LINE)
PANELBOARD CIRCUIT
BREAKER SIZE
(A)
REQUIRED
DOWNSTREAM (NF)
PANELBOARD
AMPACITY
(A)
700
800
800
800
MAXIMUM AVAILABLE
FAULT CURRENT AT
UPSTREAM (I-LINE)
PANELBOARD
(kA RMS Sym.)1
35
65
18
35
65
18
35
65
UPSTREAM (I-LINE)
PANELBOARD
FEEDER CIRCUIT
BREAKER TYPE
DOWNSTREAM (NF)
PANELBOARD BRANCH
CIRCUIT BREAKER
TYPE
PG (ET1.0)2, PG2
PJ (ET1.0)2, PJ2
PG2,4
PG2,4
PJ2,4
PG (ET1.0)2, PG2
PG (ET1.0)2, PG2
PJ (ET1.0)2, PJ2
EG
EJ
ED
EG
EJ
ED
EG
EJ
MAXIMUM AVAILABLE
FAULT CURRENT AT
DOWNSTREAM (NF)
PANELBOARD
(kA RMS Sym.)1,3
LARGEST POSSIBLE
BRANCH CIRCUIT
BREAKER
(A)
35
65
18
21.6
9
18
21.6
9
125
125
125
125
125
125
125
125
1 Available fault currents are based upon system X/R ratios less than or equal to the circuit breaker test X/R ratio. See the explanatory
note below for additional information
3 Values in red are taken from data bulletin 0100DB0501; all other values in this column generated via TCC comparison
4 Requires larger sensor size if standard rating plug is used (300A: 600A w/ LTPU=0.5, 450A: 1000A w/LTPU=0.45, 800A: 1000A
w/LTPU=0.625, 700A: 1000A w/LTPU= 0.7)

Greater than 20kA
10kA - 20kA
Less than 10kA
Test X/R
4.9
3.2
1.7
Document Number
0100DB0604
Table IIA
0.1s 1000s
FEEDER SIZE /
UPSTREAM (I-LINE)
PANELBOARD CIRCUIT
BREAKER SIZE
(A)
REQUIRED
DOWNSTREAM (NF)
PANELBOARD
AMPACITY
(A)
MAXIMUM AVAILABLE
FAULT CURRENT AT
UPSTREAM (I-LINE)
PANELBOARD
(kA RMS Sym.)1
25
100
125
65
100
110
125
125
125
25
65
100
25
42
65
100
25
42
150
250
65
100
25
175
250
42
65
100
25
42
200
250
65
100
25
42
225
250
65
100
25
42
250
250
65
100
300
400
25
42
65
UPSTREAM (I-LINE)
PANELBOARD
FEEDER CIRCUIT
BREAKER TYPE
DOWNSTREAM (NF)
PANELBOARD BRANCH
CIRCUIT BREAKER TYPE
LARGEST POSSIBLE
BRANCH CIRCUIT
BREAKER
(A)
FA2, HD, LX
PG
FH, HG, LX
PG
HJ, LX
PJ
HD
HG
HJ
HD, LA, LX
PG
LA
HG, LH, LX
PG
HJ, LX
PJ
HD
JD, LA
LX
PG
LA
HG
JG, LH
LX
PG
HJ
JJ
LX
PJ
JD, LA
LX, PG
LA
JG, LH
LX, PG
JJ
LX, PJ
JD, LA, LX,
PG
LA-MC
LA
LA-MC
JG, LH, LX
PG
LH-MC
JJ, LX
PJ
JD, LA, LX
PG
LA-MC
LA
LA-MC
JG, LH, LX
PG
LH-MC
JJ, LX
PJ
JD, LA
LA-MC
LX, PG
LA
LA-MC
JG, LH
LH-MC
LX, PG
JJ
LX, PJ
LA, MG, LX, PG5
LA
LH, MG, LX, PG5
ED
ED
EG3
EG3
EJ3
EJ3
ED
EG3
EJ3
ED
ED
EG3
EG3
EG3
EJ3
EJ3
ED
ED
ED
ED
EG3
EG3
EG3
EG3
EG3
EJ3
EJ3
EJ3
EJ3
ED
ED
EG3
EG3
EG3
EJ3
EJ3
ED
ED
ED
EG3
EG3
EG3
EG3
EG3
EJ3
EJ3
ED
ED
ED
EG3
EG3
EG3
EG3
EG3
EJ3
EJ3
ED
ED
ED
EG3
EG3
EG3
EG3
EG3
EJ3
EJ3
ED
EG3
EG3
30
50
30
50
30
50
30
30
30
30
50
30
30
50
30
50
30
35
40
70
35
30
35
40
70
30
35
40
70
40
70
40
40
70
40
70
70
80
125
70
125
70
80
125
70
80
70
110
125
70
125
70
110
125
70
110
70
125
125
70
125
70
125
125
70
125
125
125
125
Document Number
0100DB0604
FEEDER SIZE /
UPSTREAM (I-LINE)
PANELBOARD CIRCUIT
BREAKER SIZE
(A)
REQUIRED
DOWNSTREAM (NF)
PANELBOARD
AMPACITY
(A)
350
400
400
400
450
600
500
600
600
600
MAXIMUM AVAILABLE
FAULT CURRENT AT
UPSTREAM (I-LINE)
PANELBOARD
(kA RMS Sym.)1
100
25
42
65
100
25
42
65
100
25
65
100
25
65
100
25
65
100
700
800
800
800
25
65
100
25
65
100
UPSTREAM (I-LINE)
PANELBOARD
FEEDER CIRCUIT
BREAKER TYPE
DOWNSTREAM (NF)
PANELBOARD BRANCH
CIRCUIT BREAKER
TYPE
LC, MJ, LX, PJ5

LA, MG, LX
LA
LH, MG, LX
LC, MJ, LX
LA, LA-MC
LC, MJ, LX, PJ
LC, MG, LX, PG5
LC, MG, LX, PG5
LC, MJ, LX, PJ5
LC, MG, LX, PG5
LC, MG, LX, PG5
LC, MJ, LX, PJ5
LC, MG, LX, PG
(ET1.0)4, PG4
LC, MG, LX, PG
(ET1.0)4, PG4
LC, MJ, LX, PJ
(ET1.0)4, PJ4
MG, PG4,5
MG, PG4,5
MJ, PJ4,5
LARGEST POSSIBLE
BRANCH CIRCUIT
BREAKER
(A)
EJ3
ED
EG3
EG3
EJ3
ED
EG3
EG3
EJ3
ED
EG3
EJ3
ED
EG3
EJ3
125
125
125
125
125
125
125
125
125
125
125
125
125
125
125
ED
125
3
EG
125
125
125
125
125
125
125
125
EJ
ED
EG3
EJ3
ED
EG3
EJ3
2 480V-rated
3 2 Pole or 3 Pole 15 125A only. 1 Pole is available from 15 70A and has an AIR of 35kA for EG, 65kA for EJ
800A: 1000A w/LTPU=0.625, 700A: 1000A w/LTPU= 0.7)

Greater than 20kA
10kA - 20kA
Less than 10kA
Test X/R
4.9
3.2
1.7
Document Number
0100DB0604
TABLE IIB
0.01s 1000s
FEEDER SIZE /
UPSTREAM (I-LINE)
PANELBOARD CIRCUIT
BREAKER SIZE
(A)
REQUIRED
DOWNSTREAM (NF)
PANELBOARD
AMPACITY
(A)
100
125
125
125
150
250
175
250
MAXIMUM AVAILABLE
FAULT CURRENT AT
UPSTREAM (I-LINE)
PANELBOARD
(kA RMS Sym.)1
25
65
100
25
65
100
25
65
100
25
65
100
25
200
250
42
65
100
25
42
225
250
65
100
25
250
250
42
65
300
300
100
25
65
100
25
42
400
400
65
10
450
600
500
600
600
600
100
25
65
100
25
65
100
25
UPSTREAM (I-LINE)
PANELBOARD
FEEDER CIRCUIT
BREAKER TYPE
PG
PG
PJ
PG
PG
PJ
PG
PG
PJ
PG
PG
PJ
PG
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
PG
LH-MC
LH-MC
LH-MC
PJ
PG
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
PG
LH-MC
LH-MC
LH-MC
LH-MC
PJ
LA-MC
LA-MC
LA-MC
PG
LA-MC
LA-MC
LA-MC
LH-MC
LH-MC
LH-MC
PG
PJ
PG6
PG6
PJ6
LA-MC
LA-MC
PG
LA-MC
LA-MC
LH-MC
LH-MC
PG
PJ
PG6
PG6
PJ6
PG6
PG6
PJ6
PG (ET1.0)4, PG4
DOWNSTREAM (NF)
PANELBOARD BRANCH
ED
EG3
EJ3
ED
EG3
EJ3
ED
EG3
EJ3
ED
EG3
EJ3
ED
ED
ED
ED
EG3
EG3
EG3
EG3
EG3
EG3
EG3
EJ3
ED
ED
ED
ED
ED
EG3
EG3
EG3
EG3
EG3
EG3
EG3
EG3
EG3
EJ3
ED
ED
ED
ED
EG3
EG3
EG3
EG3
EG3
EG3
EG3
EJ3
ED
EG3
EJ3
ED
ED
ED
EG3
EG3
EG3
EG3
EG3
EJ3
ED
EG3
EJ3
ED
EG3
EJ3
ED
MAXIMUM AVAILABLE
FAULT CURRENT AT
DOWNSTREAM (NF)
PANELBOARD
(kA RMS Sym.)1,5
21.6
65
100
21.6
65
100
21.6
65
100
21.6
65
100
21.6
18
10
6
18
10
6
65
18
10
6
100
21.6
18
14
8
7
18
14
8
7
65
18
14
8
7
100
18
10
8
21.6
18
10
8
18
10
8
65
100
21.6
65
100
18
6
21.6
18
6
18
6
65
100
21.6
21.6
9
21.6
65
100
21.6
LARGEST POSSIBLE
BRANCH CIRCUIT
BREAKER
(A)
50
50
50
50
50
50
70
70
70
70
70
70
80
15
20
100
15
20
100
80
15
20
100
80
110
15
20
30
100
15
20
30
100
110
15
20
30
100
110
30
40
100
125
30
40
100
30
40
100
125
125
125
125
125
100
125
125
100
125
100
125
125
125
125
125
125
125
125
125
125
Document Number
0100DB0604
FEEDER SIZE /
UPSTREAM (I-LINE)
PANELBOARD CIRCUIT
BREAKER SIZE
(A)
REQUIRED
DOWNSTREAM (NF)
PANELBOARD
AMPACITY
(A)
700
800
MAXIMUM AVAILABLE
FAULT CURRENT AT
UPSTREAM (I-LINE)
PANELBOARD
(kA RMS Sym.)1
65
100
25
65
100
800
800
25
65
100
UPSTREAM (I-LINE)
PANELBOARD
FEEDER CIRCUIT
BREAKER TYPE
PG (ET1.0)4, PG4
PJ (ET1.0)4, PJ4
PG4 6
PG4,6
MG
PJ4,6
MJ
PG (ET1.0)4, PG4
DOWNSTREAM (NF)
PANELBOARD BRANCH
CIRCUIT BREAKER
TYPE
MAXIMUM AVAILABLE
FAULT CURRENT AT
DOWNSTREAM (NF)
PANELBOARD
(kA RMS Sym.)1,5
EG3
EJ3
ED
EG3
EG3
EJ3
EJ3
ED
EG3
EJ3
65
100
21.6
21.6
65
9
100
21.6
65
100
LARGEST POSSIBLE
BRANCH CIRCUIT
BREAKER
(A)
125
125
125
125
125
125
125
125
125
125
1 Available fault currents are based upon system X/R ratios less than or equal to the circuit breaker test X/R ratio. See the explanatory notes
below for additional information
2 480V-rated
3 2 Pole or 3 Pole 15 125A only. 1 Pole is available from 15 70A and has an AIR of 35kA for EG, 65kA for EJ
800A: 1000A w/LTPU=0.625, 700A: 1000A w/LTPU= 0.7)

Greater than 20kA
10kA - 20kA
Less than 10kA
Test X/R
4.9
3.2
1.7
11
Document Number
0100DB0604
TABLE IIIA
I-Line/NQOD Panelboard Selective Coordination At 208Y/120V
0.1s 1000s
FEEDER SIZE /
UPSTREAM (I-LINE)
PANELBOARD CIRCUIT
BREAKER SIZE
(A)
REQUIRED
DOWNSTREAM (NF)
PANELBOARD
AMPACITY
(A)
MAXIMUM AVAILABLE
FAULT CURRENT AT
UPSTREAM (I-LINE)
PANELBOARD
(kA RMS Sym.)1
10
100
100
22
65
110
225
10
22
65
10
125
225
22
42
65
10
150
225
22
42
65
10
175
225
22
42
65
10
200
225
22
42
65
10
225
225
22
42
65
10
250
400
22
42
65
10
300
350
12
400
22
400
42
65
10
UPSTREAM (I-LINE)
PANELBOARD
FEEDER CIRCUIT
BREAKER TYPE
DOWNSTREAM (NF)
PANELBOARD
BRANCH CIRCUIT
BREAKER TYPE
LARGEST POSSIBLE
BRANCH CIRCUIT
BREAKER
(A)
HD
FA, LX
PG
HD
FA2, LX
PG
HG
FH, LX
PG
HD
HD
HG
HD, LA
LX
PG
HD, LA
LX
PG
LA
HG, LH
LX, PG
HD
JD, LA, LX, PG
HD
JD, LA
LX
PG
LA
HG
JG, LH, LX, PG
JD, LA, LX, PG
JD, LA
LX
PG
LA
JG, LH, LX, PG
JD, LA,
LX
LA-MC, PG
JD, LA, LX,
PG
LA-MC
LA, LA-MC
JG, LH, LH-MC, LX, PG
JD, LA, LX, PG
LA-MC
JD, LA
LX
PG
LA-MC
LA, LA-MC
JD, LA, LX, PG
LA-MC
JD, LA
LX
PG
LA-MC
LA, LA-MC
LA, MG, LX
PG5
LA
MG, LX
PG5
LA
LH, MG, LX, PG5
LA
QO
QO
QO
QO-VH
QO-VH
QO-VH
QH
QH
QH
QO
QO-VH
QH
QO
QO
QO
QO-VH
QO-VH
QO-VH
QH
QH
QH
QO
QO
QO-VH
QO-VH
QO-VH
QO-VH
QH
QH
QH
QO
QO-VH
QO-VH
QO-VH
QH
QH
QO
QO
QO
QO-VH
QO-VH
QO-VH
QH
QH
QO
QO
QO-VH
QO-VH
QO-VH
QO-VH
QH
QH
QO
QO
QO-VH
QO-VH
QO-VH
QO-VH
QH
QH
QO
QO
QO-VH
QO-VH
QO-VH
QH
QH
QO
20
25
40
20
25
403
20
25
30
25
25
25
25
40
70
25
30
603
25
25
30
25
70
25
30
403
703
30
25
30
70
403
503
703
30
30
70
806
1006
503
803
1003
30
30
1006
1257
503
603
803
1253
30
30
1006
1257
603
803
1103
1503
30
30
1006
1257
903
1003
1253
30
30
1006
Document Number
0100DB0604
FEEDER SIZE /
UPSTREAM (I-LINE)
PANELBOARD CIRCUIT
BREAKER SIZE
(A)
REQUIRED
DOWNSTREAM (NQOD)
PANELBOARD
AMPACITY
(A)
MAXIMUM AVAILABLE
FAULT CURRENT AT
UPSTREAM (I-LINE)
PANELBOARD
(kA RMS Sym.)1
350
400
400
400
450
600
500
600
22
42
65
10
22
42
65
10
22
65
10
22
65
10
600
600
22
65
UPSTREAM (I-LINE)
PANELBOARD
FEEDER CIRCUIT
BREAKER TYPE
DOWNSTREAM (NQOD)
PANELBOARD
BRANCH CIRCUIT
BREAKER TYPE
MG, LX
LA, MG, LX
LA
LH, MG, LX
LA, LA-MC
LC, MG, LX, PG5
LC, MG, LX, PG5
LC, MG, LX, PG5
LC, MG, LX, PG5
LC, MG, LX, PG5
LC, MG, LX, PG5
LC, MG, LX, PG
(ET1.0)4, PG4
LC, MG, LX, PG
(ET1.0)4, PG4
LC, MG, LX, PG
(ET1.0)4, PG4
LARGEST POSSIBLE
BRANCH CIRCUIT
BREAKER
(A)
QO
QO-VH
QH
QH
QO
QO-VH
QH
QH
QO
QO-VH
QH
QO
QO-VH
QH
1257
1503
30
30
1257
1503
30
30
1257
1503
30
1257
1503
30
QO
1257
QO-VH
1503
QH
30
2 480V-rated
3 2 Pole or 3 Pole only. QO-VH 1 Pole is available up to 30A (and coordinates up to 30A)
800A: 1000A w/LTPU=0.625)
6 2 Pole or 3 Pole only. QO 1 Pole is available up to 70A (and coordinates up to 70A)
7 2p only. QO 1P is available up to 70A (and coordinates up to 70A), QO 3 Pole is available up to 100A (and coordinates
up to 100A)


Greater than 20kA
10kA - 20kA
Less than 10kA
Test X/R
4.9
3.2
1.7
13
Document Number
0100DB0604
Table IIIB
I-Line/NQOD Panelboard Selective Coordination At 208Y/120V
0.01s 1000s
FEEDER SIZE /
REQUIRED
UPSTREAM (I-LINE)
DOWNSTREAM (NQOD)
PANELBOARD CIRCUIT
PANELBOARD AMPACITY
BREAKER SIZE
(A)
(A)
MAXIMUM AVAILABLE
FAULT CURRENT AT
UPSTREAM (I-LINE)
PANELBOARD
(kA RMS Sym.)1
10
100
100
22
65
110
225
125
225
10
22
65
10
22
65
10
150
225
22
65
10
22
175
225
65
UPSTREAM (I-LINE)
PANELBOARD
FEEDER CIRCUIT
BREAKER TYPE
DOWNSTREAM (NQOD)
PANELBOARD
BRANCH CIRCUIT
BREAKER TYPE
HD
PG
HD
PG
PJ
HG
FH
PG
HD
HD
HG
HD
PG
HD
LA
PG
PJ
HG
LH
PG
HD
JD
PG
HD
JD
LA
PG
PJ
HG
JG
LH
PG
JD
PG
JD
LA
PG
PJ
JG
LH
PG
JD
LA-MC
LA-MC
LA-MC
QO
QO
QO-VH
QO-VH
QO-VH
QH
QH
QH
QO
QO-VH
QH
QO
QO
QO-VH
QO-VH
QO-VH
QO-VH
QH
QH
QH
QO
QO
QO
QO-VH
QO-VH
QO-VH
QO-VH
QO-VH
QH
QH
QH
QH
QO
QO
QO-VH
QO-VH
QO-VH
QO-VH
QH
QH
QH
QO
QO
QO
QO
LA-MC
LA-MC
LA-MC
LA-MC
PG
JD
LA
PG
PJ
LA-MC
LA-MC
LA-MC
QO
QO
QO
QO
QO
QO-VH
QO-VH
QO-VH
QO-VH
QO-VH
QO-VH
QO-VH
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
JG
LH
LH-MC
PG
JD
QO-VH
QO-VH
QO-VH
QO-VH
QH
QH
QH
QH
QH
QO
10
200
225
22
42
65
225
14
225
10
MAXIMUM AVAILABLE
FAULT CURRENT AT
DOWNSTREAM (NQOD)
PANELBOARD1.8
1.3
10
1.3
21.6
22
1.3
0.9
65
1.3
1.3
1.3
1.3
10
1.3
3.2 (2P ONLY)
21.6
22
1.3
65 (1P), 3.2 (2P, 3P)
65
1.3
2.3
10
1.3
2.3
3.2 (2P ONLY)
21.6
22
1.3
2.4
65 (1P), 3.2 (2P, 3P)
65
2.3
10
2.3
3.2 (2P ONLY)
21.6
22
2.4
65 (1P), 3.2 (2P, 3P)
65
2.3
18 (1P, 2P), 16 (3P)
18 (1P, 2P), 10 (3P)
7 (1P), 10 (2P), 6.5
(3P)
7 (1P, 2P), 6 (3P)
6 (1P, 2P), 5.5 (3P)
5 (1P, 3P), 6 (2P)
5
10
2.3
3.2 (2P ONLY)
21.6
22
22 (1P, 2P), 16 (3P)
22 (1P, 2P), 10 (3P)
7 (1P), 10 (2P), 6.5
(3P)
7 (2P), 6 (3P)
6 (2P), 5.5 (3P)
6 (2P), 5 (3P)
5
3.4
2.4
65 (1P), 3.2 (2P, 3P)
3.4
65
2.3
LARGEST POSSIBLE
BRANCH CIRCUIT
BREAKER
(A)
20
40
20
403
403
20
25
30
25
25
25
25
70
25
25
603
603
25
25
30
25
70
70
25
30
30
703
703
25
30
30
30
70
70
403
403
703
703
30
30
30
70
15
20
30
40
50
70
1006
1006
503
503
803
803
15
20
30
403
503
703
1003
30
30
30
30
30
1006
Document Number
0100DB0604
FEEDER SIZE /
REQUIRED
UPSTREAM (I-LINE)
DOWNSTREAM (NQOD)
PANELBOARD CIRCUIT
PANELBOARD AMPACITY
BREAKER SIZE
(A)
(A)
MAXIMUM AVAILABLE
FAULT CURRENT AT
UPSTREAM (I-LINE)
PANELBOARD
(kA RMS Sym.)1
10
225
225
22
42
65
10
250
400
22
42
65
10
22
300
400
65
22
350
400
65
400
400
10
UPSTREAM (I-LINE)
PANELBOARD
FEEDER CIRCUIT
BREAKER TYPE
DOWNSTREAM (NQOD)
PANELBOARD
BRANCH CIRCUIT
BREAKER TYPE
PG
LA-MC
LA-MC
LA-MC
QO
QO
QO
QO
LA-MC
LA-MC
LA-MC
QO
QO
QO
LA-MC
QO
LA-MC
LA-MC
LA-MC
JD
LA
PG
PJ
LA-MC
LA-MC
LA-MC
QO
QO
QO
QO-VH
QO-VH
QO-VH
QO-VH
QO-VH
QO-VH
QO-VH
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
JG
LH
LH-MC
PG
JD
PG
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
JD
LA
PG
PJ
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
LA-MC
JG
LH
LH-MC
PG
PG5
LA
MG
PG5
LH
MG
PG5
LA
MG
LH
MG
LA-MC
LA-MC
LA-MC
PG
QO-VH
QO-VH
QO-VH
QO-VH
QO-VH
QO-VH
QO-VH
QH
QH
QH
QH
QH
QO
QO
QO
QO
QO
QO
QO
QO
QO
QO
QO-VH
QO-VH
QO-VH
QO-VH
QO-VH
QO-VH
QO-VH
QO-VH
QO-VH
QO-VH
QO-VH
QO-VH
QH
QH
QH
QH
QH
QO
QO-VH
QO-VH
QO-VH
QH
QH
QH
QO-VH
QO-VH
QH
QH
QO
QO
QO
QO
MAXIMUM AVAILABLE
FAULT CURRENT AT
DOWNSTREAM (NQOD)
1,8
PANELBOARD
10
18
18 (1P, 2P), 16 (3P)
11 (1P), 18 (2P), 8
(3P)
10 (1P, 2P), 7.5 (3P)
10 (1P, 2P), 7 (3P)
8 (1P), 10 (2P), 6.5
(3P)
7 (1P), 10 (2P), 6
(3P)
8 (2P), 6 (3P)
6
3.825
2.3
3.2 (2P ONLY)
21.6
22
22 (1P, 2P), 18 (3P)
22 (1P, 2P), 16 (3P)
11 (1P), 22 (2P), 8
(3P)
18 (2P), 7.5 (3P)
18 (2P), 7 (3P)
13 (2P), 6.5 (3P)
10 (2P), 6 (3P)
8 (2P), 6 (3P)
6
3.825
3.825
2.4
65 (1P), 3.2 (2P, 3P)
3.825
65
2.3
10
18
18 (1P, 2P), 14 (3P)
10
10 (1P, 2P), 9 (3P)
10 (1P, 2P), 8 (3P)
10 (1P, 2P), 7.5 (3P)
10 (2P), 7.5 (3P)
4.25
2.3
3.2 (2P ONLY)
21.6
22
22 (1P, 2P), 18 (3P)
22 (1P, 2P), 14 (3P)
18 (2P), 10 (3P)
18 (2P), 9 (3P)
13 (2P), 8 (3P)
11 (2P), 7.5 (3P)
10 (2P), 7.5 (3P)
4.25
4.25
2.4
65 (1P), 3.2 (2P, 3P)
4.25
65
10
3.2 (2P ONLY)
3.6 (3P ONLY)
21.6
65 (1P), 3.2 (2P, 3P)
65 (1P), 3.6 (2P, 3P)
65
3.2 (2P ONLY)
3.6 (3P ONLY)
65 (1P), 3.2 (2P, 3P)
65 (1P), 3.6 (2P, 3P)
18
10
6
10
LARGEST POSSIBLE
BRANCH CIRCUIT
BREAKER
(A)
1006
15
20
30
40
50
60
70
806
1006
1257
503
503
803
803
15
20
30
403
503
603
703
803
1003
1253
30
30
30
30
30
1006
1006
20
30
40
50
60
70
1006
1257
603
603
1103
1103
20
30
403
503
603
803
1003
1503
30
30
30
30
30
1257
903
1003
1253
30
30
30
1503
1503
30
30
30
1006
1257
1257
15
Document Number
0100DB0604
FEEDER SIZE /
REQUIRED
UPSTREAM (I-LINE)
DOWNSTREAM (NQOD)
PANELBOARD CIRCUIT
PANELBOARD AMPACITY
BREAKER SIZE
(A)
(A)
MAXIMUM AVAILABLE
FAULT CURRENT AT
UPSTREAM (I-LINE)
PANELBOARD
(kA RMS Sym.)1
22
400
400
42
65
10
22
450
600
65
10
22
500
500
65
10
22
600
600
65
UPSTREAM (I-LINE)
PANELBOARD
FEEDER CIRCUIT
BREAKER TYPE
LA
LA-MC
LA-MC
LA-MC
MG
PG
LA-MC
LH
LH-MC
MG
PG
PG5
MG
PG5
MG
PG5
PG5
MG
PG5
MG
PG5
PG (ET1.0)4, PG4
MG
PG (ET1.0)4, PG4
MG
PG (ET1.0)4, PG4
MAXIMUM AVAILABLE
FAULT CURRENT AT
DOWNSTREAM (NQOD)
PANELBOARD BRANCH DOWNSTREAM (NQOD)
PANELBOARD
(kA RMS Sym.)1,8
QO-VH
QO-VH
QO-VH
QO-VH
QO-VH
QO-VH
QH
QH
QH
QH
QH
QO
QO-VH
QO-VH
QH
QH
QO
QO-VH
QO-VH
QH
QH
QO
QO-VH
QO-VH
QH
QH
LARGEST POSSIBLE
BRANCH CIRCUIT
BREAKER
(A)
1503
30
1003
1503
1503
1503
30
30
30
30
30
1257
1503
1503
30
30
1257
1503
1503
30
30
1257
1503
1503
30
30
3.2 (2P ONLY)

22 (1P, 2P), 18 (3P)
22 (2P), 18 (3P)
6
3.6 (3P ONLY)
21.6
6
65 (1P), 3.2 (2P, 3P)
6
65 (1P), 3.6 (2P, 3P)
65
10
3.6 (3P ONLY)
21.6
65 (1P), 3.6 (2P, 3P)
65
10
3.6 (3P ONLY)
21.6
65 (1P), 3.6 (2P, 3P)
65
10
5.4 (3P ONLY)
21.6
65 (1P), 5.4 (2P, 3P)
65
1 Available fault currents are based upon system X/R ratios less than or equal to the circuit breaker test X/R ratio. See the explanatory note
below for additional information
2 480V-rated
3 2 Pole or 3 Pole only. QO-VH 1 Pole is available up to 30A (and coordinates up to 30A)
800A: 1000A w/LTPU=0.625)
6 2 Pole or 3 Pole only. QO 1 Pole is available up to 70A (and coordinates up to 70A)
7 2 Pole only. QO 1 Pole is available up to 70A (and coordinates up to 70A), QO 3 Pole is available up to 100A (and coordinates up to 100A)


Greater than 20kA
10kA - 20kA
Less than 10kA
Test X/R
4.9
3.2
1.7
16
0600DB0601
2/2007
Cedar Rapids, IA, USA
Data Bulletin
A Comparison of Circuit Breakers and Fuses for
Low-Voltage Applications
Tony Parsons, PhD, P.E.,
Square D / Schneider Electric Power Systems Engineering
I. Introduction
Recent claims by fuse manufacturers regarding the arc-flash and simplifiedcoordination benefits of fuses do not tell the entire story regarding which
type of device is best for a given power system. In reality, not only does
the wide range of available circuit breaker types allow them to be
successfully used on nearly any kind of power system, they can be applied
so as to provide selective coordination, arc-flash protection, advanced
monitoring and control features, all in a renewable device. This paper gives
a feature-by-feature comparison of the merits of circuit breakers vs. fuses,
discussing the relative merits of fuses and circuit breakers in each section.
While both circuit breakers and fuses are available for application in
systems that operate at higher voltage levels, the focus of this guide is on
low-voltage systems operating at 600 V or below.
II. Basic Definitions and

Requirements
Article 240 of the National Electrical Code (NEC) [1] provides the basic
requirements for overcurrent (i.e., overload, short-circuit, and/or ground
fault) protection in a power system. Special requirements for overcurrent
protection of certain types of equipment are also contained in other
articlesfor example, details on protection requirements for motors and
motor circuits are given in Article 430, while transformer protection
requirements are given in Article 450.
The NEC defines the two basic types of Overcurrent Protective Devices
(OCPDs):
fuseAn overcurrent protective device with a circuit-opening fusible
part that is heated and severed by the passage of overcurrent through it.
circuit breakerA device designed to open and close a circuit by
nonautomatic means and to open the circuit automatically on a
predetermined overcurrent without damage to itself when properly
applied within its rating.
The NEC also requires that circuits be provided with a disconnecting
means, defined as a device, or group of devices, or other means by which
the conductors of a circuit can be disconnected from their source of supply.
Since fuses are designed to open only when subjected to an overcurrent,
they generally are applied in conjunction with a separate disconnecting
means (NEC 240.40 requires this in many situations), typically some form of
a disconnect switch. Since circuit breakers are designed to open and close
under manual operation as well as in response to an overcurrent, a
separate disconnecting means is not required.
Both fuses and circuit breakers are available in a variety of sizes, ratings,
and with differing features and characteristics that allow the designer of an
electrical system to choose a device that is appropriate for the system under
consideration.
A Comparison of Circuit Breakers and Fuses for Low-Voltage Applications

Data Bulletin
0600DB0601
2/2007
Low-voltage fuses are available in sizes from fractions of an amp to

thousands of amps, at voltage ratings up to 600 V, and with short-circuit
interrupting ratings of 200 kA or more. Fuses are inherently single-pole
devices (i.e., an individual fuse can only operate to open one phase of a
multi-phase circuit), but two or three individual fuses can be applied together
in a disconnect to protect a multi-phase system. Low-voltage fuses are
tested and rated according to the UL 248 series of standards. Several types
can be classified as current-limiting, which per the NEC definition means
that they ...reduce the current flowing in the faulted circuit to a magnitude
substantially less than that obtainable in the same circuit if the device were
replaced with a solid conductor having comparable impedance. In other
words, the current-limiting fuses open very quickly (within 1/2 cycle) in the
presence of a high-level fault, allowing them to provide excellent protection
for distribution system components or load equipment. Fuses can be
applied in equipment such as panelboards, switchboards, motor control
centers (MCCs), disconnect switches/safety switches, equipment control
panels, etc.
Circuit breakers are also available with a wide range of ratings10 A to
thousands of amps, also with short-circuit interrupting ratings to 200 kA
and are available as 1, 2, 3, or 4-pole devices. The three basic types of LV
circuit breakers are the molded-case circuit breaker (MCCB), low-voltage
power circuit breaker (LVPCB), and insulated-case circuit breaker (ICCB).
MCCBs are rated per UL 489, have all internal parts completely enclosed in
a molded case of insulating material that is not designed to be opened
(which means that the circuit breaker is not field maintainable), and can be
applied in panelboards, switchboards, MCCs, equipment control panels,
and as stand-alone disconnects inside a separate enclosure. LVPCBs,
which are rated per ANSI standards and are applied in low-voltage drawout
switchgear, are larger, more rugged devices that may be designed to be
fully field maintainable. ICCBs can be thought of as a cross between
MCCBs and LVPCBsthey are tested per UL 489 but may share some
characteristics with LVPCBs, including two-step stored energy mechanism
availability in drawout construction and partial field maintainability [2].
Both types of OCPDs can meet the basic requirements of the NEC, but are
circuit breakers or fuses best suited for a particular application?
Unfortunately, there is no simple answer to this questionseveral other
factors must be taken into account, such as the level of protection provided
by the OCPD, selective coordination requirements, reliability, renewability,
and flexibility. The remainder of this guide will provide a discussion of each
of these topics.
III. System Protection
As discussed above, both circuit breakers and fuses meet the basic NEC
requirements for overcurrent protection of electric power distribution
systems and equipment. Any type of OCPD must be sized and installed
correctly after taking all derating factors and other considerations into
account. Particularly for overloads and phase faults, both circuit breakers
and fuses provide excellent protection and either is suitable for most
applications. A bit more consideration is warranted for some other aspects
of system protection, as discussed in the remainder of this section.
A. Ground-Fault Protection
Conventional wisdom states that the most common type of fault in a power
system (by far) is a single-phase-to-ground fault. On solidly-grounded power
systems, the available ground-fault current level can be significant. In some
situations, ground fault current levels that are even higher than the
maximum three-phase fault current level are theoretically possible.
However, many ground faults produce only relatively low levels of fault
current due to impedance in the fault path (due to arcing or to some other
2007 Schneider Electric All Rights Reserved
0600DB0601
2/2007

Data Bulletin
source of impedance from phase to ground). While such faults can cause
significant equipment and facility damage if not cleared from the system
quickly, phase overcurrent protective devices may not respond quickly to
the lower fault levelsif they detect the fault at all. For example, an 800 A
ground fault might simply appear as an unbalanced load to a 4000 A fuse or
circuit breaker not equipped with ground-fault protection. Because of this,
NEC 230.95 requires supplementary ground-fault protection on service
disconnects rated 1000 A or more on solidly-grounded, wye systems
operating at more than 150 V to ground but not more than 600 V phase-tophase (e.g., 277/480 V systems). The NEC also defines special ground-fault
protection requirements for health care facilities and emergency systems.
See the appropriate NEC articles for more details.
Circuit breakers can be equipped with integral ground-fault protection
through addition of either electronic trip units that act as protective relaying
to detect the ground fault and initiate a trip, or through addition of add-on
ground-fault protection modules. Ground-fault trip units typically use the
current sensors internal to the circuit breaker to detect the ground fault
condition, though an external neutral sensor is normally required to monitor
current flowing on the neutral conductor in a 4-wire system. If desired,
external relaying and current transformers (CTs) can also be used for
ground-fault detection provided that the circuit breaker is equipped with a
shunt trip accessory that can be actuated by the external relay.
By themselves, fuses cannot provide ground-fault protection except for
relatively high-level ground faults. When ground-fault protection is required
in a fusible system, the disconnecting means (usually a switch, sometimes a
contactor) must be capable of tripping automatically, and external relaying
and a zero-sequence CT or set of residually-connected phase CTs must be
installed to detect the ground faults and send the trip signal to the
disconnecting means.
While either system can function well if installed properly, extra care must
be taken with a fusible system (or circuit breaker-based system with
external ground relaying) to ensure that all external sensors are oriented
correctly and that all sensor and relay wiring is installed correctly.
Performance testing of the ground-fault system, as required in NEC
230.95(C) when the system is installed, should allow for identification of any
installation issues.
B. Device Interrupting Ratings
NEC 110.9 states that equipment intended to interrupt current at fault

levels shall have an interrupting rating sufficient for the nominal circuit
voltage and the current that is available at the line terminals of the
equipment. Protective devices that are inadequately rated for either the
system voltage or available fault current levels present a safety hazard, as
there is no guarantee that they will be able to interrupt faults without
damage either to themselves or to other equipment in the system. This
could result in extended downtime and present a significant fire hazard.
Several types of low-voltage fuses (class R, class J, etc.) carry interrupting
ratings of 200 kA or more at up to 600 V. This is typically high enough to
interrupt even the most severe fault in the stiffest system. In addition, since
fuses are single-pole devices, their single-pole interrupting capability equals
the full rating of the fuse. Note that the withstand rating of the equipment
(e.g., panelboards, switchboards) in which fuses are applied may not
always be equal to the ratings of the fuses themselvesequipment
manufacturers should be consulted, particularly when system fault currents
exceed 100 kA. Note also that some LV fuses have interrupting ratings as
low as 10 kA, so care should always be taken to ensure that fuses selected
are appropriate for the installation.

Data Bulletin
0600DB0601
2/2007
Circuit breakers of all types are also available with interrupting ratings up to
200 kA. In the not-too-distant past, fused circuit breakers were required to
achieve the 200 kA interrupting ratings, but modern circuit breakers can
achieve this rating without fuses. Circuit breakers with lower ratings are also
available, typically at a lower cost. Circuit breakers have single-pole
interrupting ratings that are adequate for installation on the majority of power
systems, though special consideration may be required in some cases. See
[3] for additional information.
C. Motor Protection
Overcurrent Protective Devices (OCPDs) in motor circuits have a relatively

difficult job to perform. They must not trip on motor inrush current, but should
be sensitive enough to provide both overload protection and short-circuit
protection to the motor and its associated branch circuit. In many cases, the
fuse/circuit breaker (or motor circuit protectorMCP which is essentially a
molded-case circuit breaker with no overload element), is oversized to
accommodate motor inrush current and a separate overload relay is added
that will open the motor contactor during overload conditions. These two
devices then combine to provide overload and short-circuit protection for the
motor circuit.
Motors can also be damaged by conditions other than short-circuits and
overloads. On three-phase systems, one of the most problematic abnormal
conditions is system voltage unbalance, which can cause an increase in
phase currents and create high negative-sequence currents that flow in the
motor windings. Both of these cause increased heating in the motor
windings, which can cause insulation degradation or breakdown that can
ultimately result in failure of the motor. Unbalance from system sources such
as unbalanced load in a facility or voltage unbalance on the utility system is
potentially problematic whether circuit breakers or fuses are used as motor
OCPDs. However, the use of fuses has the potential to produce a severe
unbalance condition commonly referred to as single-phasing.
Single-phasing occurs when one phase in a three-phase motor circuit opens
but the other two phases remain in service. If the single-phasing occurs
upstream of the motor but at the same voltage level, then zero current flows
on the phase with the open fuse and elevated current levels flow in one or
both of the remaining phases, depending on whether the motor is wye or
delta-connected. Single-phasing on the primary side of a transformer feeding
the motor can produce elevated currents in all three phases, with two being
slightly elevated and the third current roughly double that of the other two.
To help guard against motor damage or failure due to single-phasing:
Use a circuit breaker-based protection system. If properly maintained, all

three phases of a circuit breaker will open in response to a fault or overload,
so single-phasing in the facility will be far less likely to occur. However, note
that if the utility supply is protected by fuses, this possibility still exists.
Apply phase-failure or current unbalance relaying, either at the facility main

(in smaller installations) or at high-value loads (e.g., larger motors that are
more expensive to replace, critical loads where the downtime associated
with a motor failure cannot be tolerated, etc.)
Size motor circuit fuses closer to the full-load current rating of the motor.
One fuse manufacturer recommends sizing dual-element, time-delay fuses
at 100125% of the motor's actual load level (not the nameplate rating) to
provide better levels of protection against damage resulting from singlephasing [4]. Note that this does not eliminate the possibility of single-phasing
occurring, and could increase the possibility of nuisance fuse operation on
sustained overloads. In applications where loading on a particular motor
varies widely, or in new facilities where actual current draw of a motor may
not be known, sizing the fuses properly could be a challenge. Application of
external relaying at high-value loads may still be warranted.
0600DB0601
2/2007
D. Component Protection

Data Bulletin
One of the great advantages of a current-limiting overcurrent protective

device is that it can literally limit the peak magnitude of fault current that
flows through it by opening within the first half-cycle after fault initiation,
before the fault current has a chance to reach its peak value. This helps
provide a degree of protection for downstream equipment that could
otherwise be damaged by the magnetic or thermal effects produced by the
high-level faults. Several types of low-voltage fuses are current-limiting to
one degree or another. Highly current-limiting fuses for special applications,
such as semiconductor fuses that are designed to protect power electronic
equipment, are also available. Same is true of breakers, only that fuses are
often more current-limiting.
Current-limiting molded-case circuit breakers are also available in a range
of sizes and with interrupting ratings of 200 kA. As with current-limiting
fuses, these circuit breakers are tested to determine the peak-let-through
current (ip) and let-through energy (i2t). While these circuit breakers are not
as current-limiting as the faster-acting current-limiting fuses (e.g., class J or
class RK-1), they do provide a degree of protection beyond that of a noncurrent-limiting circuit breaker or fuse, and may be appropriate for many
applications.
Proper protection, whether of conductors, motors, or other equipment,
depends on OCPDs being applied appropriately. This includes ensuring that
devices are sized properly and that they are installed on systems where
none of the equipment ratings are violated.
To help prevent misapplication of fuses, NEC 240.60(B) requires that
fuseholders are designed to make it difficult to insert fuses intended for
application on higher amperage or lower voltage circuits. Additionally,
fuseholders intended for current-limiting fuses should reject insertion of a
non-current-limiting fuse.
Switchboards and panelboards where circuit breakers are applied do not
typically have rejection features that prevent installation of a circuit breaker
that is of a compatible frame type but that has a lower interrupting rating.
Realistically, any device can be improperly appliedand improper use of
protective devices is an application issue, not an equipment issue. In the
real world, inadequately-rated circuit breakers can be installed, fuses of a
given cartridge size but of a higher ampere rating can be installed into a
rejection fuseholder, fuses can be replaced with slugs (produced by the
manufacturer or of the homemade variety), or fuseholders or circuit
breakers can be jumpered out altogether by a creative electrician with a
relatively short length of wire. Proper selection, installation, and
maintenance of all OCPDs are all key requirements in providing good
system protection.

Data Bulletin
E. Arc-Flash Protection
0600DB0601
2/2007
With the increased interest in arc-flash hazards in recent years, the ability of
OCPDs to provide protection against arcing faults has received much
interest. The potential severity of an arc-flash event at a given location in a
power system depends primarily on the available fault current, the distance
of the worker away from the source of the arc, and the time that it takes the
upstream OCPD to clear the arcing fault from the system. In many cases,
little can be done about the first two factorsthe available fault current
levels depend on utility system contribution, transformer impedance values,
etc.; while the working distance is limited by the fact that a worker working
on a piece of equipment must, in most cases, be physically close to the
equipment.
Proper selection and application of OCPDs can have a great deal of impact
on the fault clearing time. Clearing the fault more quickly can provide a great
deal of protection for workers, as the available incident energy is directly
proportional to the duration of the arcing faulti.e., the incident energy can
be cut in half if the fault can be cleared twice as quickly. Equations
appearing in IEEE Standard 1584-2002 [5] provide the present state-ofthe-art methods for determining the arc-flash hazard levels in a system and
for evaluating the impact of potential arc-flash mitigation options.
For low-voltage systems, which OCPDs provide the best protection against
arc flash?
Circuit breakers, with adjustable trip units that can be set to strike a
balance between providing selective coordination and arc-flash
protection?
Current-limiting fuses, which can clear high-level faults very quickly and
minimize damage to both equipment and personnel?
Unfortunately, there is no simple answer to this question, despite claims

made by manufacturers of both types of OCPDs. In some cases, both circuit
breakers and fuses provide excellent protection. There are situations when
circuit breakers can perform better than fuses, and there are situations
where fuses can perform better than circuit breakers. And there are
situations where neither circuit breakers nor fuses provide much arc-flash
protection at all, requiring either use of other means of protection
(alternative system designs, installing systems that allow for remote
operation of equipment, etc.) or a total prohibition of work on or near
energized parts.
When evaluating OCPDs in terms of the arc-flash protection that they may
provide, three general principles are important to consider:
Evaluate Specific Devices
Evaluate specific devices when possible

Evaluate devices at the actual system fault current levels
Evaluate adjustable-trip circuit breakers at their chosen settings
The IEEE 1584 standard contains three basic calculation models that can
be used to determine arc-flash hazard levelsan empirically-derived,
general model; simplified equations based on testing of current-limiting
(class RK-1 and class L) low-voltage fuses; and simplified equations based
on calculations performed on typical low-voltage circuit breakers. The
general equations require information on available fault current levels in the
system as well as knowledge of the trip characteristics of OCPDs in the
circuit, but can provide accurate results for any type of OCPD and for a wide
range of system conditions. The simplified circuit breaker and fuse
equations require little to no knowledge of actual device trip characteristics,
but differences in the way these equations were developed mean that they
should not be used to conduct a direct apples-to-apples comparison of
specific protective devices.
0600DB0601
2/2007

Data Bulletin
As discussed above, the simplified fuse equations are based on field testing
of specific types of fuses, the simplified circuit breaker equations are based
on classes of circuit breakers and on the assumption that the relevant trip
settings are maximized, and not on specific devices or actual trip settings.
The circuit breaker equations are meant to allow calculation of the worstcase arc-flash levels allowed by any example of a circuit breaker within a
given classe.g., 100400 A MCCBs. If the IEEE 1584 empirical equations
are used to calculate arc-flash levels downstream of such a circuit breaker,
the values should never be higher than (and in many cases will be well
below) those shown by the simplified circuit breaker equations. This is
particularly true when using the equations to analyze larger LVPCBsthe
simplified IEEE 1584 equations assume that the circuit breaker's
instantaneous and/or short-time pickup and delay settings are set to the
maximum levels, which can result in the calculation of very conservative
arc-flash levels if the circuit breakers are actually set differently. For
example, Figure 1 shows the incident energy levels vs. bolted fault current
values for 2000 A circuit breakers in a 480 V, solidly-grounded system.
Figure 1:
Incident Energy vs. Bolted Fault Current for 2000 A Circuit

Breakers Simplified Equations vs. Actual Data
Incident Energy (cal/cm^2)
NW-LF
NW-L
LVPCB w/INST
LVPCB w/ST
120
90
60
30
0
0
20
40
60
80
100
120
Bolted Fault Current (kA)
The LVPCB w/ST and LVPCB w/INST curves are based on the IEEE
1584 simplified equations for low-voltage power circuit breakers with shorttime and instantaneous pickup, respectively. The NW-L and NW-LF
curves show arc-flash values based on actual devices (2000 A Masterpact
NW-L and NW-LF circuit breakers set to trip instantaneously for an arcing
fault, respectively).
As shown in the plot, the simplified equations (particularly for the LVPCB
w/ST curve) are well above the results calculated based on the actual
device characteristics. When possible, a comparison of the level of arc-flash
protection a given device can provide, should be based on actual device
characteristics, not generic equations.

Data Bulletin
What is the system fault current range?
0600DB0601
2/2007
Current-limiting fuses can provide excellent protection and reduce the

available incident energy downstream to minimal levels . . . as long as they
are operating within their current-limiting range. For lower fault current
levels, the arc-flash levels can elevate.
Thermal-magnetic MCCBs can provide excellent protection as long as they
trip instantaneously, but arc-flash levels can escalate for low-level faults that
require operation of the thermal element to clear the arc. For higher levels of
fault current, RK-1 and L fuses tend to allow a lower level of incident energy
than a similarly-sized circuit breaker, but both devices provide an excellent
level of protectionallowing for the use of Category 0 PPE in many cases.
For example, see Figure 2, which shows incident energy levels vs. bolted
fault current for a 400 A Square D LH circuit breaker, a 400 A Square D LC
circuit breaker, and a 400 A class RK-1 low-voltage fuse. The circuit
breakers are assumed to trip instantaneously.
Figure 2:
Incident Energy vs. Bolted Fault Current for 400 A Circuit

Breakers and 400 A Class RK-1 Fuses.
400A LH
400A LC
400A RK-1 Fuse
2.0
1.6
1.2
0.8
0.4
0.0
0
20
40
60
80
100
120
As shown in Figure 2, the relative performance of the circuit breakers is

better for low-level faults, while the incident energy allowed by the fuses is
lower for higher fault current levels. However, the incident energy levels for
each device over the entire range of fault currents considered is less than
2.0 cal/cm2 the maximum level allowed for Category 0 PPE [6], indicating
that both circuit breakers and fuses provide excellent protection.
For larger devices, the relative performance of circuit breakers and fuses
follows these same guidelines, though the impact can be quite a bit larger.
See Figure 3, which shows the incident energy levels allowed by 1600 A
Class L current-limiting fuses, as well as two varieties of 1600 A
Masterpact NW circuit breakers. Again, the circuit breakers are assumed to
trip instantaneously for an arcing fault so circuit breaker settings must be
considered, (see Consider Circuit Breaker Settings below), but this does
show that 1600 A circuit breakers can perform significantly better than fuses
for systems with relatively low available fault current levels.
0600DB0601
2/2007

Data Bulletin
Figure 3:
Incident Energy Comparison for 1600 A Protective Devices
NW-LF
NW-H
1600L Fuse
30
25
20
15
10
5
0
0
20
40
60
80
100
120
Consider Circuit Breaker Settings
For circuit breakers with adjustable trip settings, proper selection of setting
levels is important for both arc-flash protection and for system coordination.
The best protection will be provided when the circuit breakers can be set to
trip instantaneously. Little to no protection may be provided by a circuit
breaker when the settings are blindly set to maximum, as is sometimes
done after a nuisance trip of the device. Arc-flash studies can be
performed to determine optimum settings for circuit breakers and other
devices in a system, but even then, it may not be possible to reduce circuit
breaker settings below a certain level to provide additional arc-flash
protection if system coordination is to be maintained.
However, an adjustable circuit breaker still gives the flexibility to provide arcflash protection in such situations, if only on a temporary basis. For
example, the instantaneous pickup level of a circuit breaker feeding an MCC
can be turned down to the minimum setting when workers are present at the
MCC, then turned back up when work is complete. This could allow the
circuit breaker to trip instantaneously and provide the best possible level of
protection at the MCC when workers are present and exposed to the
hazard, while the normal setting allows for proper coordination during
normal operation. While this can provide an obvious benefit, it also has its
drawbacks, including:
Requirement for analysis to determine to what level the circuit breaker

settings should be reduced to provide additional protection, as well as what
level of protection is actually provided.
Uncertainty over how to provide arc-flash warning labels for such a

locationshould labels show the available incident energy and required
PPE with the normal circuit breaker settings, the reduced settings, or
both?
Temporary loss of selectivity can become semi-permanent if the circuit

breaker settings are not restored to normal when work is complete.
While a full discussion of issues surrounding arc-flash hazards and their

mitigation is beyond the scope of this paper, many other references are
available which discuss the subject in more depth, including [7] and [8].

Data Bulletin
IV. Selective Coordination
0600DB0601
2/2007
Selective coordination of overcurrent protective devices is required to

ensure that two somewhat mutually-exclusive goals are metfaults should
be cleared from the system as quickly as possible in order to minimize
damage to equipment, while the act of clearing the faults from the system
should interrupt power to as small a portion of the system as possible.
Selective coordination is defined in the NEC as localization of an
overcurrent condition to restrict outages to the circuit or equipment affected,
accomplished by the choice of overcurrent protective devices and their
ratings or settings. See the simple power system shown in Figure 4, which
will be used to illustrate a few example cases.
Figure 4:
Sample One-Line Diagram

Utility
Utility Transformer
Switchboard
Main
Switchboard
Feeder-A
Feeder-B
Panel-B
P
Transformer-A
S
Feeder-C
Chiller
Panel-A Main
Panel-A
Panel-C
Chiller Motor
Suppose that a foreign object produces a bus fault on the main switchboard.
The Switchboard Main circuit breaker will detect the fault, then open to clear
it from the systemand interrupt power to the entire facility in the process.
However, since there are no protective devices (not including those on the
utility system) upstream of the main circuit breaker, this device operates as
intended and coordination is not an issue. If the fault occurs at Panel-C
instead, then the Feeder-C circuit breakerand only the Feeder-C circuit
breakershould open to clear the fault. If so, then Feeder-C is said to be
selectively coordinated with both of the upstream OCPDs that would also
carry the fault current. If the switchboard main circuit breaker opens either
before or at the same time as Feeder-C, then power is unnecessarily
interrupted to other parts of the systemnamely, Panel-A, Panel-B, and the
Chiller Motorand the system is not selectively coordinated.
In some situations, even though individual devices are not coordinated, the
system may still be considered to be well-coordinated. Referring again to
Figure 4, consider a fault at Panel-A. The Feeder-A circuit breaker on the
primary side of the step-down transformer and the Panel-A Main circuit
breaker on the transformer secondary will typically not coordinate well with
each otherthat is, for a fault at the Panel-A main bus, either or both of the
panel main circuit breaker and the transformer feeder circuit breaker may
open to clear the fault. However, since the two devices are in series,
operation of either/both devices interrupts power to the exact same portion
10
0600DB0601
2/2007

Data Bulletin
of the power systemnamely, Panel-A. In this case, the system is

coordinated as long as the Feeder-A circuit breaker coordinates with the
switchboard main and the Panel-A Main circuit breaker coordinates with
branch devices in Panel-A, even though the two devices, strictly speaking,
do not coordinate with one another.
Selective coordination, while always desirable, is not required by the NEC
except in certain situations:
In health-care facilities, per NEC 517.17(C): Ground-fault protection for

operation of the service and feeder disconnecting means shall be fully
selective such that the feeder device, but not the service device, shall
open on ground faults on the load side of the feeder device.
In elevator circuits when more than one elevator motor is fed by a single
feeder. See NEC 620.62.
In emergency and legally-required standby power systems (including

those in hospitals and other health-care facilities where so required), per
NEC 700.27 and NEC 701.18.
The requirements for selective coordination in emergency and legallyrequired standby systems, new in the 2005 edition of the NEC, call for each
overcurrent device to be selectively coordinated with all supply side
overcurrent protective devices.
This requirement can be problematic for system designers because it
recognizes only device coordination and not system coordination, and
because it means that special consideration must be given to circuit
breaker-based systems.
Normally, coordination between devices on a time-current plot is
demonstrated by white space on the plot between the devicesideally,
the upstream device's trip curve will appear above and to the right of the
downstream device with no overlap between the curves. This indicates that
the downstream device would trip first when both saw the same fault. Any
overlap between devices indicates an area (i.e., a range of fault currents)
where it cannot be conclusively determined, at least from examination of the
plot, which device would trip first. For circuit breakers and relays, this
graphical comparison of trip characteristics is the primary way that system
coordination is assessed.
For fuses, coordination down to 0.01 second can be assessed by a
comparison of trip curves, while fuse let-through characteristics must be
compared to verify coordination beyond this point. Alternatively, tables
produced by fuse manufacturers show minimum ampere ratios between
pairs of load-side/line-side fuses that will insure coordinationfor fuses with
a 2:1 ratio, for example, the amp rating of the line-side fuse must be at least
2X the size of the load-side fuse for them to be properly coordinated.
11

Data Bulletin
0600DB0601
2/2007
Fuse manufacturers assert that fuses are often the only type of OCPD that
can truly be coordinated over all ranges of fault current, and that the fuse
ratio tables make selective coordination of fuses a simple prospect. While
this is true in some cases, things are not always this simple. Let us return to
the example system of Figure 4. Figure 5 that shows the time-current trip
characteristics for the Feeder-A and Panel-A Main circuit breakers.
A 125 A circuit breaker feeds the 480 V primary of the 75 kVA transformer,
while a 250 A main on the 208 V panel is selected.
Figure 5:
Time-Current Characteristics for Feeder-A and Panel-A

Main Circuit Breakers.
1000
100
FDR 'A'
Time in Seconds
10 PANEL 'A' MAIN
0.10
0.01
0.5 1
10
100
1K
10K
Current in Amperes
Figure 5 shows that the trip curves of the two circuit breakers overlap,
indicating a lack of coordination between them. If the fault current falls into
the range where the device curves overlap, it is unclear which will trip first.
However, as discussed above, since these devices are in series, system
coordination is preserved even though device coordination is not.
Unfortunately, a strict interpretation of NEC 700.27 and 701.18 does not
recognize system coordination, and so this series installation would be a
code violation if installed in an emergency or legally-required standby
system.
What if fuses were used instead? The fuse ratio tables do not address
coordination between devices operating at different voltage levels, as in this
case, so a graphical evaluation of coordination would be required. Selecting
a typical 125 A, class RK-1, 600 V fuse for the primary feeder, and a 250 A,
RK-1, 250 V fuse for the secondary main will result in overlap between the
two devices. The size of the primary fuse must be increased to 175 A for the
fuses to coordinate, at least for durations above 0.01 seconds. This still
meets the NEC requirements for transformer protection in NEC 450, but
could make coordination with upstream devices more difficult depending on
the system design.
12
0600DB0601
2/2007

Data Bulletin
Figure 6 shows the time-current characteristics of the Feeder-B and FeederC circuit breakers in Figure 4.
Figure 6:
Time-Current Characteristics for the Feeder-B and

Feeder-C Circuit Breakers.
1000
100
Time in Seconds
10
FDR 'B'
1 FDR 'C'
0.10
0.01
0.5 1
10
100
1K
10K
Current in Amperes
As shown in the plot, the two devicesa 600 A Square D LC circuit

breaker (Feeder-B) and a 200 A Square D LH circuit breaker (Feeder-C)
coordinate well, except for currents above approximately 4200 A where the
two device curves overlap. If a fault downstream of the Feeder-C circuit
breaker drew more than 4200 A fault current, both Feeder-B and Feeder-C
would try and respond instantaneously, and it is not clear from the timecurrent curve (TCC) plot which device would open first to clear the fault. In
many cases, this level of coordination between the circuit breakers (i.e., no
overlap except for relatively high-level faults) is considered to be
acceptable. However, it does not meet the requirements of NEC 700.27 or
701.18.
Does this mean that system designers have to use only fuses in emergency
systems? Not necessarily! In light of the new NEC requirements, Schneider
Electric has begun to re-evaluate the performance of its low-voltage circuit
breaker product line for the selectivity of specific combinations of circuit
breakers at high fault current levels. The test results have shown that in
many cases the published circuit breaker trip curves, due to dynamic
impedance and current limiting effects, are actually somewhat conservative
in the instantaneous region when considering selectivity between circuit
breakers, and that many line/load combinations of circuit breakers actually
do coordinate even if their trip curves indicate otherwise.
13

Data Bulletin
0600DB0601
2/2007
For example, see Figure 7, which shows the time-current characteristics for
two Square D thermal-magnetic circuit breakersan 800 A MJ and a
125 A EJB, both at 208 V.
Figure 7:
Trip Curves for 800 A MJ and 125A EJB.
1000
100
800A MJ
Time in Seconds
10
125A EJB
0.10
0.01
0.5 1
10
100
1K
10K
Current in Amperes
While the curve shows mis-coordination between the circuit breakers in the
instantaneous trip region, the test results presented in Data Bulletin
0100DB0501, Short-Circuit Selective Coordination for Low Voltage Circuit
Breakers, [9] indicates that this particular combination does actually
coordinate all the way up to 100 kA, the full interrupting rating of both
devices. Not all circuit breaker combinations tested coordinated this well
and some testing remains to be completed, but the fact is that fused
systems are not the only ones that can meet the strictest NEC requirements
for selective coordination.
Selective coordination may also be enhanced through simply designing the
power system (whether fuses or circuit breakers are used) with selective
coordination in mind. As examples of the latter, situations where OCPDs are
applied in series should be avoided as should application of devices
upstream/downstream of one another that are close in size (e.g., 800 A
panelboard with 600 A circuit breaker feeding a sub-panel), neither of which
lends itself to easy selective coordination between those devices. See Data
Bulletin 0100DB0403, Enhancing Short Circuit Selective Coordination with
Low Voltage Circuit Breakers [10] and [11] for additional discussion of
selective coordination in circuit breaker systems.
14
0600DB0601
2/2007
V. Reliability

Data Bulletin
Circuit breakers, being mechanical devices, require periodic maintenance to

ensure that they can operate within expected tolerances when called upon
to clear a fault or overload from a system. If a circuit breaker is not properly
maintained, it may still be able to operate as intended, it may operate more
slowly than intended, or it may not be able to operate at all. Following proper
maintenance and testing practices and using modern, durable circuit
breakers such as the Square D Masterpact NW, which is rated for up to
12,500 mechanical or 2,800 electrical operations before maintenance is
required, can help to ensure that circuit breakers will correctly operate when
called upon to do so and that potentially defective devices are found and
repaired or replaced before they create larger problems.
While fuses themselves require no maintenance, this does not mean that a
fusible system requires no preventative maintenance or testing. Fuse
holders, cable connections, and disconnect switches (whether manually or
automatically operated) must be periodically tested and maintained, just as
in circuit breaker systems. Neglecting periodic operation of such devices,
periodic maintenance requirements, and infrared scanning can lead to
switch contacts that have welded shut, hot spots at conductor
connections, etc.
If reliability and maintenance requirements of only the overcurrent protective
devices are considered, it is true that fuses have a clear advantage over
circuit breakers. In reality, however, both fusible and circuit breaker-based
systems require at least some degree of periodic maintenance, giving
neither type system a clear advantage in this area. For details on
recommended maintenance procedures and intervals, contact the
equipment manufacturer or see NFPA 70B, Recommended Practice for
Electrical Equipment Maintenance [12].
VI. Rerating
Both fuses and thermal-magnetic circuit breakers MCCBs operate based on

heating produced by overload or fault currents flowing through them. As a
result, the ambient temperature can have an effect on the trip characteristics
of both types of devices. Square D LV MCCBs will require rerating for
ambient temperatures above 40C. They are actually capable of carrying
higher-than-rated currents for ambient levels below 24C, which may
require special consideration to ensure proper conductor protection. See
Data Bulletin 0100DB0101, Determining Current-Carrying Capacity in
Special Applications [13]. Fuses may also require rerating above
approximately 25C, as the elevated ambient decreases both their effective
continuous current rating and opening time. Like MCCBs, fuses may carry
more than rated current in low-ambient environments, again possibly
meriting special consideration to ensure that conductor protection is
provided. The response time of thermal-based devices can also be affected
by pre-loading (i.e., heating produced by flow of current through an OCPD
before an overcurrent condition is present) and harmonic distortion (highfrequency distortion can be problematic for semiconductor fuses in
particular; the effect of harmonics on general-purpose fuses and MCCBs is
generally not a reason for concern). Use of electronic-trip circuit breakers
may be warranted when facing such difficult conditions, as trip units with
true RMS metering are relatively insensitive to harmonic current levels (at
least for lower-order harmonics), and ambient temperature levels do not
have an effect on Square D electronic-trip circuit breakers [13].
15

Data Bulletin
VII. Renewability
0600DB0601
2/2007
Fuses clear faults from the system by virtue of the melting of the fusible
element. Once that element has melted and current can no longer pass
through the fuse, the fault is removed from the system. This melting is a
one-way processthe fusible link can no longer carry current and must be
replaced. For non-renewable fuseson low-voltage systems, this
encompasses all but certain types of Class H fusesthis means that the old
fuse cartridge must be removed from the fuseholder and a new one installed
before the circuit can be re-energized. Even for renewable fuses, the fuse
link itself must be replaced. Stocking spare fuses can help keep potential
system downtime to a minimum, but can mean that a substantial inventory
of spare fuses must be maintained.
A circuit breaker, on the other hand, clears faults from the system through
opening of a set of contacts. As long as the circuit breaker does not sustain
damage in the process of clearing the overcurrent, the contacts can be reclosed and the circuit re-energized by manually closing the circuit breaker. A
circuit breaker should always be inspected after a high fault, and testing
may also be wiseparticularly if any damage or stress is seen when the
circuit breaker is inspectedto ensure that the device will function properly.
In many cases, and particularly if the circuit breaker is properly applied
within its ratings, the circuit can be re-energized after only minimal
downtime.
Fuse manufacturers have argued that the non-renewability of fuses is
actually an advantage over circuit breakers in some situations. OSHA
regulations state that:
After a circuit is de-energized by a circuit protective device, the circuit
may not be manually re-energized until it has been determined that the
equipment and circuit can be safely energized. The repetitive manual
reclosing of circuit breakers or reenergizing circuits through replaced
fuses is prohibited.
NOTE: When it can be determined from the design of the circuit and the
overcurrent devices involved that the automatic operation of a device
was caused by an overload rather than a fault condition, no examination
of the circuit or connected equipment is needed before the circuit is reenergized. (OSHA 1910.334(b)(2))
The argument is that since fuses must be replaced, the temptation for a
worker to simply reset a circuit breaker and re-energize the circuit (thereby
possibly violating OSHA regulations) is removed. Realistically, though, a
worker who is willing to bypass OSHA regulations and proper work practices
in order to quickly get a circuit back in service is just as likely to do this with
fused circuits as with circuits protected by circuit breakers. In the real
world, for better or for worse, installations have been found where a single
disconnect contains more than one type and/or size of fuse; fuses have
been jumpered out or replaced with solid copper or steel bars, etc. Likewise,
circuit breakers have been misapplied, bypassed, etc. The type of worker
who operates and maintains an electric power system can have just as
much, if not more, impact on its performance as the type of overcurrent
protective device that is used.
16
0600DB0601
2/2007

Data Bulletin
Replacing fuses involves working on or near exposed, energized

equipment, which per NFPA 70E-2004 is only allowed if de-energizing
creates additional or increased hazards or is infeasible due to equipment
design or operational limits. [6] Therefore, in most situations, replacing
fuses in a panelboard or switchboard would require that the entire
panel/switchboard be de-energized. If energized work can be justified per
130.1 of NFPA 70E-2004, appropriate flash protection PPE is still required.
While use of appropriate PPE is also recommended when switching circuit
breakers, as most power distribution equipment is not rated to contain
arcing faults (the exception being Arc Resistant gear), the NFPA 70E rules
governing energized work would not apply as long as the enclosure door
remains closed, as workers would not be exposed to energized parts. While
switching circuit breakers with equipment covers and doors in place is not
inherently safe, the fact that the worker is not exposed to energized parts
should help reduce the likelihood of occurrence of arc-flash events.
17

Data Bulletin
VIII. Flexibility
A wide variety of circuit breakers are availablefrom relatively basic

molded-case circuit breakers to the top of the line low-voltage power circuit
breakerswith optional features that make them appropriate for nearly any
application. A summary of some of the more advanced features available on
circuit breakers is provided in this section. Many of these features are not
available on fusible systems without addition of external metering
equipment, relays, or other accessories.
1, 2, 3, or 4-pole Construction: a circuit breaker is available that will fit

nearly any circuit, even those where providing neutral protection or
having a switched neutral may be of benefit. The switched neutral can
help to simplify ground-fault protection system design in multi-source
systems, for example.
Integral Ground-fault Protection available: no external relaying and only

minimal associated wiring required.
Adjustable Trip Characteristics: for all but the smallest MCCBs,

adjustable trip settings are available that can help provide optimal levels
of selective coordination and arc-flash protection in a system. Electronic
trip units provide the highest degree of setting flexibility.
Advanced Protection and Monitoring Features: when applied on a

Masterpact circuit breaker, the state-of-the-art Micrologic H trip units
can provide a wide range of protection and control/monitoring features,
including:
18
0600DB0601
2/2007
Neutral conductor protection

Demand current alarm/trip
Undervoltage alarm/trip
Overvoltage alarm/trip
Voltage unbalance alarm/trip
Current unbalance alarm/trip
Reverse power alarm/trip
Overfrequency alarm/trip
Underfrequency alarm/trip
Phase rotation alarm
Available control signal for load-shed schemes
Metering capabilities:
voltage
current
power
power factor
energy
harmonic distortion
waveform captures
Trip/alarm history: records type of fault, observed levels of trip

quantity (e.g., peak fault current level recorded)
Condition monitoring of circuit breaker: contact wear indicator
Support for communication protocols that allow trip unit to be tied
into facility-wide power monitoring or SCADA system
Subsets of these features also available with other circuit breaker/trip
unit types
Interrupting ratings available to 200 kA without fuses.
0600DB0601
2/2007
IX. References

Data Bulletin
[1] NFPA 70-2005, National Electrical Code, National Fire Protection

Association, Quincy, MA.
[2] IEEE Std. 1015-1997, IEEE Recommended Practice for Applying LowVoltage Circuit Breakers Used in Industrial and Commercial Power
Systems.
[3] Gregory, G. D., Single-pole Short-Circuit Interruption of Molded Case
Circuit Breakers, IEEE Transactions on Industry Applications, vol. 35,
no. 6, Nov.Dec. 1999, p. 1265-70.
[4] SPDSelecting Protective Devices (Based on the 2005 NEC), Cooper
Bussmann, Available: http://www.bussmann.com
[5] IEEE Std. 1584-2002, IEEE Guide for Performing Arc-Flash Hazard
Calculations.
[6] NFPA 70E-2004, Standard for Electrical Safety in the Workplace,
National Fire Protection Association, Quincy, MA.
[7] Square D Data Bulletin 0100DB0402, Arc-Flash Application Guide: Arcflash Calculations for Circuit Breakers and Fuses. Available:
http://www.us.squared.com
[8] Brown, W.A., Shapiro, R., A Comparison of Arc-Flash Incident Energy
Reduction Techniques using Low-Voltage Power Circuit Breakers,
presented at the 2006 IEEE Industrial and Commercial Power Systems
Technical Conference, Dearborn, MI.
[9] Square D Data Bulletin 0100DB0501, Short-Circuit Selective
Coordination for Low Voltage Circuit Breakers. Available:
[10] Square D Data Bulletin 0100DB0403, Enhancing Short Circuit Selective
Coordination with Low Voltage Circuit Breakers. Available:
[11] Square D Data Bulletin 0100DB0403 Guide to Power System Selective
Coordination 600 V and Below. Available: http://www.us.squared.com
[12] NFPA 70B-2006, Recommended Practice for Electrical Equipment
Maintenance, National Fire Protection Association, Quincy, MA.
[13] Square D Data Bulletin 0100DB0101, Determining Current-Carrying
Capacity in Special Applications. Available: http://www.us.squared.com
19

Data Bulletin
Schneider Electric USA

3700 Sixth St SW
Cedar Rapids, IA 52404 USA
1-888-SquareD (1-888-778-2733)
www.us.SquareD.com
0600DB0601
2/2007
Electrical equipment should be installed, operated, serviced, and maintained only by

qualified personnel. No responsibility is assumed by Schneider Electric for any
consequences arising out of the use of this material.

PRM DesignGuideComplete v1

Uploaded by

Copyright:

Available Formats

PRM DesignGuideComplete v1

Uploaded by

Document Information

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

PRM DesignGuideComplete v1

Uploaded by

Copyright:

Available Formats

Section 1:

Introduction and Basic Principles

Bill Brown, P.E., Square D Engineering Services

The purpose of this guide

Applications of electric power in industrial and commercial facilities

Semiconductor Preparation Processes

Pulp-and-Paper Preparation Processes

Water Treatment Processes

Basic design philosophy

Code Compliance: All applicable codes must be complied with.

E n h a n c e d R e l i a b i l i t y : The ability to maintain service continuity during abnormal system conditions.

E n h a n c e d S i m p l i c i t y : The power system is easy to understand and operate.

Figure 1-1: Power System Design Consideration Heuristics

Electric Power Fundamentals

Bill Brown, P.E., Square D Engineering Services

All angles are measured in degrees

is the A-phase voltage

Figure 2-1: Graphical representation of 3 voltages

is the RMS value of the periodic function f(t)

Figure 2-2: Plot for phasors per (2-15) - (2-17)

A very general representation of a 3 system is shown in figure 2-3:

Figure 2-3: General 3 system representation

Figure 2-4: DC Circuit element for power calculation

For the circuit element of figure 2-4 the following is true:

The relationship between P, Q, S, S and (v-j) can be shown graphically:

Figure 2-5: Graphical depiction of AC power

A.) The Ideal Transformer

Figure 2-6: Basic transformer model

From (2-35) and (2-36),

B.) A Practical Transformer Model

Figure 2-7: Practical transformer model

Figure 2-8: Standard transformer symbolic representation

C.) 3 Transformer Connections

Figure 2-9: Wye-Wye transformer connection

Figure 2-10: Delta-Delta transformer connection

Figure 2-11: Delta-Wye transformer connection

Figure 2-12: Wye-Delta transformer connection

Basic electrical formulae

is the DC voltage across the circuit element under consideration

B.) Passive Energy Storage Elements

is the voltage across the capacitor

Inductors store energy in the form of current, with governing equations:

is the voltage across the inductor

C.) AC Voltages and Currents, Time-Domain Form

All angles are measured in degrees

If the frequency is considered to be 60 Hz, 3 voltages can be written as:

is the A-phase voltage

The RMS value of a perfectly sinusoidal ac voltage or current is

C.) AC Power, Time-Domain Form

E.) AC Currents, Voltages and Circuit Elements, Frequency-Domain Form

is the AC voltage in frequency-domain form

Capacitor and Inductor impedances in frequency domain form are

is the impedance of the capacitor

is the system frequency

AC Voltage and current for an impedance Z are related as follows:

F.) Basic Transformers

Bill Brown, P.E., Square D Engineering Services

= maximum demand of load i, regardless of time of occurrence.

= total connected load of load group i

Load Development/Build-Up Schedule Peak load requirements, temporary/construction power