Top Banner
Table of contents Section 1: Power Systems Design – Introduction & Basic Principles Section 2: Electric Power Fundamentals Section 3: Load Planning – Basic Principles Section 4: System Voltage Considerations – Basic Principles Section 5: System Arrangements Section 6: Electrical System Grounding Section 7: Electrical System Protection Section 8: AC Motors, Motor Control & Motor Protection Section 9: Power Distribution Equipment Types Section 10: Emergency & Standby Power Systems Section 11: Power Quality Considerations Section 12: Arc Flash Hazard Considerations Section 13: Utility Interface Considerations Section 14: Electrical Energy Management Section 15: Project Coordination
258
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: @Electrical

Table of contents

Section 1: Power Systems Design – Introduction & Basic Principles

Section 2: Electric Power Fundamentals

Section 3: Load Planning – Basic Principles

Section 4: System Voltage Considerations – Basic Principles

Section 5: System Arrangements

Section 6: Electrical System Grounding

Section 7: Electrical System Protection

Section 8: AC Motors, Motor Control & Motor Protection

Section 9: Power Distribution Equipment Types

Section 10: Emergency & Standby Power Systems

Section 11: Power Quality Considerations

Section 12: Arc Flash Hazard Considerations

Section 13: Utility Interface Considerations

Section 14: Electrical Energy Management

Section 15: Project Coordination

Page 2: @Electrical

11

Section 1: Introduction and Basic PrinciplesBill Brown, P.E., Square D Engineering Services

IInnttrroodduuccttiioonnWith the increasing sophistication of modern power systems, it is easy to overlook the fact that the basic functionof a power distribution system has been the same for over 100 years – the safe, reliable distribution of power froma source to the connected loads. This basic function has not changed, although the complexity of the loadsthemselves, 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 optimalsystem design. Because the characteristics of each load, process, etc., served are unique, so too will each designbe unique in order to match the requirements imposed.

TThhee ppuurrppoossee ooff tthhiiss gguuiiddeeThis guide is intended to present the fundamentals of power system design for commercial and industrial powersystems. It is not designed as a substitute for educational background and experience in this area, nor is itdesigned to replace the multitude of detailed literature available about this subject. It does, however, bring intoone volume much material which has previously been available only by referencing a number of different sourceswith different formats and terminologies.

This guide is also intended to present the state of the art with regard to power system design for commercial andindustrial 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 commercialand industrial power systems, this guide can serve as a starting point for learning the different aspects of theprofession. For the licensed design professional, this guide does present a number guidelines in a handy andconvenient reference.

This guide is not intended to substitute for the services of a licensed design professional, but can be of aid whenworking with such professionals on commercial and industrial power system design.

AApppplliiccaattiioonnss ooff eelleeccttrriicc ppoowweerr iinn iinndduussttrriiaall aanndd ccoommmmeerrcciiaall ffaacciilliittiieessIn both industrial and commercial environments, electric power is used for a wide number of applications. Thefollowing is a brief list of the most common uses for electric power. This list is taken in part from[1], which providesan expanded treatment of this subject.

� Illumination – Whether for providing light for an office environment or a manufacturing shop floor, illumination isone of the most important applications of electric power, and the oldest.

� Environmental systems – Electric heating, ventilation, and air-conditioning are a large application for electricpower, and also an area in which electric power receives direct competition from other energy sources such asnatural 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 ofprocess applications:� Pumping

� Chemical Processes

� Semiconductor Preparation Processes

� Furnaces

[1] Standard Handbook for Electrical Engineers, New York: McGraw-Hill, 2001, pp. 21-1 - 21-99.

Page 3: @Electrical

2

� 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 reliablepower 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 pumpsare vital to any commercial or industrial facility.

� Communication Systems – Systems such as telephone and intrusion detection and monitoring arecritically important.

BBaassiicc ddeessiiggnn pphhiilloossoopphhyyThe following basic considerations are fundamental to any power system design:

� BBaassiicc SSaaffeettyy:: The power system must be able to perform all of its basic functions, and withstand basicabnormal conditions, without damage to the system or to personnel.

� BBaassiicc FFuunnccttiioonnaalliittyy:: The power system must be able to distribute power from the source to the connectedloads in a reliable manner under normal conditions.

� RReeaassoonnaabbllee CCoosstt:: The power system cost to obtain basic safety and functionality should be reasonable.

� CCooddee CCoommpplliiaannccee:: All applicable codes must be complied with.

Above and beyond the basics are a multitude of considerations, some of which will apply to each particularsystem design:

� EEnnhhaanncceedd SSaaffeettyy:: The ability to withstand extremely abnormal conditions with a minimum of risk to personnel.

� EEnnhhaanncceedd RReelliiaabbiilliittyy:: The ability to maintain service continuity during abnormal system conditions.

� EEnnhhaanncceedd MMaaiinnttaaiinnaabbiilliittyy:: The system can be maintained with minimum interruption to service and withminimum personnel protective equipment.

� EEnnhhaanncceedd FFlleexxiibbiilliittyy:: The ability to add future loads to the system, and with loads of a different nature thancurrently exist on the system.

� EEnnhhaanncceedd SSppaaccee EEccoonnoommyy:: The power system takes up the smallest possible physical space.

� EEnnhhaanncceedd SSiimmpplliicciittyy:: The power system is easy to understand and operate.

� RReedduucceedd CCoosstt:: The power system costs, both first cost and operating cost, are low.

� EEnnhhaanncceedd PPoowweerr QQuuaalliittyy:: The power system currents and voltages are sinusoidal, without large amounts ofharmonics present. System voltage magnitudes do not change appreciably.

� EEnnhhaanncceedd TTrraannssppaarreennccyy:: The power system data at all levels is easily acquired and interpreted, and the powersystem is easily interfaced with other building systems. Enhanced control of the system is also possible.

Page 4: @Electrical

3

While it should be the goal of every power system design to meet the above basic considerations, no systemdesign can yield all of the enhanced characteristics listed. The relationship between the considerations listed isshown in figure 1-1.

As can be seen, some of the enhanced characteristics mentioned are mutually exclusive, and to obtain acombination 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

Page 5: @Electrical

1

Section 2: Electric Power FundamentalsBill Brown, P.E., Square D Engineering Services

IntroductionAn understanding of the fundamentals of electric power is vital to successful power system design. It is assumedthat the reader has a degree in electrical engineering or electrical engineering technology, however the followingdiscussion is presented as review and reference material for completeness.

Basic ConceptsCommercial 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 beexpressed as follows:

(2-1)

(2-2)

where

v(t) is an AC voltage

i(t) is an AC current

Vmax is the voltage amplitude

Imax is the current amplitude

f is the system frequency

Øv is the voltage phase shift in degrees

Øi is the current phase shift in degrees

t 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 canbe 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 “3Ø”) power is characterized by three different phases, each with a phase shift 120 degrees from theother two phases. The three phases are typically referred to as “A”, “B”, and “C”. Further, the standard frequencyfor the United States is 60Hz. Therefore, three-phase voltages in the United States can be mathematicallydescribed as follows:

(2-3)

(2-4)

(2-5)

where

va(t) is the A-phase voltage

vb(t) is the B-phase voltage

vc(t) is the C-phase voltage

Page 6: @Electrical

2

The voltages from (2-3) - (2-5) are shown graphically in figure 2-1:

The peaks of the voltage waveforms are 120˚ (5.5 ms at 60 Hz) apart. Note that the peak of phase A occursbefore the peak of phase B, which in turn occurs before the peak of phase C. This is referred to as an ABC phasesequence 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 ofthe waveforms.

Because the full mathematical representation of AC voltages and currents is not practical, a shorthand notation isusually used. This shorthand notation treats the sinusoids as complex quantities based upon the followingmathematical 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 quantityfor a periodic function f(t) is defined as follows:

(2-10)

where

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

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)

0 12 24 t (ms)

Vmax

VaVbVc

Figure 2-1: Graphical representation of 3Ø voltages

Page 7: @Electrical

Because the RMS value is so useful in the calculation of power-related quantities, any time an AC voltage orcurrent 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 quantitiesknown as phasors:

(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 andcalculation, 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:

In most instances the Re and Im axes are omitted since the definition of time zero (and thus angle zero) isarbitrary; 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 acounter-clockwise direction about the 0,0 point it can be seen that the peak of va(t), represented by the tip ofphasor 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 acounter-clockwise phasor rotation. If angles are defined as positive in the clockwise direction a clockwise phasorrotation would indicate an ABC phase sequence. Both are encountered in practice. In this guide all angles inphasor diagrams will be assumed to be positive in the counter-clockwise direction.

3

tjrmsa eVV 21600=

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

Page 8: @Electrical

4

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

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 discussionabove 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 phasesand 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 groundfault current as discussed in subsequent sections of this guide. In practice, when 3Ø voltages are discussed, theyare assumed to be phase-to-phase voltages unless otherwise noted.

AC powerWith the basic concepts per above, AC electrical power can be described.

Consider the following DC circuit element:

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

(2-18)

where

Vdc is the DC voltage across the circuit element under consideration, with polarity as shown

Idc is the DC current through the circuit element under consideration, considered positive for the direction shown

P is the power generated by, or dissipated through, the circuit element under consideration

Figure 2-3: General 3Ø system representation

Figure 2-4: DC Circuit element for power calculation

Page 9: @Electrical

5

The sign of P in (2-18) is dependent upon the direction of current flow with respect to the DC voltage. A positivevalue for P indicates power dissipated, while a negative value for P indicates power generated. DC power ismeasured 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 underconsideration, with and AC current and voltage as defined by (1-1) and (1-2) respectively. The expression for theinstantaneous power, after some manipulation, is:

(2-19)

Thus, the instantaneous power consists of two parts: A DC component and an AC component with a frequencytwice that of the system frequency. The quantity (Øv - Øi) is defined as the power angle or power factor angle andis the angle by which the current peak lags behind the voltage peak on their respective waveforms. The quantityP= cos(Øv-Øi) 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(Øv-Øj) will always be less than VrmsIrms unless (Øv-Øj) = 0.Further, if (Øv-Øj) = ±90˚ , as is the case with a purely inductive or capacitive load, VrmsIrms cos(Øv-Øj) = 0.Because energy is required to force current to flow, and energy is always conserved, AC power must haveanother component. This component is most easily defined if AC power is treated as a complex quantity. To dothis, 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 isdone 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. ratherthan Watts. S is the Apparent Power and is also expressed in Voltamperes.

Page 10: @Electrical

6

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

The depiction in figure 2-5 is referred to as the power triangle since P, Q and S form a right triangle. It is alsoimportant to note that the power factor angle is the same as the load impedance angle of the circuit. The powerfactor is referred to as a lagging power factor if the current lags the voltage (i.e., (Øv-Øl) is positive up to 90˚)and as a leading power factor if the current leads the voltage (i.e., (Øv-Øl) is negative down to -90˚). For a laggingpower factor, the real and reactive power flow in the same direction; for a lagging power factor they flow inopposite directions. Of the passive circuit elements, resistors exhibit a unity power factor, inductors exhibit a zeropower factor lagging, and capacitors exhibit a zero power factor leading.

The foregoing discussion considers only single-phase circuits. For 3Ø circuits the power quantities for all threephases 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 canbe represented as S1Ø, S1Ø, P1Ø, and Q1Ø; equations (2-27) - (2-30) can then be simplified as:

(2-31)

(2-32)

(2-33)

(2-34)

Figure 2-5: Graphical depiction of AC power

Page 11: @Electrical

7

TransformersTransformers are vital components for AC power systems. They are used to change the voltage and currentmagnitudes to suit the application.

A.) The Ideal TransformerTransformers are relatively simple devices that utilize Faraday’s law of electromagnetic induction. In its simplestform, this law can be written:

(2-35)

where ξ is the voltage induced in a coil of N turns that is linked by a magnetic flux ψ .

In turn, the magnetic flux ψ for a coil of N turns which through which a current I passes and linked by a magneticpath with reluctance ℜ can be expressed as:

(2-36)

Consider the simple transformer shown in the following figure:

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

(2-37)

(2-38)

(2-39)

(2-40)

(2-41)

Dividing (2-40) by (2-41),

(2-42)

dt

dN

ψξ −=

Figure 2-6: Basic transformer model

Page 12: @Electrical

8

Equations (2-38) and (2-42) are the basic equations for a single-phase transformer. The voltage ratio (V1/V2) isequal 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 powerflowing out of the transformer.

The usefulness of the transformer lies in the fact that it can adjust the voltage and current to the application. Forexample, on a transmission line it is advantageous to keep the voltage high in order to be able to transmit thepower with as small a current as possible, in order to minimize line losses and voltage drop. At utilizationequipment, it is advantageous to work with low voltages that are more conducive to equipment design andpersonnel safety.

Another important aspect of the transformer is that it changes the impedance of the circuit. For example, if animpedance 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 outputwinding multiplied by the square of the turns ratio.

Figure 2-7: Practical transformer model

B.) A Practical Transformer ModelThe idealized transformer model just presented is not sufficient for practical electric power applications due to thefact that the core is not lossless and not all of the magnetic flux links both sets of windings. To take this intoaccount, a more realistic model is used:

Page 13: @Electrical

9

The resistance Rc represents the core losses due to hysteresis, and inductance Lc represents the magnetizinginductance. 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 quickcalculations, the core losses and magnetizing inductance are often ignored, and the model is treated as animpedance in series with an ideal transformer.

To insure the proper polarity, the circuit representation for a transformer includes polarity marks as shown in figure2-8. If the current for one winding flows into its terminal with the polarity mark, the current for the other windingflow 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.

C.) 3Ø Transformer ConnectionsTo be useful in 3Ø systems transformers must be connected for use with 3Ø voltage. This is accomplished by theuse of 3Ø transformer connections.

The wye-wye connection is shown in figure 3-9. This could be a bank of three single-phase transformers or one3Ø transformer which consists of all three sets of windings on a common ferromagnetic core. Polarity markingsfor three single-phase transformer connections are shown at the individual transformers, and polarity markings fora 3Ø 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 ofthe line-to-neutral voltage multiplied by √3.

For convenience the transformer turns ratio is taken as 1:1 on the phasor diagram.

If a three-phase transformer is used, the wye-wye connection has the disadvantage of requiring a four-leggedcore to allow for a magnetic flux imbalance. Further, the solidly-grounded neutrals allow for ground currents to flowthat can create interference in communications circuits [2]. Both the primary and secondary neutrals terminalsmust be solidly-grounded to allow for triplen-harmonic currents to flow; if the neutrals are allowed to float harmonicovervoltages will be developed from phase to neutral on each winding. These overvoltages can damage the

Figure 2-8: Standard transformer symbolic representation

Figure 2-9: Wye-Wye transformer connection

Page 14: @Electrical

10

transformer insulation. Wye-wye transformers are often used on systems above 25 kV to minimize a problemknown as ferroresonance. Ferroresonance is a condition which results from the transformer magnetizingimpedance resonating with the upstream cable charging capacitance, resulting in destructive overvoltages as thetransformer core moves into and out of saturation in a non-linear manner. Single-phase switching is usually thecause 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 remainingtransformers can still provide three-phase service at a reduced capacity (57.7% of the capacity with all threetransformers in service).

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 3Ø transformers and 3Ø banks of single-phase transformers are alwaysreferenced as the phase-to-phase magnitude. Therefore, for a delta-wye transformer the winding turns ratios foreach 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-phasevoltages 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.3Ø transformers do not require a four-legged core like the wye-wye connection, but the advantages of a wyesecondary winding (elaborated on in section 6 of this guide) are obtained. Further, the secondary neutral can beleft unconnected in this arrangement, unlike the wye-wye arrangement.

Figure 2-10: Delta-Delta transformer connection

Figure 2-11: Delta-Wye transformer connection

Page 15: @Electrical

11

The wye-delta connection is shown in figure 2-12. This connection is seldom used in commercial and industrialapplications. Note that the delta is arranged differently from the delta-wye connection, in order to satisfy therequirement from [1] to have the phase-to-phase voltages on the low-voltage side of the transformer lag thecorresponding voltages on the primary side by 30˚.

Basic electrical formulaeThe following formulae are given as a convenient reference for the reader. These formulae include both formulaederived 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 is the DC voltage across the circuit element under consideration

Idc is the current through the circuit element under consideration. Idc is considered positive if it flows fromthe circuit element terminal at the higher voltage to the terminal at the lower voltage

Rdc is the DC resistance of the circuit element under consideration, measured in ohms

P is the power dissipated or generated by the circuit element. A positive power from (2-49) indicatespower dissipated by the circuit element, and a negative value indicates power generated by the circuitelement. The sign of P in (2-50) is lost due to the squaring of the current or voltage

Figure 2-12: Wye-Delta transformer connection

Page 16: @Electrical

12

B.) Passive Energy Storage ElementsCapacitors store energy in the form of voltage, with a governing equations:

(2-51)

(2-52)

where

vc is the voltage across the capacitor

ic is the current through the capacitor, considered positive if it flows toward the terminal from which vc is referenced

C is the capacitance value of the capacitor, measured in Farads

E is the energy stored in the capacitor

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

(2-53)

(2-54)

where

vi is the voltage across the inductor

ii is the current through the inductor, considered positive if it flows toward the terminal from which vI is referenced

L is the inductance value of the inductor, measured in Henries

E is the energy stored in the inductor

CC..)) AACC VVoollttaaggeess aanndd CCuurrrreennttss,, TTiimmee--DDoommaaiinn FFoorrmmSingle-phase AC voltage and current can be expressed as follows

(2-55)

(2-56)

where

v(t) is an AC voltage

i(t) is an AC current

Vmax is the voltage amplitude

Imax is the current amplitude

f is the system frequency

Øv is the voltage phase shift in degrees

Øi is the current phase shift in degrees

t is the time in ms

All angles are measured in degrees

Page 17: @Electrical

13

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

(2-57)

(2-58)

(2-59)

where

va(t) is the A-phase voltage

vb(t) is the B-phase voltage

vc(t) is the C-phase voltage

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

(2-60)

(2-61)

CC..)) AACC PPoowweerr,, TTiimmee--DDoommaaiinn FFoorrmmThe instantaneous single-phase AC power resulting from a current per (2-56) flowing through a circuit elementwith 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 FormAssuming only the real part is kept when converting to time-domain and that the same frequency appliesthroughout, AC currents and voltages can be written in the frequency domain as

(2-65)

(2-66)

where

V is the AC voltage in frequency-domain form

I is the AC current in frequency-domain form

Øv is the voltage phase shift

ØI is the current phase shift

Page 18: @Electrical

14

Capacitor and Inductor impedances in frequency domain form are

(2-67)

(2-68)

where

Zc is the impedance of the capacitor

Zi is the impedance of the inductor

j = √-1

f is the system frequency

C is the capacitance of the capacitor

L 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 3Ø 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 asfollows by considering the power quantities per phase to be S1Ø, S1Ø, P1Ø, and Q1Ø:

(2-79)

(2-80)

(2-81)

(2-82)

Page 19: @Electrical

15

F.) Basic TransformersThe input voltage and current V1 and I1 and the output voltage and current V2 and I2 for an ideal single-phasetransformer 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 thetransformer output terminals by the following equation:

(2-85)

ReferencesBecause the subject matter for this section is basic and general to the subject of electrical engineering, it isincluded in most undergraduate textbooks on basic circuit analysis and electric machines. Where material isconsidered 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.

Page 20: @Electrical

1

Section 3: Load PlanningBill Brown, P.E., Square D Engineering Services

Basic PrinciplesThe most vital, but often the last to be acquired, pieces of information for power system design are the loaddetails. 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 demandupon 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, usually15 min., 30 min., or 1 hour based upon the particular utility’s demand interval. Demand may be expressed inamperes, 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 bethe 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 asone-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 totalconnected 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 asystem to the maximum demand of the whole system, i.e.,

(3-2)

where

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

DG = 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 = total connected load of load group i

DFi = the demand factor of load group i

Page 21: @Electrical

2

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 requirementsif 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:

� 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 demandfactors for different types of loads. In addition, the National Electrical Code® [3] gives minimum requirements forthe computation of branch circuit, feeder, and service loads.

NEC Basic branch circuit requirementsNEC [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 mustbe 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 otherpower 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 systemof the premises served.

Outlet: The point on the wiring system at which current is taken to supply utilization equipment.

Page 22: @Electrical

3

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 receptacleis 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.

� 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)

a See NEC Article 220.14(J)b 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:):� The full load current rating for a single motor used in a continuous duty application is 125% of the motor’s

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).

� The load calculation for several motors, or a motor(s) and other loads, is 125% of the full load current ratingof the highest rated motor per a.) above plus the sum of the full-load current ratings of all the other motors inthe group, plus the ampacity required for the other loads (Article 430.24).

Type of Occupancy Unit Load Volt-Amperes Per quare Meter

Unit Load Volt-Amperes per Square Foot

Armories and auditoriums 11 1

Banks 39b 3.5b

Barber shops and beauty parlors 33 3

Churches 11 1

Clubs 22 2

Court Rooms 22 2

Dwelling Unitsa 33 3

Garages – commercial (storage) 6 0.5

Hospitals 22 2

Hotels and motels, including apartment houses without provision for cooking by tenantsa

22 2

Industrial commercial (loft) buildings 22 2

Lodge rooms 17 1.5

Office buildings 39b 3.5b

Restaurants 22 2

Schools 33 3

Stores 33 3

Warehouses (storage) 3 0.25

In any of the preceding occupancies except one- family dwellings andindividual dwelling units of two-family and multi-family dwellings:

Assembly halls and auditoriums 11 1

Halls, corridors, closets, stairways 6 0.5

Storage Spaces 3 0.25

Page 23: @Electrical

4

� For hermetic refrigerant motor compressors or multi-motor equipment employed as part of air conditioning orrefrigerating equipment, the equipment nameplate rated load current should be used instead of the motorhorsepower rating (Article 440.6).

� Luminaires (lighting fixtures) (Article 220.14(D)): An outlet supplying luminaire(s) shall be calculated based uponthe 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 aminimum of 600 volt-amperes.

� Sign and outline lighting (Article 220.14(F)): Sign and outline lighting loads shall be calculated at a minimum of1200 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:� The unit load per outlet as required in other provisions of article 220.14.

� 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 hotelsor motels must be calculated as follows (Article 220.14(H)):� 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.

� Where appliances are likely to be used simultaneously, each 300mm (1 ft.) or fraction thereof must beconsidered 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:� 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 includedin the general lighting load per above. No additional load calculations are required for these.

� Banks and office buildings (Article 220.14(K)): Receptacle outlets must be calculated to be the larger ofeither 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 thebranch 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:� Two or more 20A small-appliance branch circuits (Article 210.11(C)(1)).

� 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 leastthe 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), thebranch 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 applicablesizing per above.

Page 24: @Electrical

5

NEC Basic Feeder Circuit Sizing RequirementsOnce the branch circuit loads are calculated, the feeder circuit loads may be calculated by applying demandfactors to the branch circuit loads.

� General Lighting Loads (Article 220.42): The feeder general lighting load can be calculated by multiplying thebranch 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)

� Show window or track lighting (Article 220.43): Show windows must use a calculated value of 660 volt-amperes per linear meter (200 volt-amperes per linear foot), measured horizontally along its base. Tracklighting in other than dwelling units must be calculated at an 150 volt-amperes per 660mm (2 ft.) of lightingtrack or fraction thereof.

� Receptacles in other than dwelling units (Article 220.44): Demand factors for non-dwelling receptacle loadsare given in table 3-3 (NEC table 220.44).

Table 3-3: Demand factors for non-dwelling receptacle loads (NEC [3] table 220.44)

Type of Occupancy Portion of Lighting Load to Which Demand Factor Applies(Volt-Amperes)

Demand Factor(Percent)

Dwelling units First 3,000 or less atFrom 3,001 to 120,000 atRemainder over 120,000 at

1003525

Hospitals* First 50,000 or less atRemainder over 50,000 at

4020

Hotels and motels, including apartment houses withoutprovision for cooking by tenants*

First 20,000 or less atFrom 20,001 to 100,000 atRemainder over 100,000 at

504030

Warehouses (storage) First 12,500 or less atRemainder over 12,500 at

10050

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.

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:� 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 othermotors in the group, plus the ampacity required for the other loads (Article 430.24).

� The load calculation for factory-wired multimotor and combination-load equipment should be based upon theminimum 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 theduty cycles of the motors. No demand factors are given in the NEC for this situation.

Page 25: @Electrical

6

� 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 usesimultaneously, it is permissible to use only the largest loads that will be used at one time to be used incalculating the feeder demand.

� Feeder neutral load (Article 220.61): The feeder neutral load is defined as the maximum load imbalance on thefeeder. The maximum load imbalance for three-phase four-wire systems is the maximum net calculated loadbetween the neutral and any one ungrounded conductor. A demand factor of 70% may be applied to thiscalculated 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 benecessary 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 feedercircuit load calculation, the continuous portion of the load should therefore be multiplied by 1.25 unless theovercurrent device for the circuit is 100%-rated. Note that motor loads are not included in this calculation as the125% 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. Thisdata 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 NECitself when in doubt about a specific load sizing application. Computer programs are commercially available toautomate the calculation of feeder and branch circuit loads per the NEC methodology described above.

ReferencesBecause the subject matter for this section is basic and general to the subject of electrical engineering, it isincluded in most undergraduate textbooks on basic circuit analysis and electric machines. Where material isconsidered 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.

Page 26: @Electrical

1

Section 4: System Voltage ConsiderationsBill Brown, P.E., Square D Engineering Services

Basic PrinciplesThe selection of system voltages is crucial to successful power system design. Reference [1] lists the standardvoltages 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 arereferred 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 highvoltage,” 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 Four-wire

Low Voltage240480600

208 Y/120240/120

480 Y/277

Medium Voltage 2,4004,1604,8006,900

13,800

23,000

34,50046,00069,000

4,160 Y/2400

8,320 Y/480012,000 Y/6,93012,470 Y/7,20013,200 Y/7,62013,800 Y/7,97020,780 Y/12,00022,860 Y/13,200

24,940 Y/14,40034,500 Y/19,920

High Voltage 115,000138,000161,000230,000

Extra-High Voltage 345,000500,000765,000

Ultra-High Voltage 1,100,000

Page 27: @Electrical

2

The choice of service voltage is limited to those voltages which the serving utility provides. In most cases only onechoice of electrical utility is available, and thus only one choice of service voltage. As the power requirementsincrease, 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 ofthe various options would be required to arrive at the correct choice. In general, the higher the service voltage themore expensive the equipment required to accommodate it will be. Maintenance and installation costs alsoincrease with increasing service voltage. However, equipment such as large motors may require a service voltageof 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 definedby 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 orload-tap changing transformers at the service equipment to give adequate voltage regulation. This situationtypically only occurs for service voltages above 34.5 kV, however it can occur on voltages between 20 kV and34.5 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 commercialfacilities this will be 480 Y/277 V, although 208 Y/120 V is also required for convenience receptacles and smallmachinery. Large motors may require 4160 V or higher. Distribution within a facility may be 480 Y/277 V or, forlarge distribution systems, medium voltage distribution may be required. Medium voltage distribution implies amedium voltage (or higher) service voltage, and will result in higher costs of equipment, installation, andmaintenance than low voltage distribution. However, this must be considered along with the fact that mediumvoltage 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 givenload requirement for a single service, as shown in table 4-2.

Voltage drop considerationsBecause all conductors exhibit an impedance to the flow of electric current, the voltage will not be constantthroughout the system, but will tend to drop as one moves closer to the load. Ohm’s Law, expressed in phasorform for AC circuits, gives the basic relationship for voltage drop vs. the load current:

(4-1)

where

Vdrop is the voltage drop along a length of conductor or across a piece of equipment in volts

Il is the load current in amperes

Zcond 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 NECarticle 210.19(A)(1), FPN No. 4. Because this is a note only, it is not a requirement per se but is the commonlyaccepted guideline.

Page 28: @Electrical

3

Table 4-2: Power equipment design limits to service voltage vs. load requirements, for a single service

Because conductor impedance increases with the length of the conductor, it can be seen that unless the powersource is close to the center of the load the voltage will vary across the system, and, further, it can be more costlyto maintain the maximum voltage drop across the system to within 5% of the service voltage since largerconductors must be used to offset longer conductor lengths.

Voltage (V) Equipment Type Maximum EquipmentAmpacity (A)

Maximum Load (kVA)

208480600

Switchboard or Low VoltagePower Switchgear

5000 1,8004,1575,196

2,4004,1604,800

Metal-Enclosed Switchgear,w/Fuses 69,000 1080

4,4897,7828,979

6,9008,32012,00012,47013,20013,800

Metal-Enclosed Switchgear,w/Fuses

720 8,60510,37614,96515,55116,46117,210

20,78022,86023,00024,940

Metal-Enclosed Switchgear,w/Fuses

175 6,2996,9296,9727,560

34,500 Metal-Enclosed Switchgear,w/Fuses

115 6,872

2,4004,1604,8006,9008,32012,00012,47013,20013,800

Metal-Clad Switchgear 3000 12,47121,61624,94238,85343,23262,35464,79668,58971,707

20,78022,86023,00024,940

Metal-Clad Switchgear 2000 71,98479,18979,67486,395

Page 29: @Electrical

4

Also from equation (4-1) it can be seen that as load changes, so does the voltage drop. For a given maximumload, a measure of this change at a given point is the voltage regulation, defined as

(4-2)

where

Vno load is the voltage, at a given point in the system, with no load current flowing from that point to the load.

Vload 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 dropper 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 toavoid 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.

Page 30: @Electrical

1

Section 5: System ArrangementsBill Brown, P.E., Square D Engineering Services

IntroductionThe 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 figuresfor each of these assume that the distribution and utilization voltage are the same, and that the service voltagediffers from the distribution/utilization voltage. The symbology (low voltage circuit breaker, low voltage drawoutcircuit breaker, medium voltage switch, medium voltage breaker) reflects the most commonly-used equipment foreach arrangement. The symbology used throughout this section is shown in figure 5-1:

Radial systemThe radial system is the simplest system topology, and is shown in figure 5-2. It is the least expensive in terms ofequipment first-cost. However, it is also the least reliable since it incorporates only one utility source and the lossof the utility source, transformer, or the service or distribution equipment will result in a loss of service. Further, theloads must be shut down in order to perform maintenance on the system. This arrangement is most commonlyused 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.

Figure 5-1: Symbology

Page 31: @Electrical

2

Radial system with primary selectivityThis arrangement is shown in figure 5-3. If two utility sources are available, it provides almost the same economicadvantages of the radial system in figure 2 but also gives greater reliability since the failure of one utility sourcewill not result in a loss of service (note that an outage will occur between the loss of the primary utility source andswitching to the alternate source unless the utility allows paralleling of the two sources). The loss of thetransformer or of the service or distribution equipment would still result in a loss of service. Maintenance on thesystem requires all loads to be shut down.

An automatic transfer scheme may optionally be provided between the two primary switches toautomatically switch from a failed utility source to an available source. Most often metal-clad circuitbreakers are used, rather than metal-enclosed switches, if this is the case. More about typicalequipment application guidelines follows in a subsequent section of this guide.

Figure 5-2: Radial System

Figure 5-3: Radial System with Primary Selectivity

Page 32: @Electrical

3

Expanded radial systemThe 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 lossof service for all of the loads.

Figure 5-4 shows an expanded radial system utilizing multiple substations, but still with only one utility source andonly one primary feeder:

A more reliable and maintainable arrangement utilizing multiple primary feeders is shown in figure 5-5. In thesystem 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 endsfor maintenance purposes.

Typical service entrance equipment consists of a metal-clad switchgear main circuit breaker and metal-enclosedfused feeder switches. Metal-Clad circuit breakers may be used instead of metal-enclosed feeder switches if required.

Figure 5-4: Expanded Radial System with one utility source and a single primary feeder

Figure 5-5: Expanded Radial System with one utility source and multiple primary feeders

Page 33: @Electrical

4

Figure 5-6 shows an expanded radial system utilizing multiple substations and two utility sources, again withmetal-clad primary switchgear but with a duplex metal-enclosed switchgear for utility source selection:

Of the arrangements discussed this far, the arrangement of figure 5-6 is the most reliable – it does not dependupon a single utility source for system availability, nor does the failure of one transformer or feeder cause a loss ofservice to the entire facility. However, the loss of a transformer or feeder will result in the loss of service to a partof the facility. More reliable system arrangements are required if this is to be avoided.

Figure 5-6: Expanded Radial System with two utility sources and multiple primary feeders

Loop systemThe loop system arrangement is one of several arrangements that can allow one system component, such as atransformer 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-mentionedarrangements are that a failure of one feeder cable will not cause one part of the facility to experience a loss ofservice and that one feeder cable can be maintained without causing a loss of service (note that an outage to partof the system will be experienced after the failure of a feeder cable until the loop is switched to accommodate theloss of the cable).

In figure 5-7 metal-clad circuit breakers are used as the feeder protective devices. Fused metal-enclosed-feederswitches could be utilized for this, but caution must be used if this is considered since the feeder fuses wouldhave to be able to serve both transformers and the feeder and transformer fuses would have to coordinate formaximum selectivity.

It must be noted that the system arrangement of figure 5-7 is designed to be operated with the loop open, i.e., oneof the four loop switches shown would be normally-open. If closed-loop operation were required, metal-clad circuitbreakers 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 ofthe facility can be accomplished with metal-enclosed loop switches, however, if caution is used in the systemdesign and maintenance.

Page 34: @Electrical

5

Secondary-Selective systemAnother method of allowing the system to remain in service after the failure of one component is the secondary-selective 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 lossof service to one part of the plant. This is a characteristic none of the previously-mentioned system arrangementsexhibit. The system can be run with the secondary bus tie breaker normally-open or normally-closed. If the bus tiebreaker is normally-closed the failure of one transformer, if directional overcurrent relays are supplied on thetransformer secondary main circuit breakers, will not cause an outage, however care must be taken in the systemdesign 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 circuitbreakers, both for reasons of coordination and maintenance. However, a low voltage switchboard may be utilizedalso if care is taken in the system design and the system coordination is achievable. For a normally-closed bus tiebreaker, low voltage power switchgear is essential since the breakers lend themselves more readily externalprotective relaying.

Note that if one transformer fails the other transformer and its associated secondary main circuit must carry theentire load. This must be taken into account in sizing the transformer and secondary switchgear for this type ofsystem to be effective.

Figure 5-7: Primary Loop System

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 tiebreakers except one (the sparing transformer secondary main/tie breaker) normally-open. The sparing transformer

Page 35: @Electrical

6

secondary main/tie breaker) normally-open. The sparing transformer supplies one load bus if a transformer fails oris taken off-line for maintenance. A transformer is switched out of the circuit by opening its secondary mainbreaker and closing the tie breaker to allow the sparing transformer to feed its loads. The sparing transformer maybe allowed to feed multiple load busses if it is sized properly. Care must be used when allowing multipletransformers to be paralleled as the fault current is increased with each transformer that is paralleled, anddirectional 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 indifferent locations from one another. A properly-designed interlocking system will allow for the addition of futuresubstations without modification of the existing interlocking.

With both types of secondary-selective system, an automatic transfer scheme may be utilized to switch between afailed transformer and an available transformer.

Primary-Selective systemA selective system arrangement may also utilize the primary system equipment. Such an arrangement is shown infigure 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 benormally-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 notcause the entire system to be taken off-line. The available fault current with the tie breaker normally closedincreases 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 5-9: Transformer Sparing Scheme

Figure 5-10: Primary-Selective System

Page 36: @Electrical

7

Secondary Spot-Network systemIn large municipal areas where large loads, such as high-rise buildings, must be served and a high degree ofreliability is required, secondary network systems are often used. In a secondary network system several utilityservices 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 backfedthrough the low voltage system. These devices are designed to automatically isolate a faulted transformer whichis backfed from the rest of the system. The transformers typically have higher-than-standard impedances toreduce the available fault current on the low voltage network. The common secondary bus is often referred to asthe “collector bus.” An example of a secondary spot-network system is shown in figure 5-11.

Ring Bus systemEssentially a loop system in which the loop is normally closed, the ring bus is a highly reliable systemarrangement. 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 isrecommended for optimum reliability with this scheme. The bus differential relaying will open both breakersfeeding 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 incorporateadditional utility sources. As with the primary-selective system with a normally-closed bus tie breaker, the availablefault current is increased with each utility source added to the system.

Figure 5-11: Secondary Spot Network

Figure 5-12: Primary Ring Bus System

Page 37: @Electrical

8

Composite systemsThe above system arrangements are the basic building blocks of power distribution system topologies, but arerarely used alone for a given system. To increase system reliability it is usually necessary to combine two or moreof 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 withoutloss of service to either load bus (but with an outage to part of the system until the system is switched toaccommodate the failure). In addition, a single section of the primary loop or one transformer can be taken out ofservice 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 tiebreaker to form an even more reliable system, as shown in figure 5-14. With this arrangement, the failure of asingle utility source, a single primary circuit breaker, a single loop feeder cable, or a single transformer can beaccommodated without loss of service. And, any one primary circuit breaker, any one section of the primarydistribution 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.

A logical expansion of this system, resulting in a further increase in system reliability, can be had by replacing theprimary distribution loop with dedicated feeder circuit breakers from each primary bus, as shown in figure 5-15. Inthis system arrangement multiple primary feeder cable failures can be accommodated without jeopardizing serviceto 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-selectiveconfiguration shown in figure 5-3 and a variant of the transformer sparing scheme given in figure 5-9. This systemarrangement gives good flexibility in switching for maintenance purposes, and also allows any one utility, primaryswitchgear bus, or transformer fail without loss of service to any of the loads (again, an outage may be taken untilthe system is switched to accommodate the failure, depending upon the failure under consideration). It also allowsany three primary feeders to be faulted without loss of service to any of the loads. Other composite arrangementsare possible.

Figure 5-13: Composite System – Primary Loop/Secondary-Selective

Page 38: @Electrical

9

SummaryVarious system arrangements have been presented in this section, starting with the least complex andprogressing to a very complex, robust system arrangement. In general, as reliability increases so does complexityand cost. It must be remembered that economic considerations will usually dictate how complex a systemarrangement can be used, and thus will have a great deal of impact on how reliable the system is. Tables 5-6 and5-7 show the features of each system arrangement given in this section.

Figure 5-14: Composite System – Primary Selective/Primary Loop/Secondary Selective

Figure 5-15: Composite System – Primary Double-Selective,/Secondary Selective

Figure 5-15: Composite System – Primary Double-Selective,/Secondary Selective

Page 39: @Electrical

10

Please note that the formulas given in these tables are for the systems as shown in the figures above. They willhold true for expanded versions of these system arrangements where the expansion is made symmetricallywith respect to the configuration shown. They will not hold true when modifications are made to the systemarrangements with respect to symmetry, with altered numbers of switching/protective devices, or for concurrentfailures 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 thesame 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 theya guarantee of loss of service to a particular load after a component failure while the system is being switched toan alternate configuration. However, they are a guide to the relative strengths and weaknesses of each of thesystem arrangements presented.

Table 5-6: Power system arrangement summary for the basic arrangements as shown in this section

U = Number of Utility SourcesPB = Number of Primary Circuit BreakersSF = Number of Primary FeedersT = Number of TransformersSB = Number of Secondary Main and Tie Circuit Breakers$ = Relative Cost, with $=Least Expensive

Arrangement UtilityFailuresAllowed

Pri. BkrFailuresAllowed

Pri. FeederFailuresAllowed

TransformerFailuresAllowed

Sec.Main/Tie

BkrFailuresAllowed

Cost

Radial 0 0 0 0 0 $

Radial w/ Primary Selectivity U-1 � 0 0 0 0 $+

Expanded Radial, Single Primary Feeder

0 0 0 0 0 $$

Expanded Radial, Multiple Primary Feeders

0 0 0 0 0 $$

Expanded Radial, Multiple Utility Sources,Multiple Primary Feeders

U-1 � 0 0 0 0 $$+

Primary Loop System 0 1 1 0 0 $$$

Secondary-Selective System 0 0 0 1 1 $$$

Transformer Sparing Scheme 0 0 0 Varies;Maximum of T-1

T � $$$$

Primary Selective U-1 �,�Δ PB-F-U �,�,ø 0 0 0 $$$$$

Secondary Spot Network U-1 �,�,�,† PB-1 �,�,�,† F-1 �,�,�,† T-1 �,�,�,† SB-1 �,�,�,† $$$$$

Primary Ring Bus U-1 �,�,� U �,�,ø,� 0 0 0 $$$$$$

� Assumes that each utility source has sufficient capacity to supply the entire system.� Assumes that all secondary circuit breakers, including feeder breakers, are interchangeable.� Assumes that each primary main and bus tie (if applicable) circuit breakers has sufficient capacity to

supply the entire system.ø 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.� Assumes that the ring bus has sufficient capacity to supply the entire system.

Page 40: @Electrical

11

Table 7: Power system arrangement summary for the composite arrangements as shownin this section

U = Number of Utility SourcesPB = Number of Primary Circuit BreakersSF = Number of Primary FeedersT = Number of TransformersSB = Number of Secondary Main and Tie Circuit Breakers$ = Relative Cost, with $=Least Expensive

Arrangement UtilityFailuresAllowed

Pri. BkrFailuresAllowed

Pri. FeederFailuresAllowed

TransformerFailuresAllowed

Sec.Main/Tie

BkrFailuresAllowed

Cost

Primary Double-Selective /Secondary-Selective

U-1 �,� PB-F/2-U�,�,ø

F/2 T-1 † T-1 †,� $$$$$$$$

Primary Ring Bus / Primary-Selective/Secondary-Selective

U-1 �,�,� PB-F/2-U+1�,�,ø,�

F/2 T-1 † T †,� $$$$$$$$+

� Assumes that each utility source has sufficient capacity to supply the entire system.� Assumes that all secondary circuit breakers, including feeder breakers, are interchangeable.� Assumes that each primary main and bus tie (if applicable) circuit breakers has sufficient capacity to

supply the entire system.ø 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.� Assumes that the ring bus has sufficient capacity to supply the entire system.

Page 41: @Electrical

1

Section 6: System GroundingBill Brown, P.E., Square D Engineering Services

IntroductionThe topic of system grounding is extremely important, as it affects the susceptibility of the system to voltagetransients, determines the types of loads the system can accommodate, and helps to determine the systemprotection requirements.

The system grounding arrangement is determined by the grounding of the power source. For commercial andindustrial 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 theupstream 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 transformersecondary winding configuration.

D Static Power Converter – For devices such as rectifiers and inverters, the system grounding is determined bythe 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 separately-derived 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 thereare two common ways of connecting device windings: wye and delta. These two arrangements, with their systemvoltage relationships, are shown in figure 6-1. As can be seen from the figure, in the wye-connected arrangementthere are four terminals, with the phase-to-neutral voltage for each phase set by the winding voltage and theresulting 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 beexplained further below.

Figure 6-1: Wye and delta winding configurations and system voltage relationships

Page 42: @Electrical

2

Solidly-grounded systemsThe 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-to-neutral 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:

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 ofthe power system, such as equipment frames, are grounded, and the rest of the environment essentially is atground potential also, this has big implications for the system. It means that the line-to-ground insulation level ofequipment 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 connectedbetween 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 alsoon most utility distribution systems.

While the solidly-grounded wye system is by far the most common solidly-grounded system, the wye arrangementis not the only arrangement that can be configured as a solidly grounded system. The delta system can also begrounded, as shown in figure 6-3. Compared with the solidly-grounded wye system of figure 6-2 this systemgrounding arrangement has a number of disadvantages. The phase-to-ground voltages are not equal, andtherefore the system is not suitable for single-phase loads. And, without proper identification of the phases there isthe risk of shock since one conductor, the B-phase, is grounded and could be mis-identified. This arrangement isno longer in common use, although a few facilities where this arrangement is used still exist.

The delta arrangement can be configured in another manner, however, that does have merits as a solidly-grounded system. This arrangement is shown in figure 6-4. While the arrangement of figure 6-4 may not appear atfirst 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

Figure 6-2: Solidly-Grounded Wye System arrangement and voltage relationships

Figure 6-3: Corner-Grounded Delta System arrangement and voltage relationships

Page 43: @Electrical

3

used for small services which require both 240 VAC three-phase and 120/240 VAC single-phase. Note that thephase A voltage to ground is 173% of the phase B and C voltages to ground. This arrangement requires the BCwinding to have a center tap.

A common characteristic of all three solidly-grounded system shown here, and of solidly-grounded systems ingeneral, is that a short-circuit to ground will cause a large amount of short-circuit current to flow. This condition isknown as a ground fault and is illustrated in figure 6-5. As can be seen from figure 6-5, the voltage on the faultedphase 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 willsupport a large ground fault current is an important characteristic of this type of system grounding and does affectthe 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.

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. Typicalutility practice is to ground the neutral at many points, usually at every line pole, creating a multi-grounded neutralsystem. Because a separate grounding conductor is not run with the utility line, the resistance of the earth limitsthe circulating ground currents that can be caused by this type of grounding. Because separate groundingconductors are used inside a commercial or industrial facility, multi-grounded neutrals not preferred for powersystems in these facilities due to the possibility of circulating ground currents. As will be explained later in this

Figure 6-4: Center-Tap-Grounded Delta System arrangement and voltage relationships

Figure 6-5: Solidly-Grounded System with a ground fault on phase A

Page 44: @Electrical

4

section, multi-grounded neutrals in NEC jurisdictions, such as commercial or industrial facilities, are actuallyprohibited 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 loadsmust be supplied, and has the most stable phase-to-ground voltage characteristics. However, the large groundfault currents this type of system can support, and the equipment that this necessitates, are a disadvantage andcan be hindrance to system reliability.

Ungrounded systemsThis 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 inherentcharging capacitance to ground. To illustrate this point and its effect on the system voltages to ground, the deltawinding 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˚ (bothare reasonable operations since the voltage magnitudes and phase angles for the phase-to-phase voltage werearbitrarily chosen), the results are the same voltage relationships as shown in figure 6-4 for the solidly-groundedwye 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 theungrounded delta system. This becomes important when the effects of a ground fault are considered. The lack ofa grounded system neutral also makes this type of system unsuitable for single-phase phase-to-neutral loads.

In figure 6-7, the effects of a single phase to ground fault are shown. The equations in figure 6-7 are notimmediately practical for use, however if the fault impedance is assumed to be zero and the system capacitivecharging impedance is assumed to be much larger than the phase impedances, these equations reduce into aworkable 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 achange in the system phase-to-ground voltages. The phase-to-ground voltage on the faulted phase is zero, andthe phase-to-ground voltage on the unfaulted phases are 173% of their nominal values. This has implications forpower equipment – the phase-to-ground voltage rating for equipment on an ungrounded system must be at leastequal the phase-to-phase voltage rating. This also has implications for the methods used for ground detection, asexplained later in this guide.

Figure 6-6: Ungrounded Delta System winding arrangement and voltage relationships

Page 45: @Electrical

5

The ground currents with one phase is faulted to ground are essentially negligible. Because of this fact, from anoperational standpoint ungrounded systems have the advantage of being able to remain in service if one phase isfaulted to ground. However, suitable ground detection must be provided to alarm this condition (and is required inmost cases by the NEC [1] as described below). In some older facilities, it has been reported that this type ofsystem has remained in place for 40 years or more with one phase grounded! This condition is not dangerous inand of itself (other than due to the increased phase-to-ground voltage on the unfaulted phases), however if aground fault occurs on one of the ungrounded phases the result is a phase-to-phase fault with its characteristiclarge 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 resonantground fault condition occurs when the inductive reactance of the ground-fault path approximately equals the

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

Page 46: @Electrical

6

system capacitive reactance to ground. Arcing introduces the phenomenon of current-chopping, which can causeexcessive overvoltages due to the system capacitance to ground.

The ground detection mentioned above can be accomplished through the use of voltage transformers connectedin wye-broken delta, as illustrated in figure 6-9.

In figure 6-9, three ground detection lights “LTA,” “LTB” and “LTC” are connected so that they indicate the A, B andC phase-to-ground voltages, respectively. A master ground detection light “LTM” indicates a ground fault on anyphase. With no ground fault on the system “LTA,” “LTB” and “LTB” will glow dimly. If a ground fault occurs on onephase, the light for that phase will be extinguished and “LTM” will glow brightly along with the lights for the othertwo phases. Control relays may be substituted for the lights if necessary. Resistor “R” is connected across thebroken-delta voltage transformer secondaries to minimize the possibility of ferroresonance. Most ground detectionschemes 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 notwhere 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 asdiscussed below. However, older ungrounded systems are occasionally encountered.

Figure 6-9: A Ground Detection method for ungrounded systems

B

A

C

VT

VT

VT

LT A

LT B

LT C

LT M

GROUND FAULT

LOCATION LTA LTB LTC LTM

PHASE A

PHASE B

PHASE C

NONE DIM DIM DIM OFF

OFF BRIGHT BRIGHT BRIGHT

BRIGHT OFF BRIGHT BRIGHT

BRIGHT BRIGHT DIM BRIGHT

R

High-resistance grounded systemsOne ground arrangement that has gained in popularity in recent years is the high-resistance groundingarrangement. For low voltage systems, this arrangement typically consists of a wye winding arrangement with theneutral connected to ground through a resistor. The resistor is sized to allow 1-10 A to flow continuously if aground fault occurs. This arrangement is illustrated in figure 6-10.

Figure 6-10: High-Resistance Grounded System with no ground fault present

Page 47: @Electrical

7

The resistor is sized to be less than or equal to the magnitude of the system charging capacitance to ground. Ifthe resistor is thus sized, the high-resistance grounded system is usually not susceptible to the large transientovervoltages that an ungrounded system can experience. The ground resistor is usually provided with taps toallow 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 wyesystem, as shown in figure 6-10. However, if a ground fault occurs on one phase the system response is asshown 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 approximatelyequal to the phase-to-neutral voltage of the faulted phase divided by R. The faulted phase voltage to ground inthat case would be zero and the unfaulted phase voltages to ground would be 173% of their values without aground fault present. This is the same phenomenon exhibited by the ungrounded system arrangement, exceptthat the ground fault current is larger and approximately in-phase with the phase-to-neutral voltage on the faultedphase. The limitation of the ground fault current to such a low level, along with the absence of a solidly-groundedsystem neutral, has the effect of making this system ground arrangement unsuitable for single-phase line-to-neutral loads.

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-resistancegrounded 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 groundresistor is shown with a tap between two resistor sections R1 and R2. When a ground fault occurs, relay 59 (theANSI standard for an overvoltage relay, as discussed later in this guide) detects the increased voltage across theresistor. 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 contactP to close and re-open approximately once per second. When P closes R2 is shorted and the “pulse” indicator isenergized. R1 and R2 are sized so that approximately 5-7 times the resistor continuous ground fault current flowswhen 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 bedetermined. 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 groundfaults, but is generally more expensive than the ungrounded system ground fault indicator in figure 6-10.

Figure 6-11: High-Resistance Grounded System with a ground fault on one phase

Page 48: @Electrical

8

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.

Reactance groundingIn industrial and commercial facilities, reactance grounding is commonly used in the neutrals of generators. Inmost generators, solid grounding may permit the level of ground-fault current available from the generator toexceed the three-phase value for which its windings are braced [2]. For these cases, grounding of the generatorneutral through an air-core reactance is the standard solution for lowering the ground fault level. This reactanceideally limits the ground-fault current to the three-phase available fault current and will allow the system to operatewith phase-to-neutral loads.

Low-resistance grounded systemsBy sizing the resistor in figure in 6-11 such that a higher ground fault current, typically 200-800 A, flows during aground fault a low-resistance grounded system is created. The ground fault current is limited, but is of highenough magnitude to require its removal from the system as quickly as possible. The low-resistance groundingarrangement is typically used in medium voltage systems which have only 3-wire loads, such as motors, wherelimiting damage to the equipment during a ground fault is important enough to include the resistor but it isacceptable to take the system offline for a ground fault. The low-resistance grounding arrangement is generallyless expensive than the high-resistance grounding arrangement but more expensive than a solidly groundedsystem arrangement.

Creating an artificial neutral in an ungrounded systemIn some cases it is required to create a neutral reference for an ungrounded system. Most instances involveexisting 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.

Figure 6-12: Pulsing Ground Detection System

Figure 6-13: Medium Voltage implementation for high-resistance grounding

Page 49: @Electrical

9

The solution is a grounding transformer. Although several different configurations exist, by far the most popular incommercial and industrial system is the zig-zag transformer arrangement. It uses transformers connected asshown in figure 6-14:

The zig-zag transformer will only pass ground current. Its typical implementation on an ungrounded system, inorder to convert the system to a high-resistance grounded system, is shown in figure 6-15. The zig-zagtransformer distributes the ground current IG equally between the three phases. For all practical purposes thesystem, from a grounding standpoint, behaves as a high-resistance grounded system.

The solidly-grounded and low-resistance grounded systems can also be implemented by using a groundingtransformer, depending upon the amount of impedance connected in the neutral.

NEC system grounding requirementsThe National Electrical Code [1] does place constraints on system grounding. While this guide is not intended tobe a definitive guide to all NEC requirements, several points from the NEC must be mentioned and are basedupon 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 equipmentand 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.

Figure 6-14: Zig-Zag grounding transformer arrangement

Figure 6-15: Zig-Zag grounding transformer implementation

Page 50: @Electrical

10

Effectively Grounded: Intentionally connected to earth through a ground connection or connections ofsufficiently low impedance and having sufficient current-carrying capacity to prevent the buildup of voltages thatmay 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 agrounding electrode or electrodes.

Equipment Grounding Conductor: The conductor used to connect the non-current-carrying metal parts ofequipment, raceways and other enclosures to the system grounded conductor, grounding electrode conductor, orboth, 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 groundingconductor at the service.

System Bonding Jumper: The connection between the grounded circuit conductor and the equipment groundingconductor at a separately-derived system.

Grounding Electrode: The conductor used to connect the grounding electrode(s) to the equipment groundingconductor, 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 equipmentgrounding conductor, to the grounded conductor, or to both, at the service, at each building or structure wheresupplied 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 non–current-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 systemthrough normally non–current-carrying conductors, equipment, or the earth to the electrical supply source.

Effective Ground-Fault Current Path: An intentionally constructed, permanent, low-impedance electricallyconductive path designed and intended to carry current under ground-fault conditions from the point of a groundfault on a wiring system to the electrical supply source and that facilitates the operation of the overcurrentprotective 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 exceedsthe 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 damagingline-to-ground fault currents by operating to cause a disconnecting means to open all ungrounded conductors ofthe faulted circuit. This protection is provided at current levels less than those required to protect conductors fromdamage 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 theelectrical equipment and installations and has received safety training on the hazards involved.

Page 51: @Electrical

11

With these terms defined, several of the major components of the grounding system can be illustrated byredrawing the system of figure 6-2 and labeling the components:

Several key design constraints for grounding systems from the NEC [1] are as follows. These are paraphrasedfrom the code text (Note: This guide is not intended as a substitute for familiarity with the NEC, nor is it intendedas an authoritative interpretation of every aspect of the NEC articles mentioned.):

� Electrical systems that are grounded must be grounded in such a manner as to limit the voltage imposed bylightning, line surges, or unintentional contact with higher voltage lines and that will stabilize the voltage to earthduring normal operation [Article 250.4(A)(1)]. In other words, if a system is considered solidly grounded theground 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 isphysically 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 usedexclusively to supply industrial electric furnaces for melting, refining, tempering, and the like, separately derivedsystems used exclusively for rectifiers that supply only adjustable-speed industrial drives, and separatelyderived systems supplied by transformers that have a primary voltage rating less than 1000 volts provided thatcertain 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 thevoltage 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 combustiblefibers in Class III locations as provided in Article 503.155, circuits within hazardous (classified) anesthetizinglocations and other isolated power systems in health care facilities as provided in 517.61 and 517.160, circuitsfor equipment within electrolytic cell working zone as provided in Article 668, and secondary circuits of lightingsystems as provided in 411.5(A) [Article 250.22]. Some of the requirements for hazardous locations and healthcare facilities are covered in section XVI.

� For solidly-grounded systems, an unspliced main bonding jumper must be used to connect the equipmentgrounding conductor(s) and the service disconnect enclosure to the grounded conductor within the enclosurefor each utility service disconnect [Article 250.24(B)].

� For solidly-grounded systems, an unspliced system bonding jumper must be used to connect the equipmentgrounding conductor of a separately derived system to the grounded conductor. This connection must be madeat any single point on the separately derived system from the source to the first system disconnecting means orovercurrent device [250.30(A)(1)]

� A grounding connection on the load side of the main bonding or system bonding jumper on a solidly-groundedsystem is not permitted [Articles 240.24(A)(5), 250.30(A)]. The reasons for this are explained in below and insection VIII.

Figure 6-16: NEC [1] system grounding terms illustration

Page 52: @Electrical

12

� Ground fault protection of equipment must be provided for solidly grounded wye electrical services, feederdisconnects on solidly-grounded wye systems, and building or structure disconnects on solidly-grounded wyesystems under the following conditions:� The voltage is greater than 150 V to ground, but does not exceed 600 V phase-to-phase.

� 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].

� Where ground fault protection is required per Article 215.10 or 230.95 for a health care facility, an additional stepof ground fault protection is required in the next downstream device toward the load, with the exception ofcircuits on the load side of an essential electrical system transfer switch and between on-site generating units forthe essential electrical system and the essential electrical system transfer switches [Article 517.17]. Specificrequirements 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 faultprotection. For an emergency system, ground-fault indication is required [Articles 700.26, 701.17]. A latersection 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 thatcreates a permanent, low-impedance path facilitating the operation of the overcurrent device. This circuit mustbe able to safely carry the ground fault current imposed upon it. [Article 250.4(A)(5)]. The intent of thisrequirement is to allow ground fault current magnitudes to be sufficient for the ground fault protection/detectionto detect (and for ground fault protection to clear) the fault, and for a ground fault not to cause damage to thegrounding system.

� High-impedance grounded systems may utilized on AC systems of 480-1000 V where:� Conditions of maintenance and supervision ensure that only qualified persons access the installation.

� Continuity of power is required.

� Ground detectors are installed on the system.

� Line-to-neutral loads are not served.

[Article 250.36]

� For systems over 1000 V:� 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].

� 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 ispermissible 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 high-impedance 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 phaseconductors [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 groundingtransformer must not be provided with overcurrent protection independent of the main switch and common-tripovercurrent protection for the 3 phase, 4 wire system [Article 450.5 (A) (1)]. An overcurrent sensing device mustbe provided that will cause the main switch or common-trip overcurrent protection to open if the load on thegrounding 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.

Page 53: @Electrical

13

References[1] The National Electrical Code, NFPA 70, The National Fire Protection Association, Inc., 2005 Edition.

[2] IEEE Recommended Practice for Grounding of Industrial and Commercial Power Systems, IEEE Std. 142-1991, December 1991.

Page 54: @Electrical

1

Section 7: System ProtectionBill Brown, P.E., Square D Engineering Services

IntroductionAn important consideration in power system design is system protection. Without system protection, the powersystem 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 anddamage to equipment. Other concerns, such as transient overvoltages, are also considered when designingpower system protection although they are generally considered only after protection against abnormal currentshas been designed.

Characterization of power system faultsAny current in excess of the rated current of equipment or the ampacity of a conductor may be considered anovercurrent. Overcurrents can generally be categorized as overloads or faults. An overload is a condition whereload equipment draws more current that the system can safely supply. The main hazard with overload conditionsis the thermal heating effects of overloaded equipment and conductors. Faults are unintentional connections of thepower 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 connectionis referred to as a short circuit or bolted fault (the latter term is used due to the fact that a short circuit can bethought of as a bus bar inadvertently bolted across two phase conductors or from phase to ground). A fault toground is referred to as a ground fault. A fault between all three phases is referred to as a 3 phase fault. A faultbetween two phases is referred to as a phase-to-phase fault. A fault which contains enough impedance in theunintentional connection to significantly affect the fault current vs. a true short circuit is known as an impedancefault. An arcing fault has the unintentional connection made via an electrical arc through an ionized gas such asair. 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 beproduced at a given location in the system. In most cases this will be the three-phase short-circuit current, whichis the current produced if all three phases were shorted to each other and/or to ground. The simplest method forillustrating this is to reduce the power system at the point in question to its Thevenin equivalent. The Theveninequivalent is the equivalent single voltage source and impedance that produce the same short-circuit results asthe 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 withthe source voltage equal to zero. If a further simplification is made such that the system can be reduced to itssingle-phase equivalent, then a simple 3-phase fault current calculation for the three-phase fault current If3ø

can be performed as shown in figure 7-1:

Figure 7-1: Simplified 3-phase fault calculation

thZ

lnVV th = thf

Z

VI

ln=φ3

Page 55: @Electrical

2

The Thevenin impedance for a power system at a given point in the system is referred to as the short-circuitimpedance. 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 ungroundeddelta 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 theline-to-line voltage magnitude is equal to the line-to-neutral voltage magnitude multiplied by √3, and that there willbe 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 as:

(7-1)

This, however, is as far as this simplified analysis method will take us. In order to further characterize faultcurrents, 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 distinctcomponents: positive sequence, negative sequence, and zero sequence.These sequence components can bethought 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 zerosequence set all three phase components are in phase with one another. The positive, negative and zerosequence components can be further simplified by referring only to the A-phase phasor of each set; these arereferred to as V1 for the positive sequence set, V2 for the negative sequence set and V0 for the zero-sequenceset. For a given set of phase voltages Va, Vb and Vc, the sequence components are related to the phase voltagesas follows:

(7-2)

(7-3)

(7-4)

(7-5)

(7-6)

(7-7)

where

a = 1<120˚

The system may be separated into positive, negative, and zero-sequence networks depending upon the fault typeand 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 negative-sequence 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 systemmust 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

and a short-circuit impedance consisting of resistance R and inductance L, the fault

llfI −

2

3 3φf

llf

II =−

( )cba VaVaVV 21

31

++=

( )cba VaVaVV ++= 22

31

( )cba VVVV ++=31

0

021 VVVV a ++=

0212 VVaVaV b ++=

022

1 VVaVaV c ++=

)sin()( θω += tVtv m

Page 56: @Electrical

3

current for a fault initiated at time t=0 can be expressed as [2]:

(7-5)

where

ψ =

The angle ψ can be recognized to be the angle of the Thevenin impedance. Several key points can be taken from (7-5):

� When the fault occurs such that (θ - ψ )= 0 no transient will occur. For a purely inductive circuit this would meanthat θ = 90˚ and thus the fault is initiated when the voltage is at its peak.

� When the fault occurs such that (θ - ψ ) = 90˚ the maximum transient will occur. For a purely inductive circuitthis would mean that θ = 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 inductivereactance of the short-circuit impedance to its resistance. Thus the higher (X/R) or the “X/R ratio,” the longer theshort-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 smalltransient is produced is shown in figure 7-2. In contrast with this is the plot shown in figure 7-3, which is the faultcurrent for a system with a high X/R ratio and closing angle of 0 such that there is a large transient.

⎥⎥

⎢⎢

⎡−−−+

+=

⎟⎠⎞

⎜⎝⎛− tL

R

m etLR

Vti )sin()sin(

)()( ϕθϕθω

ω 22

⎟⎠

⎞⎜⎝

⎛−

R

Lω1tan

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

-1.5

-1

-0.5

0

0.5

1

1.5

0.0005 0.009 0.0175 0.026 0.0345 0.043

t(s)

i(t)

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

-1

-0.5

0

0.5

1

1.5

2

0.0005 0.009 0.0175 0.026 0.0345 0.043

t(s)

i(t)

Page 57: @Electrical

4

Figure 7-4 shows only the steady-state component of the waveform of figure 7-3, and figure 7-5 shows only thetransient component.

The fault current is often described in terms of its RMS Symmetrical and RMS Asymmetrical values. The RMSsymmetrical value is the RMS value considering the steady-state component only. The RMS asymmetrical value isthe RMS value over the first cycle considering both the steady-state and transient components at the worst-caseclosing angle. As a simplification of (7-5) an approximate asymmetry factor can be calculated as [3]

(7-9)

For example, this asymmetry factor for an X/R ratio of 25 is 1.6, meaning that the approximate worst-case RMSasymmetrical value over the first cycle for the fault current at an X/R ratio of 25 will be no greater than the RMSsymmetrical value multiplied by 1.6.

For motors and generators, which have a high X/R ratios, calculations for the transient performance during a faultare simplified by representing the short-circuit impedances differently for different time periods after the faultinitiation. The reactive component of the short-circuit impedance for the first half-cycle into the fault is thesubtransient reactance (X"d). For the first several cycles into the fault the reactance is larger and is termed thetransient reactance (X'd). For the long-term fault currents (up to 30 cycles or so into the fault) the reactance iseven larger and is termed the synchronous reactance (Xd). The synchronous reactance is much larger than eitherthe transient or subtransient reactance and models the phenomenon of AC decrement; after the DC componentdecays the AC component continues to decay, eventually reaching a value that can be less than the generatorrated 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 DCoffset. The AC decrement of the fault from generator sources is pronounced. Faults from most utility services aresufficiently far removed from generation and have enough resistance in the distribution lines that there is less DCoffset and essentially no AC decrement. The fault current contribution from induction motors has a high DC offsetbut also decays rapidly to zero over the first few cycles since there is no applied field excitation. The fault currentcontribution from synchronous motors has a large DC component and decays to zero but at a slower rate than for

Figure 7-4: Steady-state component of waveform in Figure 7-3

-1.5

-1

-0.5

0

0.5

1

1.5

0.0005 0.009 0.0175 0.026 0.0345 0.043

t(s)

ss i(

t)

Figure 7-5: Transient component of waveform in figure 7-3

00.10.20.30.40.50.60.70.80.9

1

0.0005 0.009 0.0175 0.026 0.0345 0.043

t(s)

tran

i(t)

⎟⎟⎟

⎜⎜⎜

⎛+=

RX

efactorAsymmetryπ221

Page 58: @Electrical

5

induction motors due to the applied field excitation. For a given point in the system, the fault current is the sum ofthe contributions from all of these sources and the DC offset, DC decay, and AC decrement are all dependentupon 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 subsequentsections of this guide), all the more speed and efficiency is gained by using the computer for short circuitcalculations; the various equipment rating standards can be taken into account to produce accurate results forcomparison with the equipment ratings.

Low voltage fusesThe simplest of all overcurrent protective devices is the fuse. A fuse is an overcurrent protective device with acircuit-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 withoutexceeding temperature rise limits. In accordance with NEC [1] article 210.20 [1] a fuse (or any branch-circuitovercurrent 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 itsmaximum interrupting rating, limits the clearing time at rated voltage to an interval equal to or less than the firstmajor or symmetrical loop duration, and limits peak let-through current to a value less than the peak current thatwould 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 singlecartridge. The dual-element design is a construction technique frequently used to obtain a desired time-delayresponse characteristic.

I2t: A measure of heat energy developed within a circuit during the fuse’s melting or arcing. The sum of meltingand 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 theconditions 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 thetotal clearing time.

Time delay: For Class H, K, J, and R fuses, a minimum opening time of 10s to an overload current five times theampere rating of the fuse, except for Class H, K, and R fuses rated 0-30 A, 250 V, in which case the opening timecan be reduced to 8s. For Class G, Class CC, and plug fuses, a minimum time delay of 12s on an overload oftwice the fuses’ ampere rating.

Total Clearing time: The total time between the beginning of the specified overcurrent and the final interruptionof the circuit, at rated voltage.

Voltage Rating: The RMS voltage at which the fuse is designed to operate. All low voltage fuses will operate atany 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 thevarious fuse classes and pertinent data for each class.

Fuses, like most protective devices, exhibit inverse time-current characteristics. A typical fuse time-currentcharacteristic is shown in figure 7-6. Logarithmic scales are used for both the time and current axes, in order tocover a wide range. The characteristic represents a band of operating times for which the lower boundary is theminimum melting time curve, above which the fuses can be damaged. The upper boundary is the total clearingtime curve, above which the fuse will open. For a given fault current, the actual fuse opening time will be withinthis band.

Page 59: @Electrical

6

Table 7-1: Low Voltage fuse classes [3]

FuseClass

VoltageRatings

AmpereRatings

InterruptingRatings(RMS)

CurrentLimiting?

Standards Notes

C 600 Vac

0-600 Vdc

0-12000 A

Varies

200,000 A

Varies

Varies UL 248-3-2000,CSA C22.2 NO.248.2-2000

Plug-style

CA 600 Vac

0-600 Vdc

0-30A

Varies

200,000A

Varies

Yes

Yes

UL 248-3-2000,CSA C22.2 NO.248-3-2000

No mounting holes

CB 600 Vac

0-600 Vdc

0-60 A

Varies

200,000 A

Varies

Yes

Yes

UL 248-3-2000,CSA C22.2 NO.248-3-2000

Mounting holes inend blades

CC 600 Vac

0-600 Vdc

0-30 A

Varies

200,000 A

Varies

Yes UL 248-4-2000 ,CSA C22.2 NO.248.4-2000

Rejection-style;

G 480 Vac

6000 V

480 Vdc

25-60 A

0-20 A

Varies

100,000 A

100,000 A

Varies

Yes

Yes

Yes

UL 248-5-2000,CSA C22.2 NO.248.5-2000

Non-inter-change-able dimen-sions withother fuse classes

H 250 Vac

600 Vac

0-600 Vdc

0-600 A

0-600 A

Varies

10,000 A

10,000 A

Varies

No

No

UL 248-6-2000,CSA C22.2 NO.248.6-2000

J 600 Vac

0-600 Vdc

0-600 A

Varies

200,000 A

Varies

Yes

Yes

UL 248-8-2000,CSA C22.2 NO.248.8-2000

K 250 Vac

250 Vac

250 Vac

600 Vac

600 Vac

600 Vac

0-600 Vdc

0-600A

0-600A

0-600A

0-600A

0-600A

0-600A

Varies

50,000 A

100,000 A

200,000 A

50,000 A

100,000 A

200,000 A

Varies

Yes*

Yes*

Yes*

Yes*

Yes

Yes*

UL 248-9-2000,CSA C22.2 NO.248-9-2000

Divided into low (K-1), medium (K-5),and high (K-9) Ip andI2t sub-classes;Dimen-sionsinterchange-able with class H fuses

L 600 Vac

0-600 Vdc

601-6000 A

Varies

200,000 A

Varies

Yes UL 248-10-2000,CSA C22.2 NO.248.10-2000

Bolt-on construction

R 250 Vac

600 Vac

0-600 Vdc

0-600 A

0-600 A

Varies

200,000 A

200,000 A

Varies

Yes

Yes

UL 248-12-2000,CSA C22.2 NO.248.12-2000

Divided into medium(RK-1) and high (RK-5) Ip and I2t sub-classes; Will fit classH or Class K fuseholders, but Class Rfuse holders will notfit any other type;

T 300 Vac

600 Vdc

0-600 Vdc

0-1200 A

0-1200 A

Varies

200,000 A

200,000 A

Varies

Yes

Yes

UL 248-15-2000,CSA C22.2 NO248.15-2000

Similar to Class J,but dimension-allysmaller

Plug FusesType C or

Type S

125 Vac

125 Vdc

0-30 A

0-30 A

10,000 A

10,000 A

No UL 248-11-2000,CSA NO. 248.11-2000

Type S has rejectionfeatures

* Because of their interchangeability with Class H fuses, class K-1, K-5, and K-9 fuses cannot be marked as “current-limiting.”

Page 60: @Electrical

7

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 inseries. The basic principle of system protection is that for a given fault current ideally only the device nearest thefault opens, minimizing the effect of the fault on the rest of the system. This principle is known as selectivecoordination 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 of30,000 A. Bus “A” is protected using 400 A class J fuses which supply, among others, bus “B.” Bus “B” is protectedusing 100 A class J fuses. Coordination between the 400 A and 100 A fuses can is shown via the time-currentcurves of figure 7-7, along with a one-line diagram of the part of the system under consideration. Because thetime 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 AFuse A has the potential to begin operating in its current-limiting region. Fuse B has the potential to beginoperating its current-limiting region at 1100 A. For currents above approximately 4000 A, therefore, both sets offuses have the potential to be operating in the current-limiting region. When both sets of fuses are operating thecurrent-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 I2tfor Fuse B. In practice, instead of publishing I2t data fuse manufacturers typically publish ratio tables showing theminimum ratios of fuses of a given type that will coordinate with each other.

Figure 7-6: Typical class J fuse time-current characteristic

100

100

1K1K

10K

10K

100K

100K

0 .01 0.01

0.10 0.10

1 1

10 10

100 100

1000 1000

C UR R E NT IN AMP E R E S

TIM

E IN

SE

CO

ND

S

Page 61: @Electrical

8

Low voltage fuse AC interrupting ratings are based upon a maximum power factor of .2, corresponding to amaximum X/R ratio of 4.899. In order to evaluate a low voltage fuse’s interrupting rating on a system with a higherX/R ratio the system symmetrical fault current must be multiplied by a multiplying factor [3]:

(7-10)

where

is the actual system X/R

is the test X/R

The available symmetrical fault current multiplied by the multiplying factor per (7-10) can be compared to the fuseinterrupting rating.

The use of fuses requires a holder and a switching device in addition to the fuses themselves. Because they aresingle-phase devices, a single blown fuse from a three-phase set will cause a single-phasing condition, which canlead 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 thesereasons, the use of low voltage fuses in modern power systems is generally discouraged. For circuit breakers thathave a short-time rating

Figure 7-7: Fuse coordination example

10

10

100

100

1K1K

10K

10K

100K

100K

0 .01 0.01

0.10 0.10

1 1

10 10

100 100

1000 1000

C UR R E NT IN AMP E R E S

TIM

E IN

SE

CO

ND

S

FUSE A

FUSE B

FUSE A

FUSE B

UTILITY SOURCE480 V30000.00A Available Fault

BUS A

FUSE A400.0 A

BUS B

FUSE B100.0 A

BUS C

testR

X

R

X

e

eMULT

actual

⎟⎠⎞

⎜⎝⎛−

⎟⎠⎞

⎜⎝⎛−

+

+=

π

π

1

1

actualR

X⎟⎠⎞

⎜⎝⎛

testR

X⎟⎠⎞

⎜⎝⎛

Page 62: @Electrical

9

Low voltage molded-case circuit breakersThe molded-case circuit breaker is the “workhorse” for system protection 600V and below. A circuit breaker is adevice designed to open and close by nonautomatic means and to open the circuit automatically on apredetermined 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-wiresystems. As stated in NEC article 240.85 [1], a circuit breaker with a straight voltage rating, such as 240 or 480 Vcan be used in a circuit in which the nominal voltage between any two conductors does not exceed the circuitbreaker’s voltage rating. Breakers with slash ratings, such as 120/240 V or 480 Y/277 V, can be applied in asolidly-grounded circuit where the nominal voltage of any conductor to ground does not exceed the lower of thetwo values of the circuit breaker’s voltage rating and the nominal voltage between any two conductors does notexceed the higher value of the circuit breaker’s 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 40˚C). 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 continuouscurrent 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. Circuitbreakers 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 circuitunless the circuit breaker is marked 1ø - 3ø 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 arecustom-ordered, such as motor operators.

Interrupting rating: This is the highest current at rated voltage that the circuit breaker is intended to interruptunder standard test conditions.

Short-time or Withstand Rating: This characterizes the circuit-breaker’s 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 tooperate 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, whenoperating in its current-limiting range, limits the let-through I2t to a value less than the I2t of a _-cycle wave of thesymmetrical prospective current.

HID: This is a marking that indicates that a circuit breaker has passed additional endurance and temperature risetests to assess its ability to be used as the regular switching device for high intensity discharge lighting. Per NEC240.80 (D) a circuit breaker which is used as a switch in an HID lighting circuit must be marked as HID. HIDcircuit 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 temperaturerise 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 expressedin amperes and corresponds to the largest ampere rating available in that group.

Page 63: @Electrical

10

Thermal-magnetic circuit breaker: This type of circuit breaker contains a thermal element to trip the circuitbreaker 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. Thesecircuit breakers are extremely flexible in coordination with other devices.

Sensor: An electronic-trip circuit breaker’s sensor is usually an air-core current transformer (CT) designedspecifically to work with that circuit breaker’s trip unit. The sensor size, in conjunction with the rating plug,determines the electronic-trip circuit breaker’s continuous current rating.

Rating plug: An electronic trip circuit breaker’s rating plug can vary the circuit breaker’s continuous current ratingas a function of it’s sensor size.

Typical molded-case circuit breakers are shown in figure 7-8. In figure 7-8 on the left is a thermal-magnetic circuitbreaker, and on the right is an electronic-trip circuit breaker. The thermal-magnetic circuit breaker is designed forcable connections and the electronic circuit breaker is designed for bus connections, but neither type is inherentlysuited for one connection type over another. Note the prominently-mounted operating handle on each circuitbreaker.

Circuit breakers may be mounted in stand-alone enclosures, in switchboards, or in panelboards (more informationon switchboards and panelboards is given in a later section of this guide).

A typical thermal-magnetic circuit breaker time-current characteristic is shown in figure 7-9. Note the two distinctparts of the characteristic curve: The thermal or long-time characteristic is used for overload protection and themagnetic or instantaneous characteristic is used for short-circuit protection. Note also that there is a band ofoperating times for a given fault current. The lower boundary represents the lowest possible trip time and theupper boundary represents the highest possible trip time for a given current.

Figure 7-8: Molded-Case circuit breakers

Page 64: @Electrical

11

The time-current characteristic for an electronic-trip circuit breaker is shown in figure 7-10. The characteristic foran electronic trip circuit breaker consists of the long time pickup, long-time delay, short-time pickup, short timedelay, and instantaneous pickup parameters, all of which are adjustable over a given range. This adjustabilitymakes the electronic-trip circuit breaker very flexible when coordinating with other devices. The adjustableparameters for an electronic trip circuit breaker are features of the trip unit. In many cases the trip unit is alsoavailable without the short-time function. In catalog data the long-time characteristic is listed as L, the short-time islisted as S, and the instantaneous as I. Therefore an LSI trip unit has long-time, short-time, and instantaneouscharacteristics, whereas an LI trip unit has only the long-time and instantaneous characteristics. For circuitbreakers that have a short-time rating, the instantaneous feature may be disabled, enhancing coordination withdownstream devices.

100

100

1K1K

10K

10K

100K

100K

0 .01 0.01

0.10 0.10

1 1

10 10

100 100

1000 1000

C UR R E NT IN AMP E R E S

TIM

E IN

SE

CO

ND

S

Thermal (Long-Time) Operating Region

Magnetic (Instantaneous) Operating Region

Thermal (Long-Time) Operating Region

Magnetic (Instantaneous) Operating Region

Figure 7-9: Thermal magnetic circuit breaker time-current characteristic

100

100

1K1K

10K

10K

100K

100K

0 .01 0.01

0.10 0.10

1 1

10 10

100 100

1000 1000

C UR R E NT IN AMP E R E S

TIM

E IN

SE

CO

ND

S

Long-Time Operating Region

Long Time Delay

Short Time Pickup

Short Time Delay

Instantaneous Pickup

Long-Time Operating Region

Long Time Delay

Short Time Pickup

Short Time Delay

Instantaneous Pickup

Figure 7-10: Electronic-trip circuit breaker time-current characteristic

Page 65: @Electrical

12

If the instantaneous feature has been disabled one must still be cognizant of any instantaneous override featurethe breaker has, which will engage the instantaneous function above a given level of current even if it has beendisabled 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 detaillater in this section.

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.

10

10

100

100

1K1K

10K

10K

100K

100K

0 .01 0.01

0.10 0.10

1 1

10 10

100 100

1000 1000

C UR R E NT IN AMP E R E S

TIM

E IN

SE

CO

ND

S

CB A

CB B

CB A

CB B

UTILITY SOURCE480 V30000.00A Available Fault

BUS A

BUS B

BUS C

CB A2500.0 A

CB B400.0 A

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 breakerscan be accomplished through zone-selective interlocking. This consists of wiring the two trip units such that if thedownstream circuit breaker senses the fault (typically this will be based upon the short-time pickup) it sends arestraining signal to the upstream circuit breaker. The upstream circuit breaker will then continue to time out asspecified on its characteristic curve, tripping if the downstream device does not clear the fault. However, if thedownstream device does not sense the fault and the upstream devices does, the upstream device will not havethe restraining signal from the downstream device and will trip with no intentional delay. For example, if zoneselective interlocking were present in the system of figure 7-11 and fault occurs on bus C circuit breaker B willsense the fault and send a restraining signal to circuit breaker A. Circuit breaker A is coordinated with circuitbreaker B, so circuit breaker B will trip first. If circuit breaker B fails to clear the fault, circuit breaker A will time outon its time-current characteristic per figure 7-11 and trip. If the fault occurs at bus B, circuit breaker B will notdetect the fault and thus will not send the restraining signal to circuit breaker A. Circuit breaker A will sense thefault and will trip with no intentional delay, which is faster than dictated by its time-current characteristic per figure7-11. Care must be used when applying zone-selective interlocking where there are multiple sources of power andfault 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; whenspecifying circuit breakers manufacturer’s actual catalog data should be used.

Page 66: @Electrical

13

Table 7-2: Typical characteristics of molded case circuit breakers for commercial andindustrial applications (Largely same as [3] table 7-1)

Note that the continuous current rating is set by the sensor and rating plug sizes for a given electronic-trip circuitbreaker. This can be smaller than the frame size. As can be seen from table 7-2, more than one interrupting ratingcan 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 theavailable RMS symmetrical fault current should be multiplied by the multiplying factor per equation (7-10) in orderto be compared with the circuit breaker interrupting rating.

Table 7-3: AC test circuit characteristics for molded-case circuit breakers [4]

Current-limiting circuit breakers are also available. Coordination between two current-limiting circuit breakers whenthey are both operating in the current limiting range is typically determined by test.

Frame Size (A) Number ofPoles

Interrupting Rating at AC voltage (kA, RMS symmetrical)

120 V 240 V 277 V 480 V 600 V

100 1

1

10

65

14

65

100, 150 2,3

2,3

2,3

18

65

100

14

25

65

14

18

25

225, 250 2,3

2,3

2,3

25

65

100

22

25

65

22

22

25

400, 600 2,3

2,3

2,3

42

65

100

30

65

22

25

35

800, 1000 3 42

65

200

30

50

100

22

25

65

1200 3

3

3

42

65

200

30

50

100

22

25

65

1600, 2000 3

3

65

125

50

100

42

65

3000, 4000 3

3

100

200

100

150

85

100

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

Page 67: @Electrical

14

By definition, low voltage molded case circuit breakers are not maintainable devices. Failure of a componentgenerally requires replacement of the entire circuit breaker unless the circuit breaker has been specificallydesigned for maintainability.

Magnetic-only circuit breakers which have only magnetic tripping capability are available. These are often usedas short-circuit protection for motor circuits (discussed in more detail in a later section of this guide). For thisreason 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 magneticelement which opens the switch above a specified current to protect the switch from damage due to lack of ashort-time rating.

Molded-case circuit breakers are available with several different options, such as stored-energy mechanisms, keyinterlocks, motor operators, etc. Refer to specific manufacturer’s literature for details.

Because the switching means is included with the device, molded-case circuit breakers give inherent flexibility ofoperation. This allows circuits to be reclosed without removing cover panels and exposing the operator tohazardous voltages. For three-phase circuits three-pole circuit breakers are used, which alleviates single-phasingconcerns. 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 breakersFor larger systems, those devices closest to the source of power often require the ability to coordinate withmultiple 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 ofapplication 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 olderdashpot-operated units may be encountered. The tripping characteristics are essentially identical to those forelectronic-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 withhigher interrupting ratings, although many modern-design power circuit breakers do not require fuses to obtainshort-circuit ratings up to 200kA RMS symmetrical.

The short-time rating is an important characteristic of the low voltage power circuit breaker. The rated timeduration 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 relayingapplications, as described below. Therefore, if a low voltage power circuit breaker is not equipped with a direct-acting 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

Page 68: @Electrical

15

Table 7-4: Preferred ratings for low voltage AC power circuit breakers with instantaneousdirect-acting phase trip elements (Same as [3] table 7-3)a

SystemNominalVoltage (V)

RatedMaximumVoltage (V)

Insulation(dielectric)withstand (V)

Three-Phase Short-Circuit currentrating (A)b

Frame Size (A) Range of tripdevice currentratings (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

a See IEEE Std C37.13-1990 and ANSI C37.16-2000

b Ratings in this column are RMS symmetrical values for single-phase (two pole) circuit breakers and three-phase average RMSsymmetrical values of three-phase (three-pole) circuit breakers. When applied on systems where rated maximum voltage mayappear across a single pole, the short-circuit current ratings are 87% of these values. See 5.6 in IEEE Std C37.13-1990.

c The continuous-current-carrying capability of some circuit-breaker-trip-device combinations may be higher than the trip-devicecurrent rating. See 10.1.3 in IEEE Std C37.13-1990.

Page 69: @Electrical

16

Table 7-5: Preferred ratings for low voltage AC power circuit breakers without instantaneousdirect-acting phase trip elements (Largely same as [3] table 7-4)a

As with molded-case circuit breakers, low voltage power circuit breakers are tested at a given power factor. Thetest power factor is 15% for unfused circuit breakers and 20% for fused circuit breakers. Table 7-6 shows themultiplying factors for both fused and unfused circuit breakers for various short-circuit power factors. Themultiplying 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 themultiplying factors obtained from equation (7-10) [5].

Rated MaximumVoltage (V) Frame Size (A)

Range of trip device current ratings (A)d

Setting of short-time delay trip element

Minimum timeband

Inter-mediateTime Band

MaximumTime 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

a See IEEE Std C37.13-1990 and ANSI C37.16-2000.

d 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.

Page 70: @Electrical

17

Table 7-6: Short-circuit multiplying factors for low voltage power circuit breakers (Largely same as [5] table 3)

Use of low voltage power circuit breakers allows optimum flexibility in coordination, since the instantaneousfunction 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 voltagemetal enclosed switchgear (described in a later section of this guide). This makes them ideal for low voltageautomatic 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 apremium consideration.

Medium voltage fusesThe definition of fuse in “Low voltage fuses” section above is equally applicable to medium voltage fuses. Recallfrom 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 kVthrough 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 thefocus of this section will be on medium voltage fuses through 38 kV.

The following standards apply to medium voltage fuses [3]:

� 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 arealso 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 thesame basic definition applies to medium voltage fuses. Expulsion fuses are defined as follows [3]:

System Short-CircuitPower Factor System X/R Ratio

Multiplying Factor xRMS Symmetrical Short-Circuit Current, forUnfused Power CircuitBreakers

Multiplying Factor xRMS Symmetrical Short-Circuit Current, forFused Power CircuitBreakers

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

7 14.25 1.11 1.21

5 20.0 1.14 1.26

Page 71: @Electrical

18

Expulsion fuse: A vented fuse in which the expulsion effect of the gases produced by internal arcing, eitheralone 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, appliedprimarily in stations and substations, with mechanical construction basically adapted to station and substationmountings.

Distribution fuse: Defined by ANSI C37.42-1996 as having dielectric withstand (BIL) strengths at distributionlevels, applied primarily on distribution feeders and circuits, and with operating voltage limits corresponding todistribution voltages. These are further subdivided into distribution current limiting fuses and distribution fusecutouts, as described below.

Current-limiting fuses interrupt in less than _ cycle when subjected to currents in their current-limiting range. Thisis an advantage as it limits the peak fault current to a value less than the prospective fault current as describedabove for low voltage fuses. This provides current-limiting fuses with high interrupting ratings and allows them toprotect downstream devices with lower short-circuit ratings in some cases. However, the same technologies thatcombine to give medium voltage current-liming fuses their current-limiting characteristics can also produce thermalissues when the fuses are loaded at lower current levels. For this reason, the following definitions apply tocurrent-limiting fuses [3]

Backup current-limiting fuse: A fuse capable of interrupting all currents from its maximum rated interruptingcurrent down to its rated minimum interrupting current.

General purpose current-limiting fuse: A fuse capable of interrupting all currents from the rated interruptingcurrent 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 currentdown 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 havemelting 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 withinthe range of 200% to 240% of the continuous-current rating of the fuse unit, refill unit, or use link. The current-responsive element for ratings above 100 A shall melt in 600 s at an RMS current within the range of 220% to264% 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% to240% 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 figure7-13 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 toexceed the E- or R-rating of the fuse. Failure to do this can result in the development of a hot-spot andsubsequent failure of the fuse and its mounting. For fuses enclosed in equipment, this can have disastrousconsequences since failure of the fuse and/or its mounting can lead to an arcing fault in the equipment. Note thatthe boundary of the characteristic, denoting the minimum-melting current, should be further derated to take intoaccount 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 current-limiting range below this interrupting time.

Page 72: @Electrical

19

A current-limiting power fuse consists of a fuse mounting (typically fuse clips) plus the fuse unit itself. These arefrequently mounted in metal-enclosed switchgear. A distribution current-limiting fuse may consist of adisconnecting-style holder or clips, plus the fuse unit. Distribution current-limiting fuses may also be provided withunder-oil mountings for use with distribution transformers. They are frequently used for capacitor protection aswell, with clips designed to mount to the capacitor.

Current-limiting power fuses are typically used for short-circuit protection of instrument transformers, powertransformers, and capacitor banks. Table 7-7 gives maximum ratings for medium voltage current-limiting powerfuses from 2.75 through 38 kV.

10

10

100

100

1K1K

10K

10K

100K

100K

0 .01 0.01

0.10 0.10

1 1

10 10

100 100

1000 1000

C UR R E NT IN AMP E R E S

TIM

E IN

SE

CO

ND

S

Figure 7-13: Typical E-rated current-limiting power fuse time-current characteristic.

Figure 7-14: Current-limiting power fuses and mountings

Page 73: @Electrical

20

Table 7-7: Maximum ratings for current-limiting power fuses 2.75 - 38 kV (Same as [3] table 6-3)

During interruption current-limiting fuses produce significant arc voltages. These must be taken into account inselecting equipment. They are typically compared to the BIL level of the equipment, including downstreamequipment at the same voltage level. The maximum permissible overvoltages for current-limiting power fuses areshown in table 7-8 [3]:

Table 7-8: Maximum permissible overvoltages for current-limiting power fuses (Same as [3] table 6-1)

In practice, the arc voltages for current-limiting fuses generally indicate the use of the smallest available fusevoltage 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.

Rated Maximum Voltage (kV) Continuous-Current Ratings (A),Maximum

Short-Circuit maximuminterrupting 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

a Parallel Fuses

Rated Maximum Voltage (kV)Maximum Peak Overvoltages (kV, crest)

0.5A to 12A Over 12A

2.8 13 9

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

Page 74: @Electrical

21

Power expulsion fuses generally consist of an insulating mounting plus a fuse holder which accepts the fuserefills. The fuse holder may be of the disconnecting or non-disconnecting type. Only the refill is replaced when afuse interrupts an overcurrent, and if only one phase of a three-phase set interrupted the fault only that fuse needbe 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 armmounting. 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 asthe 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, interruptingthe 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 aretherefore 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-acidexpulsion-type power fuses [3]. Because expulsion-type fuses are non-current-limiting, they do not produce thesignificant arc voltages that current-limiting fuses produce, and thus it is permissible to use a fuse with a largervoltage class than the system, for example, a 14.4 kV-rated fuse on a 4160 V system. This makes expulsion-typefuses particularly useful on systems which may be upgraded in the future to a higher voltage. However, the lowerinterrupting ratings of expulsion-type fuses are often an issue vs. current-limiting fuses in light of the fact that thelargest expulsion-type fuse interrupting ratings require larger physical dimensions which cannot always be easilyaccommodated in enclosed equipment. Further, in some cases the expulsion-type fuses prohibit some space-saving 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 typeboric-acid expulsion-type power fuses (Same as [3] table 6-6)

Rated Maximum Voltage (kV) Continuous-Current Ratings (A),maximum

Short-Circuit maximuminterrupting 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

a Parallel Fuses

Page 75: @Electrical

22

E-ratings are used for power expulsion fuses. A typical time-current characteristic for a 125E boric-acid fuse isgiven in figure 7-15.

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 isconcerned, 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 publishedinterrupting rating for power fuses is typically for a test X/R ratio of 15, and for distribution fuses the test X/R ratiois 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 fortransformers, cables, and capacitors. For more sophisticated protection at the medium voltage level, other meansmust be employed.

10

10

100

100

1K1K

10K

10K

100K

100K

0 .01 0.01

0.10 0.10

1 1

10 10

100 100

1000 1000

C UR R E NT IN AMP E R E S

TIM

E IN

SE

CO

ND

S

Figure 7-15: Typical boric acid power expulsion fuse time-current characteristic

Medium voltage circuit breakersThe 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 forboth 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.

Page 76: @Electrical

23

Medium voltage circuit breakers are generally not equipped with integral trip units as low voltage circuit breakersare. 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 ratingdefinitions 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.

Rated Dry Withstand Voltage: The RMS voltage that the circuit breaker in new condition is capable ofwithstanding for 1 minute under specified conditions.

Rated Wet Withstand Voltage: The RMS voltage that an outdoor circuit breaker or external components in newcondition are capable of withstanding for 10s.

Rated Lightning Impulse Withstand Voltage: The peak value of a standard 1.2 x 50µ s wave, as defined inIEEE 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 tocarry continuously.

Rated Interrupting Time: The maximum permissible interval between the energizing of the trip circuit at ratedcontrol 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 symmetricalcomponent of the short-circuit current in RMS amperes at the instant of arcing contact separation that the circuitbreaker shall be required to interrupt at a specified operating voltage, on the standard operating duty cycle, andwith 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 instantof arcing contact separation that the circuit breaker shall be required to interrupt at a specified operating voltageand on the standard operating duty cycle. This is based upon a standard time constant of 45ms (X/R ratio =17 for60 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 powerfrequency making current whose maximum peak is equal to or less than 2.6 (for 60 Hz power frequency; 2.5 for50 Hz power frequency) times the rated short-circuit current.

Figure 7-16: Medium voltage circuit breaker, for use In metal-clad switchgear

Page 77: @Electrical

24

Rated Short-Time Current: The maximum short-circuit current that the circuit breaker can carry without trippingfor a specified period of time.

Maximum Permissible Tripping Delay: The maximum delay time for protective relaying to trip the circuitbreaker during short-circuit conditions, based upon the rated short-time current and short-time current-carryingtime period.

Rated Transient Recovery Voltage (TRV): At its rated maximum voltage, a circuit breaker is capable ofinterrupting three-phase grounded and ungrounded terminal faults at the rated short-circuit current in any circuit inwhich the TRV does not exceed the rated TRV envelope. For a circuit breaker rated below 100kV, the rated TRVis represented by a 1-cosine wave, with a magnitude and time-to-peak dependent upon the rated maximumvoltage of the circuit breaker.

Rated Voltage Range Factor K: Defined in earlier versions of [6] as the factor by which the rated maximumvoltage may be divided to determine the minimum voltage for which the interrupting rating varies linearly with theinterrupting rating at the rated maximum voltage by the following formula:

(7-8)

where

Ivmax is the rated short-circuit current at the maximum operating voltage

Vmax is the rated maximum operating voltage

Vop is the operating voltage where Vop

Ivop 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. Becausemost modern circuit breakers employ vacuum technology, the current version of [6] assumes that K = 1., whichgives the same short circuit rating for all voltages below the rated voltage. However, in practice designs with K > 1still 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 preferredratings for circuit breakers where K > 1.

⎟⎟⎠

⎞⎜⎜⎝

⎛×=

opvvop V

VII maxmax

⎟⎟⎠⎞

⎜⎜⎝⎛≥K

Vmax

Page 78: @Electrical

25

Table 7-10: Preferred ratings for indoor circuit breakers with K=1.0 (Essentially same as [7] table 1)

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)

RatedMax.

Voltage,(kV)

RatedVoltageRange

Factor K

RatedContinuous

Current(A RMS)

RatedShort-

Circuit andShort-Time

Current(kA RMS)

Rated TRV RatedInter-

ruptingTime(ms)

Rated Max.Permissible

TrippingTime Delay Y

(s)

RatedClosing

andLatchingCurrent,

(kA Peak)

Rated PeakVoltage E2

(kV peak)

RatedTime toPeak T2,

(µs)

4.76 1.0 1200, 2000 31.5 8.9 50 83 2 82

4.76 1.0 1200, 2000 40 8.9 50 83 2 104

4.76 1.0 1200, 2000, 3000 50 8.9 50 83 2 130

8.25 1.0 1200, 2000, 3000 40 15.5 60 83 2 104

15 1.0 1200, 2000 20 28 75 83 2 52

15 1.0 1200, 2000 25 28 75 83 2 65

15 1.0 1200, 2000 31.5 28 75 83 2 82

15 1.0 1200, 2000, 3000 40 28 75 83 2 104

15 1.0 1200, 2000, 3000 50 28 75 83 2 130

15 1.0 1200, 2000, 3000 63 28 75 83 2 164

27 1.0 1200 16 51 105 83 2 42

27 1.0 1200,2000 25 51 105 83 2 65

38 1.0 1200 16 71 125 83 2 42

38 1.0 1200,2000x 25 71 125 83 2 65

38 1.0 1200, 2000, 3000 31.5 71 125 83 2 82

38 1.0 1200, 2000, 3000 40 71 125 83 2 104

RatedMax.

Voltage,(kV)

RatedVoltageRange

Factor K

RatedContinuousCurrent at

60Hz(A RMS)

Rated Short-Circuit

Current atRated Max.

kV(kA RMS)

RatedInterrupting

Time,Cycles

Rated Max.Voltage

Divided byK, kV RMS

Max.SymmetricalInterrupting

Capability andRated Short-Time Current

(kA, RMS)

Closing andLatching

Capability2.7K Times

Rated Short-CircuitCurrent

(kA Crest)

4.76 1.36 1200 8.8 5 3.5 12 32

4.76 1.24 1200, 2000 29 5 3.85 36 97

4.76 1.19 1200, 2000, 3000 41 5 4.0 49 132

8.25 1.25 1200, 2000 33 5 6.6 41 111

15.0 1.30 1200, 2000 18 5 11.5 23 62

15.0 1.30 1200, 2000 28 5 11.5 36 97

15.0 1.30 1200, 2000, 3000 37 5 11.5 48 130

38.0 1.65 1200, 2000, 3000 21 5 23.0 35 95

38.0 1.0 1200, 3000 40 5 38.0 40 108

In order to apply medium voltage circuit breakers, it is important to understand how the system X/R ratio affectsthe circuit breaker interrupting rating. As stated above, for 60Hz systems the asymmetrical interrupting capability isbased upon an X/R ratio of 17. Thus, for systems where the X/R ratio is 17 or lower the circuit breaker will haveadequate asymmetrical interrupting capability so long as 100% of the symmetrical short-circuit current rating is

Page 79: @Electrical

26

Figure 7-17: Microprocessor-based protective relay

equal to or above the available RMS symmetrical fault current. For X/R ratios above 17, the available RMSsymmetrical fault current must be compared to the short-circuit current rating of the circuit breaker multiplied by amultiplying factor determined from [8]. Because the multiplying factors from [8] do not usually exceed 1.25, thefault current may be compared to 80% of the circuit breaker interrupting rating regardless of X/R ratio in mostcases. The close and latch rating is evaluated using equation (7-9) to obtain the asymmetrical fault current at thecircuit breaker. Reference [8] contains a full method for determining the suitability of a circuit breaker for duty on agiven system, and along with the requirements for low voltage short-circuit calculations from [5] forms the basis forwhat the industry terms as ANSI short-circuit analysis. Capacitance switching and generator applications are alsoareas of concern when applying medium voltage circuit breakers. Preferred capacitance switching values aregiven in [7] and must not be exceeded. Generator applications, for generators rated above 3MVA, must beapproached with caution due to the high X/R ratios encountered. Often, breakers with longer interrupting times aredesirable in large generator applications in order to allow the fault current to decay to the point that there is anatural 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 areequipped with tripping and closing coils to allow tripping and closing operations via protective relays, manualcontrol switches, PLC’s, SCADA systems, etc. The circuit breaker internal control circuitry is arranged per IEEEC37.11-1997. Circuit breakers are also equipped with a number of auxiliary contacts to allow interlocking andexternal 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: Circuitbreaker applications are more expensive than fuse applications, both due to the inherent cost of the circuitbreakers themselves and due to the protective relays required. For many applications, however, circuit breakersare the only choice that offers the flexibility required. Large medium voltage services and distribution systems andmost applications involving medium voltage generation employ circuit breakers.

Protective relaysFor medium voltage circuit breaker applications, protective relays serve as the “brains” that detect abnormalsystem conditions and direct the circuit breakers to operate. They also serve to provide specialized protection inlow 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 olderelectromechanical devices are still available. Solid-state electronic or microprocessor-based relays offer moreflexibility and functionality than electromechanical relays, including the ability to interface with commoncommunications 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 ismost 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 insingle-phase or three-phase versions. Microprocessor-based relays are typically three-phase devices. Whileelectromechanical 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 microprocessor-based relay capable of performing the same functions that would require several electromechanical or solid-staterelays. This functionality usually makes microprocessor-based relays the best choice for new installations.

Page 80: @Electrical

27

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 aredesigned. Instrument transformers generally fall into one of two broad categories: Current Transformers (CT’s)and Voltage Transformers (VT’s). The loads on instrument transformers, such as relays and meters, are known asburdens to distinguish them from power system loads.

A current transformer consists of a coil toroidally-wound around a ferromagnetic core. The conductor for which thecurrent is to be measured is passed through the center of the toroid. The magnetic field generated by the currentthrough the conductor causes current to flow in the coil. In essence, a CT may be thought of as a conventionaltransformer with one primary turn.

CT’s 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. C57.13-1993designates certain ratios as standard, as well as a classification system for relaying performance. Theclassification system consists of a letter and a number. The letter may be C, designating that the percent ratiocorrection may be calculated, or T, denoting that the ratio correction has been determined by test. The numberdenotes the voltage that the CT can deliver to a “standard burden” (as described in IEEE Std. C37.13-1993) at 20times 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, theissue at hand is the performance of the relay during worst-case short-circuit conditions, when the CT secondarycurrents are the largest and may cause the secondary voltage to exceed the CT’s rating due to the voltagedeveloped across the relay input coil. This condition will cause the CT to saturate, significantly changing the ratioand thus the accuracy of the measurement. For cases of severe CT saturation the relay may respond in anunpredictable 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 aswindow-type CT’s. CT’s which are designed with an integral bus bar running through device are known as bus-bartype CT’s. Other designs, such as wound primary CT’s for metering applications and non-saturating air-core CT’s,are available. Additional information on CT application can be found in [3].

Voltage transformers (VT’s) 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 insection 2 of this guide, except that the design has been optimized for accuracy. Like current transformers, voltagetransformers 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 theapplication of voltage transformers.

Protective relays are classified by function. To make circuit representations easier, each function has been definedand 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.

pIsI

Quasi-Physical Arrangement

POWER CONDUCTOR

H1

H2

X1

X2

pI

sI

Circuit Representation

Figure 7-18: Current transformer

Page 81: @Electrical

28

Table 7-12: Commonly used protective relay device function numbers (Same as [3] table 4-1)

Table 7-13: Commonly used suffix letters applied to relay function numbers (Same as [3] table 4-2)

Several commonly-used protective functions are described below. It must be noted that where a protectivefunction is described it may be a dedicated relay (electromechanical, solid-state electronic, or microprocessor-based) or a single protective function contained within a microprocessor-based relay. In some manufacturer’sliterature 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 withvarious 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 withthe 50 instantaneous characteristic.

Relay DeviceFunction 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

Suffix Letter Relay Application

A Alarm only

B Bus protection

GGround fault protection [relay current transformer (CT) in a system neutral circuit] orgenerator protection]

GS Ground-fault protection (relay CT is toroidal or ground sensor)

L Line Protection

M Motor Protection

N Ground fault protection (relay coil connected in residual CT circuit)

T Transformer protection

V Voltage

Page 82: @Electrical

29

The pickup level is set by the tap setting, which is usually set in CT secondary amperes but may be set in primaryamperes 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-currentcharacteristic curve. In figure 7-19, the time dial settings are different to give enough space between the curves toshow their differences.

The above are IEEE-standard curves; others are available, depending upon the relay make and model. A solid-state 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 the50 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 timedelay is added the 50 is referred to as a definite-time overcurrent function. On solid-state electronic andmicroprocessor-based relays, the 50 function may be enabled or disabled. On electromechanical relays, the 50function can be added as an instantaneous attachment to a 51 time-overcurrent relay. If a relay has both 50 and51 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, themost common choice for ground fault protection is to add a fourth relay in the residual connection of the CT’s tomonitor the sum of all three phase currents. This relay is referred to as a residual ground overcurrent or 51N (or50/51N) relay.

The CT arrangement for 50/51 and 50/51N relays for a solidly-grounded system is shown in figure 7-20.

10

10

100

100

1K1K

10K

10K

100K

100K

0 .01 0.01

0.10 0.10

1 1

10 10

100 100

1000 1000

C UR R E NT IN AMP E R E S

TIM

E IN

SE

CO

ND

S

51, INVERSE

51, ST. INV.

51, MOD. INV.

51, VERY INV.

51, EXT. INV.

50

51, INVERSE

51, ST. INV.

51, MOD. INV.

51, VERY INV.

51, EXT. INV.

50

Figure 7-19: 50 and 51 overcurrent relay characteristics

RELAY CURRENT INPUTS

SOURCE

LOAD

50/51-A

50/51-B

50/51-C

50/51-N

A B C

Figure 7-20: Overcurrent relay arrangement with CT’s, including 50/51N

Page 83: @Electrical

30

For a low-resistance-grounded system, the use of an overcurrent relay connected to a CT in the servicetransformer or generator neutral is usually the best option. This CT should have a ratio smaller than the phaseCT’s, and the relay pickup range in conjunction with the neutral CT should allow a pickup as low as 10% of theneutral resistor rating. For a feeder circuit downstream from the service transformer, a zero-sequence CT isrecommended, 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 relaydepending upon relay type used. Figure 7-21 shows typical arrangements for both these applications.

For ungrounded systems, the ground detection methods in Section 6 are recommended since little ground currentwill 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.

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 instantaneousfunction). The bracketed [3] denotes that there are three phase overcurrent relays and three CT’s. The dotted line from the relays to the circuit breaker denotes that the relays are wired to trip the circuit breaker on anovercurrent condition.

Another type of overcurrent relay is the voltage-restrained overcurrent relay 51 V and the voltage-controlled relay51C. Both are used in generator applications to allow the relay to be set below the generator full-load current dueto the fact that the fault contribution from a generator will decay to a value less than the full-load current of thegenerator. The 51C relay does not operate on overcurrent unless the voltage is below a preset value. The 51 Vrelay 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.

RELAY CURRENT INPUT

TRANSFORMER NEUTRAL

51N

RELAY CURRENT INPUT

ZERO-SEQUENCE

A B C50G

C

A

BG

R

Figure 7-21: Transformer neutral and zero-sequence ground relaying applications for resistance-grounded systems

52

51CT600:5

[3]

[3]

51N

Figure 7-22: Typical application of overcurrent relays

Page 84: @Electrical

31

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 containa given fault.

Standard overcurrent relays cannot distinguish the direction of the current flow. Directional relays (67, 67N) arerequired to perform this function.

An important concept in the application of directional overcurrent relays is polarization. Polarization is the methodused by the relay to determine the direction of current flow. For phase directional overcurrent relays, this isaccomplished by the use of voltage transformers, which provide a voltage signal to the relay and allow it todistinguish 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 adifferent phase. Care must be used when defining CT polarities as each manufacturer typically defines a preferredpolarity 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-sequencevoltage. Electromechanical 67N relays must be polarized via either a CT in the source transformer neutral (zero-sequence current polarization) or three VT’s connected with a wye-connected primaries and broken-deltaconnected secondaries (refer to figure 6-9 in Section 6 for an example of the wye-broken delta connection withferroresonance-swamping resistor). Solid-state 67N relays usually must be polarized the same way but dosometimes offer a choice of either method. Microprocessor-based relays typically offer a choice of either methodand, in some cases, can self-polarize by calculating the zero-sequence voltage from the measured three-phaseline voltage.

As an example of the effectiveness of directional overcurrent relays, consider the primary-selective systemarrangement from Section 5 of this guide. The primary main and tie circuit breakers and an example of protectiverelaying for those circuit breakers are shown in figure 7-23.

In figure 7-23 the bus tie circuit breaker is normally-closed, paralleling the two utility feeds. Each main circuitbreaker and the bus tie circuit breaker are protected via 51 and 51N relays. The mains also have 67 and 67Nrelays. Note that the 67 relays are polarized via the line voltage transformers, and auxiliary voltage transformersconnected in wye-broken delta are supplied for polarization of the 67N relays. The polarization results in theindicated tripping directions for these relays. The need for the 67 and 67N relays can be demonstrated byconsidering a fault on one of the utility feeds. Should utility feed #2, for example, experience a fault, the faultcurrent will be supplied both from the upstream system feeding utility feed #2 and from utility feed #1 throughcircuit breakers 52-M1, 52-T, and 52-M2. Because the 51 and 51N relays for 52-M1 and 52-M2 are likely setidentically, they will both respond to the fault at the same time, tripping 52-M1 and 52-M2 and de-energizing theentire 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

N.C.

52-M252-M1

UTILITY FEED

#1

UTILITY FEED

#2

51CT600:5

[3]

[3]

51N

CT600:5

[3]

VT [3]

67

[3]

67N

AUX VT.[3]

67, 67NTRIP

DIRECTION

51 CT600:5

[3]

[3]

51N

CT600:5

[3]

VT [3]

67

[3]

67N

AUX VT.[3]

67, 67NTRIPDIRECTION

51

[3]

51N

52-T

FAULTFAULT CURRENT

FLOW

FAULT CURRENT FLOW

FAULT CURRENT FLOW

FAULT CURRENT FLOW

FAULT CURRENT FLOW

Figure 7-23: Example protective relaying arrangement for closed-transition primary-selective system

Page 85: @Electrical

32

will not trip due to the fact that the current is flowing in the direction opposite to the tripping direction. However, the67 and 67N relays on 52-M2 will sense current in the tripping direction and trip 52-M2. The downstream system isstill 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. Theyare 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. Undernormal conditions, the incoming power from the utility is measured and generator power output is controlled sothat power is not exported to the utility. Should the generator controls malfunction, the generator may begin tosource power to the utility. The 32 relay would then trip the service breaker offline, isolating the system from theutility. Of course, this would not apply if surplus power is to be intentionally sold to the utility in a co-generationarrangement.

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 ofmicroprocessor-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 inmultiple applications. The most common application is as a bus undervoltage relay, which alarms or trips a busoffline if the system voltage becomes unacceptably low. In this application, the relay should be time-delayed toride 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 momentarydips in voltage.

27 relays may also be used as permissive devices, for example to disallow closure of a circuit breaker if thesystem voltage is not above a pre-defined level. In this application the relay is typically configured to haveinstantaneous pickup, with time-delayed drop-out to insure that the system voltage has been above the presetlevel 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. Theyare 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 thelost phase high enough to avoid pickup by a 27 relay on that phase. This fact usually justifies the use of the 47relay 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 rolethe 47 relay helps insure that a circuit breaker does not close if the system phase rotation is reversed, such as bythe swapping of phase cables.

As with the 27 relay, the 47 relay should have a time delay to allow the system to ride through transientconditions. When a 27 and 47 relay are combined into the same electromechanical or solid-state electronicdevice, the device is referred to as a 27/47 relay.

Page 86: @Electrical

33

F.) Overvoltage relays (Device 59)59 relays respond to voltages above a pre-determined level. They are most often used in conjunction with 27relays in generator applications to protect voltage-sensitive devices from overvoltage. They may also be used aspermissive devices, usually in conjunction with 27 relays. Either application gives a voltage “window” within whichthe 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 systemthe 59 relay may be used across the broken-delta secondary of a ground-detection VT circuit, such as the circuitshown 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 a27/59 relay. When an electromechanical or solid-state electronic relay includes 27, 47, and 59 functions it isreferred 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 casesthe lockout relay is essentially a switch, and in fact is typically mounted in close proximity to circuit breaker controlswitches. The relay is spring-loaded, and a trip coil, when energized, causes the lockout relay to trip theconnected devices and prevent them from reclosing. There is typically a conspicuous target on the lockout relay toalert operating personnel that it has tripped. When the lockout relay is reset, the opening springs are compressedand the relay is ready for the next tripping operation.

86 relays are commonly used where one protective relay must trip several protective devices, and wherereclosure 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 currentflowing out, a fault must exist within the device.

Differential relays generally fall within one of two broad categories: Current-differential or high-impedancedifferential.

Current-differential relays are typically used to protect large transformers, generators, and motors. For thesedevices detection of low-level winding-to-ground faults is essential to avoid equipment damage. Current differentialrelays 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 issummed and the sum is directed through an operating winding. The current through the operating winding mustbe above a certain percentage (typically 15%-50%) of the current through the restraint windings for the relay tooperate. For solid-state electronic or microprocessor-based 87 relays the operating windings exist in logic onlyrather 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 CT’s are connected in delta tocounteract this phase shift for the connections to the relays. Under normal conditions the operating windings willcarry no current. For a large external fault on the load side of the transformer, differences in CT performance inthe primary vs. the secondary (it is impossible to match the primary and secondary CT’s due to different currentlevels) are taken into account by the proper percentage differential setting. Because the CT ratios in the primaryvs. secondary will not always be able to match the current magnitudes in the relay operating windings duringnormal conditions, the relays are equipped with taps to internally adjust the current levels for comparison. Thespecific 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 manysolid-state electronic and microprocessor-based relays, the phase shift is made internally in the relay and the CT’smay be connected the same on the primary and secondary sides of the transformer regardless of the transformerwinding connections. The manufacturer’s literature for a given relay make and model must be consulted whenplanning the CT connections.

Page 87: @Electrical

34

Percentage-differential characteristics are available as fixed-percentage or variable percentage. The difference isthat a fixed-percentage relay exhibits a constant percentage restraint, and for a variable-percentage relay thepercentage restraint increases as the restraint current increases. For an electromechanical relay, the percentagecharacteristic must be specified for each relay; for solid-state electronic or microprocessor-based relays thesecharacteristics 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. Theseharmonics are characteristic of transformer inrush and without harmonic restraint the transformer inrush maycause the relay to operate.

An important concept in the application of differential relays is that the relay typically trips fault interrupting deviceson both sides of the transformer. This is due to the fact that motors and generators on the secondary side of theprotected device will contribute to the fault current produced due to an internal fault in the device. An exampleone-line diagram representation of the transformer differential protection from 7-24 is given in figure 7-25:

Note that the secondary protective device is shown as a low voltage power circuit breaker. It is important that theprotective devices on both sides of the transformer be capable of fault-interrupting duty and suitable for relaytripping.

In figure 7-25 a lockout relay is used to trip both the primary and secondary overcurrent devices. The lockout relayis designated 86T since it is used for transformer tripping, and the differential relay is denoted 87T since it isprotecting the transformer. The wye and delta CT connections are also noted.

RELAY CURRENT INPUTS

SOURCE

LOAD

R

A B C

A B C

R

O

R R

O

R R

O

A-PHASE

B-PHASE

C-PHASE

Figure 7-24: Typical application of current-differential relays for delta-wye transformer protection

52

87T

CT[3]

[3]

86T

CT[3]

87T ZONE OF PROTECTION

Figure 7-25: Transformer differential relay application from figure 7-24 in one-line diagram format

Page 88: @Electrical

35

An important concept in protective relaying is the zone of protection; a zone of protection is the area that a givenprotective relay and/or overcurrent device(s) are to protect. While the zone of protection concept applies to anytype of protection (note the term zone selective interlocking as described earlier in this section), it is especiallyimportant in the application of differential relays because the zone of protection is strictly defined by the CTlocations. In figure 7-25 the zone of protection for the 87T relay is shown by the dashed-line box around thetransformer. For faults within the zone of protection, the currents in the CT’s will not sum to zero at the relayoperating windings and the relays will operate. Outside the zone of protection the operating winding currentsshould 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 foroperation. A high-impedance differential relay has a high-impedance operating element, across which the voltageis measured. CT’s are connected such that during normal load or external fault conditions the current through theimpedance is essentially zero. But, for a fault inside the differential zone of protection, the current through thehigh-impedance input is non-zero and causes a rapid rise in the voltage across the input, resulting in relayoperation. A simplified schematic of a high-impedance differential relay is shown in figure 7-26 to illustrate theconcept. Note that the relay only has one set of input terminals, without restraint windings. This means that anynumber of CT’s may be connected to the relay as needed to extend zone of protection, so long as the CT currentssum to zero during normal conditions. Also note that a voltage-limiting MOV connected across the high-impedanceinput is shown. This is to keep the voltage across the input during a fault from damaging the input.

High-impedance differential relays are typically used for bus protection. Bus protection is an application thatdemands many sets of CT’s be connected to the relays. It is also an application that demands that that relay beable to operate with unequal CT performance, since external fault magnitudes can be quite large. The high-impedance differential relay meets both requirements.

Figure 7-27 shows the application of bus differential relays to a primary-selective system. Note that in figure 7-27the zones of protection for Bus #1 and Bus #2 overlap. Here the 86 relay is extremely useful due to the largenumber of circuit breakers to be tripped. Note that all circuit breakers attached to the protected busses areequipped with differential CT’s and are tripped by that busses’ respective 86 relay. The 87 relays are denoted 87Bsince they are protecting busses. The same applies for the 86B relays. Note also that the protective zonesoverlap; 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 istypically set so that if one CT is fully saturated and the others are not the relay will not operate. By its nature, thehigh-impedance differential relay is less sensitive than the current-differential relay, but since it is typically appliedto protect bussing, where fault magnitudes are typically high, the additional sensitivity is not required.

Z

HIGH-IMPEDANCE DIFFERENTIAL RELAY

MOV

ZONE OF PROTECTION

Figure 7-26: High-impedance differential relay concept

Page 89: @Electrical

36

Ground-fault protection for solidly-grounded systems 600 V and belowBecause the ground fault is the most common type of system fault, and because low voltage systems arenecessarily the largest portion of most industrial and commercial facilities, low voltage ground-fault protection hasbecome a specialized area of development for system protection. Unlike the relayed ground-fault protectionsystems shown in VIII above, these systems are specially designed to provide sensitive protection for four-wiresystems with imbalanced loads.

As noted in Section 6, the National Electrical Code [1] requires ground-fault protection for most solidly-groundedelectrical systems 1000 A or more and above 150 volts to ground but not exceeding 600 V phase-to-phase. Forthis 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 guidelinesgiven above for relayed ground fault protection.

A.) Ground-fault protection for radial systemsGround-fault protection for low voltage radial systems is straightforward. For electronic trip units the tripping logicis typically built into the circuit breaker, and only the neutral CT or sensor must be connected to complete theground fault protection system. Such an arrangement is illustrated in figure 7-28:

I.) Other Protective Relay TypesOnly a small selection of the most commonly-used protective relay types are given here. For more in-depthdescriptions of their application, and for descriptions of other protective relay types, see reference [3]. Thoseprotective relay functions that are typically used for motors are described in section 8 of this guide.

N.C.

52-M252-M1

BUS #1

87BCT[3]

[3]

86B

CT[3] 52-T

BUS #2

CT[3]

CT[3]

CT[3]

CT[3]

87B CT[3]

[3]

86B

CT[3]

CT[3]

CT[3]

BUS #1 ZONE OF PROTECTION

BUS #2 ZONE OF PROTECTION

Figure 7-27: High-impedance differential relaying applied to a primary-selective system

FI

CB CURRENT SENSORS

SOURCE

LOAD

A B C N

ELECTRONIC-TRIP CIRCUIT BREAKER

TRIPUNIT W/

GFT(LSG,LSIG)

NEUTRAL SENSOR OR CT

SAFETY GROUND MAY OR MAY NOT

EXIST ON NEUTRAL

SENSOR/CT CIRCUIT

MAIN OR SYSTEMBONDINGJUMPER

GROUNDFAULT

FI

FI

FI

Figure 7-28: Low voltage ground fault protection for 4-wire radial system with electronic-trip circuit breaker

Page 90: @Electrical

37

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 systembonding 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 usedwith a zero-sequence sensor to trip the circuit breaker. The circuit breaker must be equipped with a shunt tripattachment in this case. Figure 7-29 shows an example of this arrangement. In figure 7-29 the external groundfault 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 includesthe phase conductors only.

These methods provide sensitive ground fault protection for solidly-grounded radial systems. However, if multiplesources are involved a more involved system is required in order to obtain reliable ground-fault protection.

B.) Modified-differential ground fault systemsBecause 4-pole circuit breakers are not in common use in the United States, the issue of multiple ground currentreturn paths has a large effect upon ground-fault protection in 4-wire systems. To illustrate this point, consider asecondary-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 orthe other source-transformer main/system bonding jumper neutral. How much ground fault current flows in eachpath 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 transformermain/system bonding jumper neutral and B x the total ground-fault current flows through the other transformermain/system bonding jumper, where A + B = 1. As can be seen from figure 7-30, the ground-fault protection forthe faulted bus can be de-sensitize or, worse, the wrong circuit breaker(s) may trip.

FI

SOURCE

LOAD

A B C N

Circuit Breaker without Electronic Trip

ZERO-SEQUENCE

SENSOR OR CT

SAFETY GROUND MAY OR MAY NOT

EXIST ON ZERO-SEQUENCE

SENSOR/CT CIRCUIT

BONDINGJUMPER

GROUNDFAULT

FI

FI

FI

ST

GS

Figure 7-29: Low voltage ground fault protection for 4-wire radial system without electronic trip circuit breaker

Page 91: @Electrical

38

The solution is the modified-differential ground fault system. A typical example of such a system is shown in figure 7-31:

In figure 7-31 the breaker internal sensors are shown, but the trip units are omitted for clarity. The ground-faultfunction for CB-M1 is noted as GM1, for CB-M2 is noted as GM2, and for CB-T is noted as GT. In thisarrangement, regardless of the ground current dividing factors A and B the correct circuit breakers will sense theground 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 CT’s 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 beexpanded. These are usually custom-engineered solutions.

C.) Four-pole circuit breakersAnother possible option, in lieu of the modified differential ground fault system, is the use of four-pole circuitbreakers. These switch the neutral as well as the phase conductors, separating the neutrals of multi-sourcecircuits. This method will not work if sources are paralleled. Four-pole circuit breakers are not common in theUnited States for this reason, as well the increased physical equipment sizes they necessitate.

SOURCE #1

SOURCE #1

LOADS

A B C N

MAIN OR SYSTEMBONDINGJUMPER

GROUNDFAULT

SOURCE #1

A B C N

SOURCE #2

LOADS

CB-M1 CB-M2

CB-T

IF

IF

A x IF

IF

B x IF

B x IF

B x IF

B x IF

Figure 7-30: Secondary-selective system with radial ground-fault protection of figure 7-28 applied

SOURCE #1

SOURCE #1

LOADS

A B C N

MAIN OR SYSTEMBONDINGJUMPER

GROUNDFAULT

SOURCE #1

A B C N

SOURCE #2

LOADS

CB-M1 CB-M2

CB-T

IF

IF

A x IF

IF

B x IF

B x IF

B x IF

GT

GM1 GM2

B x IF

B x IFA x IFIF B x IF

0IF

B x IF

A x IF

B x IF

Figure 7-31: Modified-differential ground-fault protection for secondary-selective system

Page 92: @Electrical

39

D.) Ground fault time-current characteristicsFigure 7-32 shows typical time-current characteristics for the ground fault function of an electronic-trip circuit breaker.

This characteristic is adjustable both for pickup and time delay. Discrete relays for use with non-electronic circuitbreakers are also available with similar characteristics.

Care must be taken when coordinating ground-fault protection if multiple levels of ground-fault protection do not existdownstream from the service or source of a separately-derived system. The NEC Article 230.95 (A) service-entrancerequirement [1] for a maximum of 1200 A pickup and maximum 1-second delay at 3000 A ground-fault current canlead to a lack of coordination for downstream feeder and branch-circuit ground faults. This is one of the reasons forthe use of other than solidly-grounded systems where maximum system reliability is to be achieved.

0.5 1

1

10

10

100

100

1K1K

10K

10K

0 .01 0.01

0.10 0.10

1 1

10 10

100 100

1000 1000

C UR R E NT IN AMP E R E S

TIM

E IN

SE

CO

ND

S

Figure 7-32: Typical electronic-trip circuit breaker ground-fault protection time-current characteristic

Surge protectionSurge protection is protection of conductors and equipment against the effects of voltage surges. These areusually due to lightning, although switching transients can also cause damaging overvoltages. Unlike overvoltagerelaying, surge protection is directly connected to the power circuit, and for the best protection is usually locatedas close as physically practical to the protected equipment.

A.) Medium voltage surge protectionMedium voltage surge protection is generally accomplished with surge arresters. A surge arrester exhibits animpedance which decreases with the line voltage. Older technologies included spark-gap arresters, whichprovided 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 resistancewhich 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 areused, they must be sized for the maximum anticipated phase-to-ground voltage. MOV surge arrestors have aduty-cycle voltage rating and a maximum continuous operating voltage (MCOV) rating; the MCOV rating is thequantity to be compared to the phase-to-ground voltage. Also of importance is the arrester’s classification asdistribution-class, intermediate-class, or station-class; these classes are defined in terms of the energy the

Page 93: @Electrical

40

surge arrestor can absorb without damage, in ascending order as listed. Table 7-14 gives commonly-applied MOVsurge arrestor ratings vs. the system voltage. In general, use of surge arrestors with the lowest MCOV exceedingthe anticipated line-to-ground voltage provides the best protection. Detailed insulation coordination studies canalso be performed with the use of transient analysis software. For low-resistance-grounded systems, selection ofthe lowest acceptable surge arrestor rating involves comparing the overvoltage vs. time characteristic of the surgearrestor 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 motorwindings. Care must be used when sizing surge capacitors and the effects of harmonic currents must beevaluated 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 polymerhousings will result in minimal damage should the arrester fail; the housing will simply split to relieve the internaloverpressure. Use of porcelain housings which can sustain large internal overpressures can result in severedamage should the arrester fail. In the case of surge capacitors, since they are typically filled with dielectric fluidand have steel housings they can sustain high internal overpressures, and failure of the housing due to internaloverpressure 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)` MCOV (kV)4-Wire Effectively-Grounded NeutralSystem1

3-Wire Grounded andResistance-GroundedWye Systems2

3 2.6 4160 Y/2400 2400

6 5.1 8320 Y/4800 4160

4800

9 7.7 12000 Y/6930

12470 Y/7200

6900

10 8.4 13200 Y/7620

13800Y/7970

12 10.2

15 12.7 20780 Y/12000 12000

12470

18 15.3 22860 Y/13200

24940 Y/14400

13200

13800

21 17.0

24 19.5

27 22.0 34500 Y/19920 20780

30 24.4 22860

36 29.0 24940

1 Use of this system category requires a solid ground conductor (non-earth) path back to the upstream transformer or generator neutral.

2 Includes grounded-wye systems where the path to the upstream transformer neutral includes an earth path

Page 94: @Electrical

41

B.) Low voltage surge protectionStrictly speaking, low voltage surge protection is via smaller versions of the MOV arresters used in mediumvoltage systems. Various versions of these are available, including mountings which fit the circuit-breaker spacesin 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 deviceclassification unique to UL and is intended to provide a degree of protection for sensitive utilization equipmentagainst the effects of transient voltages. TVSS’s usually include MOV’s for voltage clamping as well as filtering forsurge attenuation. They are designed and manufactured to UL Std.1449, which requires consideration for devicefailure modes due to the intended installation location. UL 1449 defines three general installation classifications:Permanently Connected, Cord Connected, and Direct Plug-In. TVSS’s are intended for use on the load side of anovercurrent protective device such as a circuit breaker. They often include status indication for identification of afailed unit. Transient Voltage Surge Suppressors are generally considered the preferred means of low voltagesurge protection.

Protection of specific system components

A.) Power cablesFor low voltage power cables, so long as they are protected at their ampacities per NEC Article 240.5 [1] they canbe considered adequately protected.

For medium voltage power cables, proper protection requires both sizing per the NEC ampacity tables andcomparing the overcurrent protective device time-current characteristic with the cable damage characteristic perIPCEA Publication P-32-382. The cable damage characteristic for a #1/0 AWG copper conductor is shown infigure 7-33 as compared with the time-current characteristic of a 300E fuse. Per NEC Article 240.100 (C) [1], thefactor to be considered is the short-circuit performance of the cable; indeed, NEC Article 240.101 (A) allows therating of a fuse protecting the conductor to be up to 3 times the ampacity of the conductors (6 times for a circuitbreaker or electronically-actuated fuse). As can be seen in figure 7-33, the 300E fuse indeed does not provideoverload protection for the cable at its ampacity (200 A), but is within the limits of NEC 240.101 (A). Even thoughthe fuse will allow the cable ampacity to be exceeded, the continuous load on the cable should not exceed thepublished conductor ampacity. The 300E fuse does, however, adequately protect the cable for short-circuits, asevidenced 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 protectionTransformer 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 thetransformer. The smallest protective device that allows the rated full-load current of the transformer gives the bestpractical overcurrent protection. Increasing the secondary overcurrent device size beyond this may be necessaryfor short-term overloads or for coordination with downstream devices, but in any case the requirements of NECArticle 450 must be met.

Page 95: @Electrical

42

Short-circuit protection involves comparison of the transformer damage curve per IEEE Std. C57.109-1993 withthe primary overcurrent device time-current characteristic. In general, the damage curve must be to the right andabove the primary overcurrent device characteristic. Another constraint on the primary overcurrent device is that itmust be capable of withstanding the inrush of the transformer without tripping (and without damage for current-limiting 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 65Ecurrent-limiting primary fuses and a 1200 A electronic-trip secondary circuit breaker. As can be seen from thefigure, the fuses do withstand the inrush without damage since the inrush point is to the left and below the fuseminimum melt curve. The transformer is protected from short-circuits by the primary fuses. The secondary circuitbreaker provides overload protection at the full-load current of the transformer. Note that the primary fuse andsecondary circuit-breaker characteristics overlap for high fault currents; this is unavoidable and is consideredacceptable. Note also that the fuse curve and the transformer damage curve overlap; this is unavoidable but theseshould overlap at the lowest current possible. For currents below the fuse/transformer damage curve overlap thesecondary circuit breaker must protect the transformer; the lower the point of overlap, the more likely the fault isan external fault on the load side of the secondary circuit breaker and therefore greater chance the secondarycircuit 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-wyetransformer, a ground-fault on the secondary side of the transformer will result in only 57.7% of the maximumthree-phase primary fault current while one secondary winding experiences the full fault current. This is illustratedin Figure 7-35, as well as the corollary for delta-delta transformers. The damage characteristic has therefore beenshifted to 57.7% of its published value to account for secondary line-to-ground faults. Also, the shifted curve hasanother, more conservative curve shown; this is the frequent-fault curve and is applicable only to the secondaryovercurrent device since faults between the transformer secondary and the secondary overcurrent protectivedevice should not be frequent.

Additional devices, such as thermal overload alarms/relays and sudden-pressure relays, are also available forprotection of transformers. These are typically specified with the transformer itself and can provide very goodprotection. However, even if these devices are installed the primary and secondary overcurrent devices must becoordinated with the transformer as described above.

#1/0 AWG CABLE

10

10

100

100

1K1K

10K

10K

100K

100K

0 .01 0.01

0.10 0.10

1 1

10 10

100 100

1000 1000

C UR R E NT IN AMP E R E S

TIM

E IN

SE

CO

ND

S

#1/0 AWG CABLE

300E C/L FUSE

#1/0 AWG CABLE

300E C/L FUSE

Figure 7-33: Medium voltage cable protection example

Page 96: @Electrical

43

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 internalfaults in this case.

C.) Generator protectionThe 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 thatare 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).

TX Inrush

XFMR

0.5 1

1

10

10

100

100

1K1K

10K

10K

0 .01 0.01

0.10 0.10

1 1

10 10

100 100

1000 1000

C UR R E NT IN AMP E R E S

TIM

E IN

SE

CO

ND

S

PRI. FUSE

XFMR

SEC. CB

PRI. FUSE

XFMR

SEC. CB

Figure 7-34: Example protection for a 1000 kVA, 13.2 kV Delta: 480 Y/277V, 5.75%Z dry-type transformer

Figure 7-35: Fault-current flow for delta-wye transformer L-N faults and delta-delta transformer L-L faults

Page 97: @Electrical

44

E.) Other devices and additional information

For protection other devices, refer to [3] and/or the applicable standards for the device in question. For additionalinformation on the protection of cables and transformers, please refer to [3].

Protection selectivityThe selectivity of protection refers to its ability to isolate an abnormal condition to the smallest portion of thesystem possible. In most cases selectivity is a function of how well-coordinated the overcurrent protective devicesin the system are. As an example, consider the system of figure 7-36:

Figure 7-36 shows a small radial system with a medium voltage utility service, a service substation consisting of aprimary switch step-down transformer protected by a primary fuse, and a secondary switchboard. One of theswitchboard 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. Thepanel main circuit breaker and all devices upstream should not be affected. If the lighting panel feeder circuitbreaker time-current characteristic does not coordinate with that of the lighting panel main, the main may trip, de-energizing the entire panelboard.

Going upstream, a fault at point F should only cause the panelboard main circuit breaker to trip and a fault at pointE should only cause the switchboard main circuit breaker to trip. A fault at point D may cause the switchboardmain circuit breaker to trip or the primary fuse to blow, but the effect on the system is the same since all of theloads 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 theoutage increases as the fault location is considered farther and farther upstream. In this example, if thetransformer primary fuses and the upstream utility recloser, protective relays, or fuses are not coordinated theentire 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 thetime-current coordination characteristics of the protective devices in the system and plots them on time currentcurves such as those illustrated in this section. Coordination is considered to be achieved between two devices iftheir time-current bands show sufficient clear space between them on the time-current curve or, in the case ofprotective relays, if sufficient margin for overtravel, manufacturing tolerances, circuit breaker speed, and safety areachieved.

UTILITY SERVICE

A

B

C

D

E

F

G

Figure 7-36: Example system for selectivity discussion

Page 98: @Electrical

45

Coordination is not always possible to maintain in the high fault-current ranges. However, in most cases anacceptable 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 panelfeeder circuit breaker fails to trip the panelboard main circuit breaker should trip as dictated by its time-currentcurve. If selective coordination exists between the panelboard main circuit breaker and the switchboard feedercircuit breaker, then the switchboard feeder circuit breaker will not trip. So, backup protection must consider onelevel upstream vs. primary protection unless additional backup protective devices are installed.

References[1] The National Electrical Code, NFPA 70, The National Fire Protection Association, Inc., 2005 Edition.

[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. 242-2001, December 2001.

[4] Molded-Case Circuit Breakers, Molded Case Switches and Circuit-Breaker Enclosures, UL 489, Underwriter’sLaboratories Inc., April 25, 2002.

[5] IEEE Standard for Low Voltage AC Power Circuit Breakers Used in Enclosures, ANSI/IEEE Standard C37.13-1990, October 1990, Reaff. April 8, 1996.

[6] IEEE Standard Rating Structure for AC High Voltage Circuit Breakers, ANSI/IEEE Standard C37.04-1999, June1999.

[7] AC High Voltage Circuit Breakers Rated on a Symmetrical Current Basis – Preferred Ratings and RelatedRequired 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 StdC37.010-1999, September 1999.

[9] Swindler, D.L., Fredericks, C.J., “Modified Differential Ground Fault Protection for Systems Having MultipleSources and Grounds,” Industry Applications, IEEE Transactions on, Volume 30, Issue 6, Nov. 1994.

Page 99: @Electrical

1

Section 8:AC Motors,motor control and motor protectionBill Brown, P.E., Square D Engineering Services

IntroductionElectric motors are an important part of any electrical system. Because they convert electrical energy tomechanical 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 commonlyused for these.

AC motor typesMotors 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 , whichproduce a magnetic field that interacts with the magnetic field from the armature windings to produce torque onthe rotor. This torque causes the rotor to rotate. For most AC motors, the armature windings are located on thestator, and the field windings are located on the rotor (one exception is the field exciter for a brushlesssynchronous motor, as described below). For this reason, in most cases the armature windings are knownsynonymously as the stator windings.

AC motors in common use today may be divided into two broad categories: Induction (asynchronous) orsynchronous. 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 therotating stator magnetic field. For synchronous motors, a field excitation is applied to the rotor windings. Thisdifference in field excitation leads to differences in motor characteristics, which leads in turn to different protectionand control requirements for each motor type.

A.) Induction motorsInduction motors are the “workhorses” of modern industry. Because they have no applied field excitation, the rotorwindings 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 oneach end, forming a structure which resembles a squirrel cage [1]. Due to the simple rotor construction, thesquirrel 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 laterin this section.

An important concept in the application of induction motors is the fact that due to the lack of field excitation, themotor speed will vary with the torque of the load. Synchronous speed for a given motor is given by the equation:

(8-1)

where

ns is the synchronous speed, in RPM

f is the frequency in Hz

p is the number of poles of the motor, which can be defined as 2 x the number of different magneticfield 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 themotor. The motor speed can be expressed as:

Page 100: @Electrical

2

(8-2)

where

n is the speed of the motor, in RPM

s is the slip

ns 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 performancecharacteristics of the motor. Table 8-1 gives typical design letter characteristics for induction motors. These aretypical characteristics only – for further details consult the specific performance standards for the completerequirements [2,3].

B.) Synchronous motorsSynchronous 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 synchronousspeed, so for starting the rotor is also equipped with damper windings that allow the motor to be started as aninduction motor.

Table 8-1: Typical characteristics and applications of fixed frequency medium AC squirrelcage motors (Essentially same as [2] table 10-1 and [3] table 3)

PolyphaseCharacteristics

Locked-rotor

torque(percent

rated loadtorque)

Pull-uptorque

(percentrated load

torque)

Breakdowntorque

(percentrated load

torque)

Locked-rotor

current(percent

rated loadcurrent)

Slip(%)

Typical Applications RelativeEfficiency

Design ANormal lockedrotor torque andhigh locked-rotorcurrent

70-275a 65-190a 175-300 NotDefined

0.5-5 Fans, blowers, centrifugalpumps and compressors,motor-generator sets, etc.,where starting torquerequirements are relatively low

Mediumor High

Design BNormal locked-rotor torque andnormal locked-rotor current

70-275a 65-190a 175-300a 600-800 0.5-5 Fans, blowers, centrifugalpumps and compressors,motor-generator sets, etc.,where starting torquerequirements are relatively low

Mediumor High

Design CHigh locked-rotortorque andnormal locked-rotor current

200-285a 140-195a 190-225a 600-800 1-5 Conveyors, crushers, stirringmachines, agitators,reciprocating pumps andcompressors, etc., wherestarting under load is required

Medium

Design DHigh locked-rotortorque and high slip

275 Notdefined

275 600-800 ≥5 High peak loads with or withoutflywheels such as punchpresses, shears, elevators,extractors, winches, hoists, oil-well pumping and wire-drawingmachines

Medium

IEC Design HHigh locked rotortorque and highlocked rotorcurrent

200-285a 140-195a 190-225a 800-1000 1-5 Conveyors, crushers, stirringmachines, agitators,reciprocating pumps andcompressors, etc., wherestarting under load is required

Medium

IEC Design NNormal locked-rotor torque andhigh locked rotorcurrent

70-190a 60-140a 160-200a 800-1000 0.5-3 Fans, blowers, centrifugalpumps and compressors,motor-generator sets, etc.,where starting torquerequirements are relatively low

Mediumor High

a Higher values are for motors having lower horsepower ratings

Page 101: @Electrical

3

Synchronous motors may be further classified as brush or brushless type. The field exciter for a brush-type motoris typically a DC generator with its rotor mounted on the motor shaft. The output of the DC generator is fed viabrushes and slip rings to the motor field windings. The field exciter for a brushless synchronous motor typicallyconsists of an AC generator with the field windings on its stator, armature windings on its rotor, and with its rotormounted on the motor shaft. The output of the generator is rectified by solid-state rectifier elements also mountedon the rotor shaft and fed directly to the motor field windings without the need for brushes or slip rings. Because ofthe proliferation of solid-state power electronic technology, and because the brushless-type motors require lessmaintenance almost all new synchronous motors are brushless-type [1], although many existing installations dohave older brush-type motors in service. In either design the field excitation to the exciter may be varied to varythe power-factor operation of the motor, and in fact power factor correction is one common use of synchronousmotors since they can be made to operate at leading power factors.

C.) Enclosure types, cooling methods and other general application informationPlease refer to [3] for more information on motor enclosure types and cooling methods, as well as additionalgeneral application information for motors.

Motor torque and driven load characteristicsMotors are rated in horsepower (hp; 1hp = 746W) or, occasionally, in watts or kilowatts. In either case, this is therated output power of the motor at the motor shaft when the motor is running at full speed. Due to losses in themotor, the input power will be higher. Due to the motor power factor and these losses, the full-load current of themotor will be larger than would be otherwise anticipated by looking only at the hp or kW rating. This will bediscussed 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 is the shaft output power in hp

T is the shaft output torque in ft-lbf

n 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 is the output accelerating torque in ft-lbf

J is the total moment of inertia of the motor shaft and rotor plus the driven load, in lb-ft2 (also referred toas wk2)_ is the shaft acceleration in rpm/min.

Because a= dn/dt, the speed of the motor shaft can be written as:

(8-5)

Page 102: @Electrical

4

The inertia of the load (and rotor), then, is crucial to the acceleration rate of the motor shaft (and the load) andthus to the output speed of the shaft. A typical design B induction motor torque-speed characteristic is as shown infigure 8-1, along with pertinent characteristics from table 8-1 labeled:

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 isplotted along with the motor speed-torque characteristic in figure 8-2. The load speed-torque characteristic is aplot of the torque required to drive a load at a given speed. Several points can be made regarding the motor andload of figure 8-2:

� The motor locked-rotor torque is greater than the load torque at zero speed. This means the motor can start withthe load connected.

� The motor pull-up torque is greater than the load torque during the acceleration period. This means that themotor can successfully accelerate the load.

� The steady-state speed of the motor is where the motor-torque and load-torque curves cross – the steady-stateoperating 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-loadsystem. The accelerating torque is the same as given in eq. (8-1) above. A plot of the accelerating torque is givenin figure 8-3.

Figure 8-1: Typical NEMA design B induction motor speed-torque characteristic

0

0.2

0.4

0.6

0.8

1

1.2

00.0

60.1

20.1

80.2

4 0.3 0.36

0.42

0.48

0.54 0.6 0.6

60.7

20.7

80.8

4 0.9 0.96

Speed (pu synchronous)

Torq

ue (p

u br

eakd

own)

Locked-RotorTorque

Pull-UpTorque

BreakdownTorque

Figure 8-2: Example motor and load speed-torque characteristics

0

0.2

0.4

0.6

0.8

1

1.2

0

0.07

0.14

0.21

0.28

0.35

0.42

0.49

0.56

0.63 0.

7

0.77

0.84

0.91

0.98

Speed (pu synchronous)

Torq

ue (p

u br

eakd

own)

MotorTorque

LoadTorque

Accelerating Torque

Steady-State OperatingPoint

Figure 8-3: Accelerating torque for motor and load of figure 8-2

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

00.0

60.1

20.1

80.2

4 0.3 0.36

0.42

0.48

0.54 0.6 0.6

60.7

20.7

80.8

4 0.9 0.96

Speed (pu synchronous)

Acc

eler

atin

g To

rque

(pu

mot

or

brea

kdow

n to

rque

)

Page 103: @Electrical

5

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 inapproximately 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.

The importance of the above analysis lies in the fact that for successful motor starting the motor must be able tosuccessfully 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.

For a synchronous motor, starting analysis is similar since the damper windings of the motor give a speed-torquecharacteristic similar to that of an induction motor. For a synchronous motor the steady-state speed is thesynchronous speed of the motor, which is achieved by applying the field excitation once the motor hasaccelerated to a speed close to synchronous speed on the damper windings.

Motor starting methodsSeveral 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 devicesThe 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.” Contactorsare designed for optimum performance and lifetime when switching loads; they are not designed for interruptingshort-circuit currents and therefore motor circuits require separate short-circuit protection. Because contactors areclosed magnetically via their control coils, the use of contactors is typically referred to as magnetic control. Forsmall motors, typically fractional-horsepower, manual control switches are also available. Motor starting contactorsand 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-determinedmanner, the electric power delivered to the apparatus to which it is connected.” Motor starting contactors areavailable 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 tostart and stop a motor in combination with suitable overload protection.”; a combination starter, which includesthe motor switching contactor as well as overload protection (described further below) and an integraldisconnecting device, is a type of combination controller. Low Voltage manual and magnetic controllers areclassified 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 forservice on 600 V or less. They are capable of interrupting operating overloads but not short circuits or faultsbeyond operating overloads.

Figure 8-4: Typical motor speed vs. time for the example above

0

0.2

0.4

0.6

0.8

1

1.2

0 4.5 913

.5 18 22.5 27 31

.5 36 40.5 45 49

.5 54 58.5 63 67

.5 72

Time (s)

Mot

or S

peed

(pu

sync

hron

ous)

Page 104: @Electrical

6

Class B: Class B controllers are DC air-break manual or magnetic controllers for service on 600 V or less. Theyare 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, andcapable 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 dutyapplications per [5]. Figure 8-5 shows a NEMA-rated low voltage contactor along with a manual motor startingswitch, a starter, and a combination starter.

Control of contactors using maintained-contact devices is referred to as two-wire control. Use of momentary-contact devices in the control of contactors is referred to as three-wire control. Three-wire control has theadvantage of allowing the contactor to open and remain open if the line voltage should fail; this arrangement istypical 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.

Medium voltage contactors are typically use vacuum as the interrupting means. Unlike a circuit breaker, a mediumvoltage vacuum contactor is specifically designed for long life in load-interrupting duty rather than for short-circuitinterrupting duty. However, unlike their low voltage counterparts a medium voltage contactor may be able tointerrupt short-circuit currents beyond operating overloads.

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 andpilot devices

a.) b.) c.) d.)

Figure 8-6: Low voltage contactor control: (full-voltage non-reversing control shown):

a.) Contactor nomenclature,

b.) Two-wire control,

c.) Three wire control

Page 105: @Electrical

7

Medium voltage air-break, vacuum, or oil-immersed controllers are classified by [7] as class E. Class E controllersare 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 interruptingshort circuits or faults exceeding operating overloads.

Class E2: Class E2 controllers employ their contacts for starting and stopping the motor and employ fuses forshort circuits or faults exceeding operating overloads.

Above 7200 V, motor control is generally accomplished using circuit breakers.

B.) Across-the-Line StartingThe most common method for starting an induction motor is across-the-line starting, where the motor is startedwith full voltage applied to the stator windings. Across-the-line starting uses contactors to energize the motor at fullline voltage. The motor acceleration will be as described above, and is dependent upon the line voltage, motoroutput 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.

Another form of across-the-line starting is full voltage reversing starting, in which the motor may be made to turnin either direction. This arrangement utilizes a full voltage reversing contactor with six poles, interlocked so thatonly one set of contacts may be closed at a given time. The contacts are connected so that in the reversedirection 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 currentwill be drawn during starting. This can cause voltage sags. Also, since the motor acceleration is dependent onlyupon the motor output torque and load torque characteristics (along with the line voltage level), the acceleration isnot as smooth as can be attained with other starting methods.

C.) Reduced-voltage autotransformer startingReduced-voltage autotransformer starting consists of initially starting the motor with an autotransformer, thenremoving the autotransformer from the circuit as the motor accelerates. This method results in a lower inrushcurrent as the motor starts, but also results in less available output torque when the autotransformer is in thecircuit. Autotransformer windings typically are tapped at 80, 65, and 50% voltage levels; the available outputtorque is related to the output torque when at full voltage by the equation:

(8-6)

where

TRV is the motor output torque at reduced voltage when the autotransformer is in the circuit

TFV is the motor output torque with full voltage applied

Figure 8-7: Low voltage across-the-Line starting implementation

Page 106: @Electrical

8

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 dutycapabilities of the autotransformer per NEMA ICS-9 must be taken into account; these will generally limit thelowest 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.

In figure 8-8 there are three contactors, labeled R, 1S, and 2S. The control scheme is designed so that the firstcontactor 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 apre-set time delay or current transition level, contactor 1S opens, leaving the motor energized through the non-common autotransformer windings. Once contactor 1S is open, contactor R closes, energizing the motor at fullvoltage. This is a closed-transition scheme; open transition schemes exist also.

D.) Reduced-voltage reactor or resistor startingReduced-voltage reactor or resistor starting consists of adding a resistor or reactor in series with the motor at thebeginning of the starting cycle, then shorting the resistor or reactor as the motor accelerates. This is generally aless expensive method than the reduced-voltage autotransformer method, but suffers the same limitations due tothe reactor or resistor thermal limits.

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

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

Figure 8-8: Low voltage implementation of a reduced-voltage autotransformer starter

Page 107: @Electrical

9

G.) Solid-state soft-startingSolid-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, whichcan 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 athyristor). An SCR is a device which conducts current in one direction when current is injected into its gateterminal, and blocks current in the other direction. The circuit symbology and nomenclature for an SCR, includingthe direction of current flows, is given in figure 8-10. In figure 8-10, if ig flows in the direction shown and Vak hasthe polarity shown, the SCR current i will flow in the indicated direction. If the gate current becomes zero, the SCRwill turn off; this is known as natural commutation. To stop the flow of current in the absence of a natural currentzero, 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, SCR’s 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 in-line contactor IC closes first. The firing circuit then causes the SCR’s to vary the motor voltage as required by thestarting parameters. Once the motor has accelerated to full speed, the shorting contactor SC closes, bypassingthe SCR’s and connecting the motor directly to full voltage.

The soft-start controller can also decelerate the motor in the same manner using the SCR’s.

Because the SCR’s dissipate heat, the equipment heat dissipation and the ambient temperature are concernswhen applying soft-start controllers and must be considered carefully.

Dedicated motor power factor correction capacitors must be switched out of the circuit during starting whenadjustable-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 andinternal diagnostic capabilities, making them a truly cutting-edge motor starting (and stopping) solution.

Figure 8-9: Circuit symbology and nomenclature for an SCR

Figure 8-10: Solid-state soft-starter, low voltage implementation

Page 108: @Electrical

10

H.) Rotor resistance startingApplicable only to wound-rotor motors, this starting method employs an adjustable external resistance which isconnected to the rotor via brushes and slip rings. The resistance in the rotor circuit dramatically alters the speed-torque characteristics of the motor during starting. The resistance in the rotor circuit is generally adjusted to behighest 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 valueof the external rotor circuit resistance.

These methods offer running speed control also. However they have been supplanted in recent years byadjustable-speed drives where speed control is required and by soft-starters where speed control is not required.

I.) Adjustable-speed drive startingAdjustable-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 adjustable-speed drive is among the best solutions available.

Because an adjustable-speed drive is not bypassed after starting, unlike a soft-starter, the harmonic currents itcauses to flow can affect the system power quality on a continuous basis. Also, power factor correction capacitorsmust not be used at the drive output to the motor.

J.) Medium voltage starting method implementationsAll of the starting methods mentioned above are generally available at the medium voltage level. However, thedisadvantages attributed to a given starting method are often exponentially more so when that method is appliedat 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 thereduced-voltage autotransformer, which at the medium voltage level typically employs a three-phaseautotransformer 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 surgearrestors 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 therequirements 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 ofthe general advantages and disadvantages for each of the starting methods discussed above.

Motor speed control methodsIt is often desirable to control the motor speed, usually for reasons process control for such variables as flow orpressure. Such applications as fans and pumps often have varying output requirements, and control of the motorspeed is more efficient than mechanically limiting the process output with such devices as throttling valves ordampers. The reason for this is due to the fact that for centrifugally-based processes (such as fans andcentrifugally-based pumps), the following relationships exist [1]:

(8-7)

(8-8)

Page 109: @Electrical

11

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 speedcontrol economically attractive [3]. To further this argument, consider the energy wasted when mechanical meanssuch as the throttling valves or dampers are used to control a process which is being driven from a motor runningat full speed. It is clear that motor speed control can be used to save energy by reducing wasted energy used tomechanically control the process.

A.) Adjustable-speed drivesBy far the most commonly-used AC motor control method is the use adjustable-speed drives. In most commercialand 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. Recallthat from eq. (8-1) the synchronous speed of a motor is a function of both the system frequency and the numberof poles of the motor. By varying the frequency, the motor speed may be varied so long as the motor is equippedto dissipate the heat at reduced speeds. Unlike soft-starting, specialized definite-purpose inverter-rated motordesigns 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

Cost-Effective

High Current Inrush

High Starting Torque

Abrupt Start

Reduced-voltageautotransformer

High output torque vs. starting current

Some Flexibility in starting characteristics due adjustabletaps on autotransformers

Limited duty cycle

Large equipment size due to autotransformers

Reduced-VoltageResistor orReactor

High output torque vs. starting current Limited duty cycle

Limited flexibility in starting characteristics

Higher inrush current than with reduced-voltageautotransformer

Large equipment size due to resistors/reactors

Wye-Delta Relatively low inrush current

Relatively simple starter construction

Good for long acceleration times

Relatively low output torque vs. starting current

Limited flexibility in starting characteristics

Requires special motor construction

Part-Winding Relatively Simple starter construction Relatively low output torque vs. starting current

Not suitable for frequent starts

Requires special motor construction

Solid-state soft starter

Smooth Acceleration

Low inrush current

High flexibility in starting characteristics

Typically offers deceleration control also

Typically integrates with industrial automationinfrastructure

Relatively Expensive

Sensitive to power quality

Heat dissipation and ambient temperature are a concern

Rotor Resistance Smooth acceleration available

Good flexibility in starting characteristics

Can be used for speed control also

Complicated controller design

Requires expensive wound-rotor motor construction

Adjustable Speed Drive

Smooth Acceleration

Low inrush current

High flexibility in starting characteristics

Offers deceleration and speed control also

Typically integrates with industrial automationinfrastructure

Cost-prohibitive unless speed control is required also

Sensitive to power quality

Heat dissipation and ambient temperature are a concern

Continuous harmonic currents can create power qualityissues

Page 110: @Electrical

12

overspeed operation can result in safety issues. Further, pulse-width modulated (PWM) drive outputs can causerepetitive voltage overshoots referred to as ringing, which can reduce the life expectancy of a general-purposemotor. As per [3], the motor manufacturer should be consulted before applying a general-purpose motor in anadjustable-speed drive application.

Various designs exist for adjustable-speed drives, however for low voltage drives the most prevalent is thevoltage-source pulse-width modulated design. As its name implies, the output is pulse-width modulated to reducethe output harmonic and noise content. The AC input to the drive is typically a diode rectifier. A simplified circuittopology for al voltage-source PWM drive is given in figure 8-11.

The output stage for the circuit in figure 8-11 consists of Insulated-Gate Bipolar Transistors (IGBT’s), which arecommonly used in low voltage PWM adjustable-speed drives instead of SCR’s due to their superior switching ratecapability.

Adjustable speed drives offer superior speed control for motors through 10,000hp, depending upon the systemvoltage [1]. They usually incorporate protection for the motor as well, allowing the omission of separate motorprotective relays if desired. Due to the high switching frequencies involved and their interaction with the cablecapacitance, the length of the cable runs between the output of the drive and the motor are limited, and, asmentioned above for soft-starters, power factor correction capacitors and surge capacitors should not be used atthe output of an adjustable speed drive. Also due to the high switching frequencies, common-mode noise on thegrounding 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-allsystem 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 withthe automation infrastructure of a facility.

With the exception of a few isolated cases, for most industrial and commercial facilities adjustable speed drivesare the speed control of choice for AC motors.

B.) Older methodsVarious other methods exist for AC motor speed control. A few of these are:

� Rotor-resistance speed control – similar to rotor-resistance starting, this method consists of varying the effectiveresistance in the rotor of a wound-rotor induction motor to vary the speed. Variants of this method include rotorpower 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 motorsand 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 adjustablespeed drive is typically the solution of choice for most commercial and industrial facilities.

Figure 8-11: Voltage-source PWM adjustable-speed drive: simplified circuit topology for low voltage implementation

Page 111: @Electrical

13

Motor stopping devicesSeveral methods for motor stopping exist. Which, if any, is to be used is dependent upon the application.

A.) Dynamic brakingOn form of dynamic braking involves the disconnection of AC power an induction motor and connecting DC powerto 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 energyback to the drive, where it is dissipated via a braking resistor.

B.) PluggingPlugging is the reversal of the phase sequence on an induction motor via switching two phase connections to themotor, which will cause the motor to come to a very rapid stop due to torque developed on the rotor in theopposite direction from the current running direction. A zero-speed switch should be used to prevent reversal ofthe motor.

C.) Mechanical brakingUnlike 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 ofan electrical power failure.

D.) Adjustable speed drive and soft-start controller decelerationFor those applications not require fast deceleration times but a controlled deceleration is required, the speedramping capabilities inherent in most adjustable speed drives and soft-start controllers may be used. For fasterstopping with an adjustable-speed drive dynamic braking is required.

Motor protectionMotor protection involves protection of a motor from abnormal conditions. The most common abnormal conditionis an overload, which can produce damaging heating effects in the motor. For this reason, overload relays are theprimary means of motor protection. However, short-circuit protection is also required to minimize damage to themotor from an internal short-circuit. Other protective devices are also available, their use depending upon the sizeof the motor and the cost of protection vs. the cost of the motor.

A.) Nameplate valuesWhile these are not the only values marked on a motor nameplate, the following are the nameplate values mostimportant 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 themotor at zero speed. It is the initial current upon full-voltage energization of the motor. If given as a code letter, thecode 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 maximumcontinuous output of the motor without exceeding a defined temperature rise in the motor.

Page 112: @Electrical

14

B.) Low voltage motor protectionLow 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 inputcurrent is always larger than would normally be dictated by the output power due to losses and the motor powerfactor. NEC Article 430 [6] gives typical full-load currents where a machine’s actual full-load current is not known.However, for overload protection purposes the motor nameplate full-load current rating must be used ([6] Article430.32 (A) (1)). The NEC basic requirement for overload protection is 125% of the nameplate rating for motorswith 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 voltagemotors fine-tuning of the relay selection should be made according to the motor manufacturer’s recommendations.

Typically, overload relays for low voltage motors are classified as melting alloy, bimetallic, or solid-state. Ingeneral, melting alloy relays are hand-reset devices, whereas bimetallic relays can be self-resetting or hand-resetting. Bimetallic relays are available as temperature-compensated or non-compensated; non-compensated isan advantage when the relay and motor are in the same ambient temperature since the relay opening timechanges with the temperature in a similar manner to the motor [2]. Temperature compensated relays are designedfor operation where the motor is at a constant ambient temperature but the relay is at a varying ambienttemperature. While melting-alloy and bimetallic overload relays must be selected to suit the motor, for a solid-staterelay the same physical relay may be used for several different types of motors, with the settings adjusted on therelay to match the motor it is protecting. Some solid-state relay models also have the advantage of providingphase-loss protection.

Note that NEMA ICS 2-2000 classifies motor overload relays into three classes, Class 10, 20, and 30, dependingupon the time delay to trip on locked-rotor current. NEMA Class 10 overload relays will trip in 10s at 6x theoverload 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 circuitprotectors). Short-circuit protection is constrained by NEC article 430-52 [6] gives limits for various types ofmotor/protective device combinations. In general, however, the lowest rating that does not cause nuisance trippingdue 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 thecontactor. Note that the motor circuit conductors, per NEC Article 430.22 (A), must be sized to have an ampacitynot less than the 125% of the motor full-load current as determined from the tables in Article 430, not from thefull-load current marked nameplate rating. The purpose of this is avoid undersized cables should the motor bereplaced in the future with a different make and model of the same hp rating, since the hp rating does not clearlydefine the full-load current of the motor. Unlike conventional branch circuits, overcurrent protective devices inmotor branch circuits do not dictate the conductor sizes. For this reason, care must be taken to insure that themotor 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 damagecurves, 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. FromNEC 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 manufacturer’s recommendations, is selectedfor overload protection. A magnetic-only circuit breaker, sized at 800 A, which is in accordance with the 800% ofmotor full-load current per NEC Table 430.52 [6], is used for short-circuit protection. The motor switching devicesis a NEMA size 6 contactor. A one-line representation of this motor branch circuit is shown in figure 8-12.

Page 113: @Electrical

15

The resulting time-current coordination is shown in figure 8-13. Note that in figure 8-13 the purple curve to theright of the overload relay curve is the motor thermal damage curve, obtained from the motor manufacturer. If thiscurve is not available the relay or motor manufacturer’s selection tables should be used for selection of theoverload relay. The thermal overload relay protects the motor from overloads, while at the same time not openingthe motor inrush current or full-load current, as denoted by the purple “MOTOR” current curve to its left. Note thehigh-current region (with current equal to the motor locked-rotor current) with an acceleration time ofapproximately 9 seconds; this curve is dependent upon the connected load and must also obtained from themotor manufacturer unless the motor acceleration time can be determined. In many cases the motor startingcurrent 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 30applied for high-inertial loads [2]. The magnetic-only circuit breaker protects the cable for short-circuits, asdenoted 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 10times without servicing per [5] which is more than the maximum trip current of the circuit breaker; the contactor istherefore adequately protected. The motor and its branch circuit is, therefore, adequately protected.

For larger motors on solidly-grounded systems low voltage ground-fault protective devices may also be required toallow coordination with upstream ground-fault protective devices. The application of these falls under the sameguidelines as given in “System protection” scetion (section 7 in this guide).

Figure 8-12: Example low voltage motor protection branch circuit

10

10

100

100

1K1K

10K

10K

100K

100K

0 .01 0.01

0.10 0.10

1 1

10 10

100 100

1000 1000

C UR R E NT IN AMP E R E S

TIM

E IN

SE

CO

ND

S

CABLE

CB

O.L.

MOTOR

CABLE

CB

O.L.

MOTOR

Figure 8-13: Time-current coordination for circuit of figure 8-12

Page 114: @Electrical

16

In addition to the overload protection described above, thermostats are commonly installed in three-phaseindustrial-service 460 V motors from 11 kW through 150 kW (14-200hp) [2]. These are bimetallic devices thatoperate 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 energizationwhen an undervoltage condition exists.

C.) Medium voltage motor protectionThe 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 protectionas well as a host of other protection features which protect the motor from other abnormal conditions. R-ratedfuses are typically used for short-circuit protection.

Some of the additional protective elements utilized for medium voltage motors include:

� RTD’s – Resistance Temperature Detectors are typically made of platinum, nickel, or copper and exhibit anincreasing resistance with increasing temperature. The RTD resistance is used to monitor the temperature atvarious points in the motor, typically in the stator windings. The temperature is used to provide precise overloadprotection for the motor. Per [2], RTD’s 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-sequencecurrents, which can be caused by unbalanced voltages.

� Phase sequence (Device 47) – This is used to prevent the single-phasing of three-phase motors, which cancause 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 differentialprotection, motors can also be equipped with self-balancing differential protection in which only one CT is usedfor 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-impedancedifferential 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-resistance-grounded systems are provided with ground-fault protection. This is accomplished with a zero-sequence CT andis almost always instantaneous.

87M 87M

87M

87M

87M

a.) b.)

87M

Figure 8-14: Motor differential protection:

a.) Traditional

b.) Self-balancing

Page 115: @Electrical

17

Typical overload and short-circuit protection for a medium voltage motor may be illustrated by considering thefollowing: A 750 hp, 4160 V motor is to be protected. The motor has a nameplate full-load current value of 96 Aand a locked-rotor current of 576 A, and a service factor of 1.15. A microprocessor-based motor protection relay isto 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 motoroverload protective device(s) are to trip. The pickup value for the overload protection is to be set equal to theservice factor times the nameplate full-load current, which is 110.4 A; the cables are copper in undergroundconduit, therefore the cable size selected is #2AWG, with an ampacity of 145A per NEC table 310.77 [6]. The CTprimary ratings for the motor protection relay are typically selected as no less than 1.5 times the motor full loadcurrent to avoid saturation (must be checked!) – in this case 200:5. To coordinate with the overload protection andprotect the motor branch circuit cables and motor, a 6R fuse is chosen (note that per NEC Article 430.225 [6] themotor overcurrent protection must be coordinated to automatically interrupt overload and fault currents in themotor, but there is not specific constraint given for the short-circuit protection, unlike the requirements for motorsunder 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.

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 andfull-load current does not trip the overload relay. Unlike the case for a low voltage motor, this is typically availablefrom the manufacturer, who has analyzed the motor’s performance when connected to the driven load. Note alsothat the purple motor thermal damage curve is to the left and above the relay overload curve, indicating the motoris protected for overloads. The same applies for the red “MV CABLE” rated full-load current marker at the top ofthe plot. The motor thermal damage curve is obtained from the motor manufacturer; if the entire curve is notavailable, 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 forshort-circuits by the fuses. Finally, note that the fuse total clearing and overload relay curves cross atapproximately 900 A; this is well above the inrush of 576 A, but well below the contactor rating of 5000 A. Thefuse will therefore clear faults above the contactor’s 5000 A rating before the contactor opens.

NEC requirements for motorsThe following are highlights from the NEC [6] requirements for motors. This is not intended to list all NECrequirements for motors, but to illustrate the major points that apply in the most common motor installations andaffect the power system design. For the full text of the complete NEC requirements for motors, consult the NEC.

A.) GeneralThe NEC basic requirements for motors, motor circuits, and controllers are given in NEC Article 430 [6], and aresupplemented by additional articles for specific motor-driven equipment. Article 430 is divided into 14 parts. Therequirements which apply to each part of a motor circuit are illustrated in figure 8-17.

Figure 8-15: Example medium voltage motor circuit, excluding ground-fault protection

Page 116: @Electrical

18

One of the main premises of Article 430 is the fact that the hp or kW rating of a motor is the output rating, asdiscussed above. The motor electrical input characteristics will vary based upon the motor design. Therequirements 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 devicessuch as encoders, tachometers, thermal switches, and detectors, air blowers, heaters, and vibration sensors.

MV CABLERELAY

10

10

100

100

1K1K

10K

10K

100K

100K

0 .01 0.01

0.10 0.10

1 1

10 10

100 100

1000 1000

C UR R E NT IN AMP E R E S

TIM

E IN

SE

CO

ND

S

MV CABLEMV FUSE

RELAY

MV MOTOR

MV CABLEMV FUSE

RELAY

MV MOTOR

Figure 8-16: Time-current coordination for circuit of figure 8-15

Figure 8-17: Illustrated NEC article 430 contents [6]

Page 117: @Electrical

19

Adjustable-Speed Drive System: An interconnected combination of equipment that provides a means ofadjusting the speed of a mechanical load coupled to a motor. A drive system typically consists of an adjustablespeed 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 themotor 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 theperformance of the controller but does not carry the main power current.

System Isolation Equipment: A redundantly monitored, remotely operated contactor-isolating system, packagedto provide the disconnection/isolation function, capable of verifiable operation from multiple remote locations bymeans 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 todetermine the motor ampacity for all purposes other than for overload protection, per Article 430.6. This does notapply to low-speed (less than 1200 RPM) motors, motors built for high torques, multispeed motors, equipmentemploying 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 motorampacity tables (Tables 430.248, 430.248, 430.29, and 430.250) are referenced in hp; for motors marked inamperes, the horsepower must be determined by finding the hp corresponding with the nameplate ampacity, usinginterpolation 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 currentrequirements of the motor, yet it is possible that a replacement for a given motor would be selected based onlyupon 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 circuitconductors be rated not less than 125% of the motor’s 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 mustbe based upon the highest of the full-load current ratings shown on the motor nameplate, with the branch-circuitconductors between the controller and the motor based upon the current rating of the winding(s) that theconductors energize. [430.22 (B)]

For a wye-start, delta-run connected motor, the selection of branch circuit conductors must be based upon themotor full-load current. The conductors between the controller and the motor must be based upon 58% of themotor full-load current. [430.22 (C)]

For a part-winding connected motor, the selection of branch-circuit conductors on the line side of the controllermust be based upon the motor full-load current. The selection of conductors between the controller and the motormust 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 thanas shown in table 430.22 (E), which gives percentages of the nameplate full-load current rating for variousclassifications of service and motor duty ratings. [430.22(E)]

For continuous-duty motors with wound-rotor secondaries, conductors between the secondary (rotor) to thecontroller must have an ampacity no less than 125% of the full-load secondary current of the motor. For other thancontinuous duty motors, the percentages in table 430.22 (E) apply. If there is a resistor separate from thecontroller, Table 430.23 (C), which is based upon the resistor duty classification, contains percentages of full-loadsecondary current to which the conductors between the controller and the resistor must be compared. [Article430.23).

Page 118: @Electrical

20

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 othermotors in the group, plus the ampacity required for other loads. Various exceptions apply, and the authority havingjurisdiction may grant permission for a lower ampacity, provided the conductors have the ampacity for themaximum load determined in accordance with the sizes and number of motors supplied and the character of theirloads and duties. [430.24, 430.25]

Where a motor installation includes a capacitor connected on the load side of the motor overload device, theeffect 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 devicethat is responsive to motor current or a thermal protector integral with the motor that will protect from dangerousoverheating and failure to start. Motors that are part of an approved assembly that does not subject the motor tooverloads may be protected by an integral device that protects the motor against failure to start. Motors largerthan 1500 hp must be protected by a protective device with imbedded temperature detectors that cause current tothe motor to be interrupted when the motor attains a temperature rise greater than marked on the motornameplate in an ambient temperature of 40˚C. [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 RTD’s and their associatedrelay(s).

If overload relays are used, for motors with a marked service factor of 1.15 or greater or a marked temperaturerise of 40˚C or less may be set at a maximum of 125% of the motor nameplate full-load current rating. For allother motors the maximum overload relay setting is 115% of the motor nameplate full-load current rating. If thesevalues do not allow the motor to start or carry the load, higher-size sensing elements or incremental settings maybe permitted to be used so long as they do not exceed 140% of the motor nameplate full-load current rating formotors with a service factor of 1.15 or greater or a temperature rise of 40˚C 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 ornon-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-faultprotective device. [430.33]

Motor overload devices for non-automatically started motors may be shunted or cut out of the circuit duringstarting. With some exceptions, an automatically-started motor cannot have their overload devices shunted or cutout 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 unlessapproved 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 orincreased hazard(s) to a person(s) and continued motor operation is necessary for a safe shutdown of equipmentor process, the motor overload protection may be permitted to be connected to a supervised alarm rather thancausing 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, therating 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].

Page 119: @Electrical

21

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 givesmaximum ratings, in percentage of the motor full-load current, which can be used for various motor and protectivedevice 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, thenext higher standard size is permitted. If the value for the protective device rating is no sufficient for the startingcurrent 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 themanufacturer’s overload relay table for use with a motor controller or are otherwise marked on equipment, theymust 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 motorcontroller 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 morewindings, so long as the rating of the protective device does not exceed the percentage per Table 430.52 of thesmallest 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 thedevices 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 longas the adjustable instantaneous trip settings do not exceed 1300% of full-load motor current for other than DesignB energy-efficient motors and 1700% of full-load current for Design B energy-efficient motors. The same appliesfor a motor short-circuit protector, so long as it is part of a listed motor controller having coordinate motor overloadprotection 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:

� The motors are not over 1hp, the branch circuit is 120 V and protected at not over 20 A 600 V or less protectedat not over 15 A, the full-load rating of each motor does not exceed 6 A, the rating of the branch-circuit short-circuit and ground-fault protective device marked on any of the controllers is not exceeded, and individualoverload protection confirms to 430.32, OR,

� If the branch circuit short-circuit and ground-fault protective device is selected not to exceed the requirements of430.52 for the smallest rated motor, each motor has individual overload protection, and it can be determined thatthe branch-circuit short-circuit and ground-fault protective device will not open under the most severe normalconditions 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 devicemust 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 ratingor 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 insight from the controller location unless the circuit is over 600 V, in which case a controller disconnecting meanscapable of being locked in the open position is permitted to be out of sight of the controller if it is marked with awarning label giving the location of the disconnecting means. A single disconnecting means is permitted for a

Page 120: @Electrical

22

group of coordinated controllers that drive several parts of a single machine or piece of apparatus. In this case thedisconnecting means must be located in sight from the controllers, and both the disconnecting means and thecontrollers 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 locationunless the controller disconnecting means is individually capable of being locked in the open position and eithera.) such a location of the disconnecting means is impracticable or introduces additional or increased hazards topersons or property, or b.) the motor is in an industrial installation where conditions of maintenance andsupervision ensure that only qualified persons service the equipment. The controller disconnecting means perabove may be permitted to serve as the disconnecting means for the motor if it is located in sight from the motorlocation and the driven machinery location. [430.102]

The disconnecting means must open all ungrounded supply conductors and must be designed so that no pole canbe 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. Thedisconnecting means may be a listed motor circuit switch rated in horsepower, a listed molded case circuitbreaker, a listed molded case switch, or an instantaneous trip circuit breaker that is part of a listed combinationcontroller. Listed manual motor controllers additionally marked as “suitable as motor disconnect” are permitted asa disconnecting means where installed between the final motor branch-circuit short-circuit protective device andthe motor. [430.103, 430.104, 430.109 (A)]

System isolation equipment must be listed for disconnection purposes. Where system isolation equipment is usedit must be installed on the load side of the overcurrent protection and its disconnecting means. The disconnectingmeans must be a listed motor-circuit switch rated in horsepower, a listed molded case circuit breaker, or a listedmolded-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 notless than twice the full-load current rating of the motor, a general-use AC snap switch for use only on AC, or alisted manual motor controller with a hp rating not less than the motor hp and marked “suitable as motordisconnect” 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 tobe 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 loessthan the motor ratings may use the attachment plug and receptacle and the disconnecting means. Cord-and-plugconnected appliances per 422.32, room air conditioners per 440.63, or a portable motor rated 1/3 hp or less donot 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 themotor, unless it is rated in hp and has a hp rating not less than the hp of the motor. For torque motors thedisconnecting means must have a an ampere rating of at least 115% of the motor nameplate current. A methodfor 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 ofa single machine or piece of apparatus, a group of motors is under the protection one set of branch-circuitprotective devices as permitted by 430.53 (A), or where a group of motors is in a single room within sight from thelocation of the disconnecting means. [430.112]

Where a motor or motor-operated equipment receive electrical energy from more than one source, each sourcemust be provided with a disconnecting means from each source of electrical energy immediately adjacent to theequipment served. The disconnecting means for the main power supply to the motor is not required to beimmediately 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 interruptingthe locked-rotor current of the motor. An autotransformer controller must provide an “off” position, a runningposition, and at least one starting position, and designed so that it cannot rest in the starting position or in any

Page 121: @Electrical

23

position that will render the overload device in the circuit inoperative. Motor starting rheostats must be designedso 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 bedamaged 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 atthe application voltage not lower than the horsepower rating of the motor. A branch circuit inverse time circuitbreaker or molded case switch is permitted as a controller for all motors. For stationary motors 2hp or less and300 V or less, a general-use switch having an ampere rating not less than twice the full-load current rating of themotor or an AC – only snap switch where the motor full-load current rating is not more than 80% of the ampererating 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 inwhich the nominal voltage between any two conductors does not exceed the controller’s voltage rating. A controller with a slash rating, for example, 480 Y/277 V, may only be applied on a solidly-grounded circuit wherethe nominal voltage to ground from any conductor does not exceed the lower of the two values of the controller’svoltage rating and the nominal voltage between any two conductors does not exceed the higher value of thecontroller’s 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 thepole which disconnects the grounded conductor cannot open without simultaneously opening all conductors of thecircuit. [430.85]

Each motor must have its own individual controller, unless a number of motors drive several parts of a singlemachine or piece of apparatus, a group of motors is under the protection one overcurrent device as permitted by430.53 (A), or where a group of motors is in a single room within sight from the location of the disconnectingmeans. An air-break switch, inverse time circuit breaker, or oil switch may be permitted to serve as the controllerand disconnecting means if it complies with the requirements for controllers in 430.83, opens all ungroundedconductors to the motor, and is protected by an overcurrent device in each ungrounded conductor. Anautotransformer type controller must be provided with a separate disconnecting means. Inverse-time circuitbreakers 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 thedisconnecting 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 therequirement to the need for the two disconnecting means to be adjacent to each other. Control transformers incontroller 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 controlcircuit that are outside the control device itself must be installed in a raceway or otherwise suitably protected fromphysical damage. Where one side of the motor control circuit is grounded, the motor control circuit must bearranged so that an accidental ground in the control circuit remote from the motor controller will not start the motoror bypass manually operated shutdown devices. [430.73]

H.) Adjustable-speed drive systemsBranch/feeder circuit conductors that supply power conversion equipment included as part of an adjustable-speeddrive 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 thanrequired by 430.6 (see above). [430.122]

Page 122: @Electrical

24

Where the power conversion equipment is marked to indicate that motor overload protection is included, additionaloverload 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 overloadprotection per part III is required. [430.124]

Adjustable speed drive systems must protect the motor against overtemperature conditions by means of a motorthermal protector per 430.32, an adjustable speed drive controller with load and speed-sensitive overloadprotection and thermal memory retention upon power loss, overtemperature protection relay utilizing thermalsensors embedded in the motor and meeting the requirements of 430.32 (A)(2) or (B)(2), or a thermal sensorembedded in the motor that is received and acted upon by an adjustable speed drive. Motors that utilize externalforced-air or liquid cooling systems must be provided with protection that will be continuously enabled or enabledautomatically if the cooling system fails. For multiple motor applications, individual motor overtemperatureprotection must be provided. The provisions of 430.43 and 430.44 apply to motor overtemperature protectionmeans. [430.24]

The disconnecting means is permitted to be in the incoming line conversion equipment and must have a rating ofnot 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. Theampere rating or setting of the overcurrent protective device must not exceed the rating of the common powerbus. This overcurrent protection may be provided by an overcurrent protective device located ahead of the motorcontrol 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 beprovided with a protective device having a rating or setting not greater than the largest rating or setting of thebranch-circuit short-circuit and ground-fault protective device for any motor supplied by the feeder, plus the sum ofthe full-load currents of the other motors of the group. The largest rating or setting of the branch-circuit short-circuit and ground-fault protective device is based upon the maximum permitted size per Article 430.52. and Table430.52. Where one or more instantaneous trip circuit breakers or motor short-circuit protectors are used for motorbranch-circuit and ground fault protection as permitted in 430.52(C), each instantaneous trip circuit breaker ormotor short-circuit protector must be assumed to have a rating not exceeding the maximum percentage of motorfull-load current permitted by Table 430.52 for the type of feeder protective device employed. Where the feederovercurrent protective device also provides overcurrent protection for a motor control center, the provisions of430.94 apply. [430.62]

Where a feeder supplies a motor load and, in addition, a lighting or lighting and appliance load, the feederprotective device must have a rating sufficient to carry the lighting and appliance load, plus the rating permitted by430.52 for a single motor, the rating permitted by 440.22 for a single hermitic refrigerant motor-compressor, or therating 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 overloadprotective device(s) is selected to trip [430.224]

Each motor circuit must include coordinated protection to automatically interrupt overload and fault currents in themotor, the motor circuit conductors, and the motor control apparatus. This may be a thermal protector integral tothe motor or external current-sensing devices, or both. The secondary circuits of wound-rotor AC motors areconsidered to be protected against overcurrent by the motor overload protection means. Operation of the overloadinterrupting device must simultaneously disconnect all ungrounded conductors. Automatic reset of overloadsensing devices is prohibited after trip unless resetting does not cause automatic restarting or there is no hazardto persons due to automatic restarting. Where a motor is vital to operation of the plant and the motor should

Page 123: @Electrical

25

operate to failure if necessary to prevent a greater hazard to persons, the sensing device(s) are permitted to beconnected 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 withouthazard, or fuses. A circuit breaker must open each ungrounded conductor. Fuses must be placed in eachungrounded conductor and must be furnished with a disconnecting means (or be of the type that can serve as thedisconnecting means) and arranged so that they cannot be serviced while energized. Automatic reclosing of thefault-current interrupting device is not permitted unless the circuit is exposed to transient faults and suchautomatic reclosing does not create a hazard to persons. Overload and fault-current protection may be providedby the same device. [430.225]

The ultimate trip current of overload relays or other motor-protective devices must not exceed 115% of thecontroller’s continuous current rating. Where the motor branch-circuit disconnecting means is separate from thecontroller, the disconnecting means current rating must not be less than the ultimate trip setting of the overcurrentrelays in the circuit.

The controller disconnecting means must be capable of being locked in the open position.

L.) Other articlesOther NEC [6] articles which apply to motors and augment or amend the provisions in article 430 are given intable 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 andStairway 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, NEMAStandards 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, NEMAStandards Publication ICS 2-2000.

[6] The National Electrical Code, NFPA 70, The National Fire Protection Association, Inc., 2005 Edition.

[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.

Page 124: @Electrical

1

Section 9: Power Distribution EquipmentBill Brown, P.E., Square D Engineering Services

IntroductionPower 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 powerdistribution equipment: Panelboards, Switchboards, Low Voltage Motor Control Centers, Low Voltage Switchgear,Medium Voltage Power and Distribution Transformers, Medium Voltage Metal Enclosed Switchgear, MediumVoltage Motor Control Centers, and Medium Voltage Metal-Clad switchgear. Each has its own unique standardsand application guidelines, and one facet of good power system design is the knowledge of when to apply eachtype of equipment and the limitations of each type of equipment. All of these equipment described herein aretypically custom-engineered on a per-order basis.

NEMA enclosure typesOne common characteristic of all of the equipment types covered in this section is that they are all enclosed forsafety. 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 powerequipment 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 tohazardous 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 personnelagainst access to hazardous parts; to provide a degree of protection of the equipment inside the enclosureagainst ingress of solid foreign objects (falling dirt and windblown dust); to provide a degree of protection withrespect to harmful effects on the equipment due to the ingress of water (rain, sleet, snow); and that will beundamaged 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 personnelagainst access to hazardous parts; to provide a degree of protection of the equipment inside the enclosureagainst ingress of solid foreign objects (falling dirt and windblown dust); to provide a degree of protection withrespect to harmful effects on the equipment due to the ingress of water (rain, sleet, snow, splashing water, andhose 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 personnelagainst access to hazardous parts; to provide a degree of protection of the equipment inside the enclosureagainst ingress of solid foreign objects (windblown dust); to provide a degree of protection with respect to harmfuleffects 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 externalformation of ice on the enclosure.

Type 5: Enclosures constructed for indoor use to provide a degree of protection to personnel against access tohazardous parts; to provide a degree of protection of the equipment inside the enclosure against the ingress ofsolid foreign objects (falling dirt and settling airborne dust, lint, fibers, and flyings); and to provide a degree ofprotection with respect of harmful effects on the equipment due to the ingress of water (dripping and lightsplashing).

Type 12: Enclosures constructed (without knockouts) for indoor use to provide a degree protection to personnelagainst access to hazardous parts; to provide a degree of protection of the equipment inside the enclosureagainst ingress of solid foreign objects (falling dirt and circulating dust, lint, fibers, and flyings); and to provide adegree of protection with respect to harmful effects on the equipment due to the ingress of water (dripping andlight splashing).

Page 125: @Electrical

2

PanelboardsTable 9-1: Quick reference – Panelboards

Panelboards are the most common type of power distribution equipment. A panelboard is defined as “a singlepanel or group of panel units designed for assembly in the form of a single panel, including buses and automaticovercurrent 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; andaccessible 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.

Panelboards are used to group the overcurrent protection devices for several circuits together into a single pieceof equipment. In small installations they may serve as the service equipment. The NEC [2] divides panelboardsinto two categories:

Lighting and Appliance Branch-Circuit Panelboard: A panelboard having more than 10 percent of itsovercurrent devices protecting lighting and appliance branch circuits.

Power Panelboard: A panelboard having 10 percent or fewer of its overcurrent devices protecting lighting andappliance branch circuits.

Lighting and appliance branch-circuit panelboards are limited to a maximum of 42 overcurrent devices, excludingmains. UL 67 [3] designates Class CTL Panelboard as the marking for appliance and branch circuit panelboards;CTL stands for “circuit limiting.” In some manufacturer’s literature lighting and appliance branch-circuitpanelboards for residential or light commercial use are referred to as loadcenters.

Figure 9-1: 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

Page 126: @Electrical

3

Panelboards are available with built-in main devices or as main lugs only (MLO). The NEC [2] requires applianceand branch circuit panelboards to be individually protected on the supply side by not more than two main circuitbreakers or two sets of fuses having a combined rating no greater than the rating of the panelboard. Lighting andappliance branch circuit panelboards are not required to have individual protection if the feeder overcurrent deviceis no greater than the rating of the panelboard. Power panelboards must be protected by an overcurrent devicewith 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 forthe panelboard itself are purchased separately as well. This is typically true of larger panelboards and gives agreat 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 panelboardbus without the need for overcurrent devices. Circuit breaker locking devices allow locking of circuit breakers inthe open or closed position (note that the breakers will still trip on an overcurrent condition). Various types of trimsare available, with various locking means available for trims that are equipped with doors.

SwitchboardsTable 9-2: Quick reference – Switchboards

The definition of a switchboard is “a large single panel, frame, or assembly of panels on which are mounted onthe 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 wellas from the front. They may consist of multiple sections, connected by a common through-bus. Unlikepanelboards, 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 levelupstream from panelboards in the electrical system, and in some small to medium-size electrical systems theyserve as the service equipment. Figure 9-2 shows an example of a switchboard.

Switchboards are available with a main circuit breaker or fusible switch, or as main-lugs only. The availableampacities and multi-section availability makes them more flexible than panelboards. They are generally available

Available voltage ratings 120-600 V

Available current ratings 800-5000 A

Available short-circuit ratings Through 200 kA

Major industry standards UL 891, NEMA PB 1, NEC

Typical enclosure types 1, 3R

Primary NEC requirements Article 408

Figure 9-2: Switchboards

Page 127: @Electrical

4

utilizing either copper or aluminum bussing, and with a variety of bus plating options. Custom bussing for retrofitapplications is also possible.

Switchboard circuit breakers may be stationary-mounted (also referred to as fixed-mounted), where they can beremoved only by unbolting of electrical connections and mounting supports, or drawout-mounted, where they canbe without the necessity of removing connections or mounting supports. It is possible to insert and removedrawout devices with the main bus energized. The section which contains the main circuit breaker(s) or servicedisconnect devices is referred to as a main section. A section containing branch or feeder circuit breakers isreferred to as a distribution section.

Devices mounted in the switchboard may be either panel mounted (also referred to as group mounted), wherethey are mounted on a common base or mounting surface, or individually mounted, where they do not share acommon 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 accessibleto personnel unless special means for access are used is referred to an isolated device. Figure 9-3 showsexamples 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 thethrough-bus is referred to as the section bus (also known as vertical bus). The bussing that connects the sectionbus to the overcurrent devices is referred to as the branch bus. Section and branch busses may be smaller thanthe main through-bus; if this is the case UL 891 [2] gives the required section bus size as a function of the numberof overcurrent devices connected to it.

Switchboards are available with a number of accessories, including custom-engineered options such as utilitymetering compartments, automatic transfer schemes, and modified-differential ground fault for switchboards withmultiple mains. However, the internal barriering requirements are minimal.

Low voltage motor control centersTable 9-3: Quick reference – Low voltage motor control centers

Figure 9-3:

a.) Group-mounted devices

b.) Individually-mounted devices

Available voltage ratings 120-600 V

Available current ratings 600-2500 A

Available short-circuit ratings Through 100 kA

Major industry standards NEMA ICS-18, UL 845, NEC

Typical enclosure types 1, 3R, 12

Primary NEC requirements Article 430

a.) b.)

Page 128: @Electrical

5

A motor control center (MCC) is defined as “a floor-mounted assembly of one or more enclosed vertical sectionstypically 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 locationwith a common power bus. Figure 9-4 shows an example of a 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 diagramssupplied 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 connectedeither 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 ofvertical sections which contain combination motor control units or control assemblies; the field load wiring isconnected to master power terminal blocks mounted at the top or bottom of vertical sections or directly to thedevice terminals.

MCCs generally consist of a common power bus and a vertical bus for each section to which combination motorcontrollers are plugged on. The individual plug-in units are often referred to as buckets and may be inserted andremoved 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 footprintvs. individual control cabinets, and like switchboards are available with many options. Removable plug-on unitsallow 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.

Figure 9-4: Low voltage motor control center

Page 129: @Electrical

6

Low voltage switchgearTable 9-4: Quick reference – Low voltage switchgear

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:

� 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 theadvantages of low voltage power circuit breakers, discussed in Section 7, can be utilized to enhance coordinationand reliability. It is typically used as the highest level of low voltage equipment in a facility of this type and, if theutility service is a low voltage service, the service entrance switchgear as well. Figure 9-5 shows an example oflow voltage switchgear.

Low voltage switchgear, although it performs the same functions and has comparable availability of voltage andampacity ratings as switchboards, represents a different mode of development from switchboards and is, ingeneral, 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 overswitchboards where coordination, reliability, and maintenance are a primary concern.

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 usuallyavailable to separate the main bus from the cable terminations, forming separate bus and cable compartmentswithin a section, as well as side barriers to separate adjacent cable and bus compartments.

Available voltage ratings 120-600 V

Available current ratings 1600-5000 A

Available short-circuit ratings Through 200 kA

Major industry standards ANSI/IEEE C37.20.1, ANSI/IEEE C37.51, NEMA SG-5,CAN/CSA C22.2 NO 31-M89, UL 1558

Typical enclosure types 1, 3R

Figure 9-5: Low voltage Switchgear

Page 130: @Electrical

7

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 morenumerous than those for switchboards due to the nature of switchgear service conditions.

Medium voltage power and distribution transformersTable 9-5: Quick reference – Medium voltage power and distribution transformers

Medium voltage power and distribution transformers are used for the transformation of voltages for the distributionof 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 windingsmay 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 thewindings than the VPI or VPE types. In some cases the primary windings are cast-resin and the secondarywindings 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 reviewedto insure that installation requirements can be adequately met, and maintenance must be taken underconsideration 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 havelow first-costs, have longer lifetimes than liquid-filled units, and are much easier than liquid-filled types to installindoors; however, consideration should be given to the absorption of moisture by the windings if these are usedoutdoors. Installed indoors, these have expected service lifetimes of around 30 years. Cast-resin, dry-typetransformers have generally high first-costs compared to the other types, but have the same installationrequirements 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 typegenerally mounted outdoors and with specific tamper-resistance features to prevent inadvertent access by thegeneral public, and unit substation type, which is an industrial-type enclosure suitable for close-coupling into anintegrated unit substation lineup with primary and secondary equipment (note that unit substation-styletransformers 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. Coolingtypes are listed as AA (ambient air) for dry-type transformers without fans, and AA/FA (ambient air/forced air) fordry-type transformers with fans, for an increase of 33% in kVA capacity. The cooling type for a liquid-filledtransformer is listed as OA for units without fans, OA/FA for units with fans, with an increase of 15% kVA capacityfor units 225-2000 kVA, and 25% for units 2,500-10,000 kVA. “FFA” (future forced air) options are usuallyavailable for both dry and liquid-filled types, although experience has shown that the fans are almost never addedin the field.

Available primary voltage ratings 2400 - 38 kV

Available secondary voltage ratings 120 - 15 kV

Available kVA ratings Through 10,000 kVA

Major industry standards ANSI/IEEE C57 Series (All Types)UL 1562 (Dry and Cast-Resin Types)

Typical enclosure types 1, 3R

Page 131: @Electrical

8

Table 9-6 gives typical BIL levels for medium voltage power and distribution transformers. These apply to both theprimary and secondary windings. Table 9-7 gives typical design temperature rises.

Table 9-6: Typical BIL levels for medium voltage power and distribution transformers

Table 9-7: Typical design temperature rises for medium voltage power and distributiontransformers (over A 30˚C average/ 40˚C maximum ambient)

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, integralfuses for transformers with padmount-style enclosures, etc.

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

a.)

b.)

d.)

c.)

kV class VPI/VPE dry-type BIL (kV) Liquid-filled and cast-resindry-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

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

* VPI/VPE dry-type transformers are typically not available above 25.0 kV Class

Page 132: @Electrical

9

Medium voltage metal-enclosed interrupter switchgearTable 9-8: Quick referencee: Medium voltage metal-enclosed switchgear

Metal-enclosed power switchgear is defined by [8] as “a switchgear assembly enclosed on all sides and top withsheet metal (except for ventilating openings and inspection windows) containing primary power circuit switching orinterrupting 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 interrupterswitchgear is defined by [8] as “metal-enclosed power switchgear including the following equipment as required:

� 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 asservice-entrance equipment in small- to medium- size facilities. Figure 9-7 shows an example of metal-enclosedinterrupter switchgear.

As with all fusible equipment, overcurrent protection flexibility is limited, however with current-limiting fuses thisequipment has high (up to 65 kA rms symmetrical) short-circuit interrupting capability. The load interrupterswitches in this class of switchgear are designed to interrupt load currents only, and may use air as theinterrupting medium or SF6. They may be arranged in many configurations of mains, but ties, and feeders asrequired by the application.

This type of switchgear is frequently used as the primary equipment of a unit substation line-up incorporatingprimary 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 thistype of switchgear.

Available voltage ratings 2400 V - 38 kV

Available current ratings 600 - 2000 A

Available short-circuit ratings Through 65 kA

Major industry standards ANSI/IEEE C37.20.3

Typical enclosure types 1, 3R

Figure 9-7: Metal-enclosed interrupter switchgear

Page 133: @Electrical

10

Table 9-9: Voltage withstand levels for metal-enclosed interrupter switchgear, per [8]

Internal barriering requirements for medium voltage areas within the switchgear are minimal. All low voltagecomponents 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 toprevent closing their respective switch while they are accessible. In the rare case that this type of switchgearcontains drawout devices, shutters must be provided to prevent accidental contact with live parts when thedrawout element is withdrawn.

Available options for this type of switchgear include shunt trip devices for the switches, motor operators for theswitches, blown fuse indication, etc. Relaying of any type, including voltage relaying, must be carefully reviewed toavoid exceeding the limits of the switches. The application of overcurrent relaying to this type of switchgear is notrecommended unless a short-circuit interrupting element is included, such as a vacuum interrupter.

Medium voltage motor control centersTable 9-10: Quick referencee: Medium voltage motor conrol centers

Medium voltage motor controllers are used to control the starting and protection for medium voltage motors. Theygenerally utilize vacuum contactors rated up to 400 A continuous, in series with a non-load-break isolation switchand R-rated fuses, fed from a common power bus. The motor starting methods in Section 8 are all generallysupported, 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 aregenerally of roll-out design to allow quick replacement.

Above 7200, metal-clad switchgear is generally used for motor starting.

Rated Maximum Voltage (kV) 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

Available voltage ratings 2400 V – 7.2 kV

Available current ratings Through 3000 A

Available short-circuit ratings Through 50 kA

Major industry standards NEMA ICS-3, UL 347

Typical enclosure types 1, 3R

Figure 9-8: Medium voltage MCC

Page 134: @Electrical

11

Medium voltage metal-clad switchgearTable 9-11: Quick referencee: Medium voltage metal-clad switchgear

Metal-clad switchgear is defined by [10] as “metal-enclosed power switchgear characterized by the followingnecessary features:

� 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, voltagetransformers, and control power transformers, are completely enclosed by grounded metal barriers that have nointentional openings between compartments. Specifically included is a metal barrier in front of, or a part of, thecircuit interrupting device to ensure that, when in the connected position, no primary circuit components areexposed 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 barriersfrom all primary circuit elements with the exception of short lengths of wire such as at instrument transformerterminals.

� The door through which the circuit interrupting device is inserted into the housing may serve as an instrument orrelay 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- tolarge-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.

Available voltage ratings 2400 V – 38 kV

Available current ratings Through 3000 A

Available short-circuit ratings Through 50 kA

Major industry standards ANSI/IEEE C37.20.2

Typical enclosure types 1, 3R

Figure 9-9: Metal-clad switchgear

Page 135: @Electrical

12

Metal-clad switchgear uses high-voltage circuit breakers, as described in Section 7, fed from a common powerbus. 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 ofswitchgear is the preferred means for accomplishing automatic transfer control and complex generator parallelingapplications; the control may be placed in the switchgear itself or in a separate panel, depending upon theapplication and specific end-user preferences.

The construction requirements per [10] insure that metal-clad switchgear is the safest type of switchgear in termsof operator safety.

The BIL and withstand voltage requirements for this switchgear are the same as for metal-enclosed switchgear asgiven in table 9-9 above.

This type of switchgear has many options available to suit the application, such as electric racking for circuitbreakers, ground and test units that allow the grounding/testing of stationary contacts with a circuit breakerwithdrawn, etc.

References[1] Enclosures for Electrical Equipment (1000 Volts Maximum), NEMA Standards Publication 250-2003.

[2] The National Electrical Code, NFPA 70, The National Fire Protection Association, Inc., 2005 Edition.

[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 StandardsPublication ICS 3-1993.

[10] IEEE Standard for Metal-Clad Switchgear, IEEE Std. C37.20.2-1999, July 2000.

Page 136: @Electrical

1

Section 10:Emergency and Standby Power SystemsBill Brown, P.E., Square D Engineering Services

IntroductionEmergency and standby power systems are designed to provide an alternate source of power if the normal sourceof power, most often the serving utility, should fail. As such, reliability of these types of systems is critical and gooddesign practices are essential.

Code and standards

A.) Classification of emergency and standby power systemsThe 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 electricenergy that, upon failure or outage of the normal source, automatically provides reliable electric power within aspecified time to critical devices and equipment whose failure to operate satisfactorily would jeopardize the healthand 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 andclassed as emergency by municipal, state, federal, or other codes, or by any governmental agency havingjurisdiction. These systems are intended to automatically supply illumination, power, or both, to designated areasand equipment in the event of failure of the normal supply or in the event of accident to elements of a systemintended 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 oroutage of the normal source, provides electric power of acceptable quality so that the user’s facilities maycontinue 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 bymunicipal, state, federal, and other codes or by any governmental agency having jurisdiction. These systems areintended to automatically supply power to selected load (other than those classed as emergency systems) in theevent 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 thenormal 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 propertywhere life safety does not depend on the performance of the system. Optional standby systems are intended tosupply on-site generated power to selected loads either automatically or manually. FPN: Optional standby systemsare typically installed to provide an alternate source of electric power for such facilities as industrial andcommercial buildings, farms, and residences and to serve loads such as heating and refrigeration systems, dataprocessing 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 446-1995IEEE Standard 446-1995, IEEE Recommended Practice for Emergency and Standby Power Systems forIndustrial and Commercial Applications [1], is a general engineering reference for the design of these systems.

Page 137: @Electrical

2

C.) The National Electrical CodeThe National Electrical Code [2] contains requirements for emergency systems in Article 700, Legally-RequiredStandby 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 ElectricalPower 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 110NFPA 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 ofthe 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))

The Class of an emergency power system refers to the minimum time, in hours, for which the system is designedto operate at its rated load without being refueled or recharged. The Classes for emergency power systems areshown in table 10-2 [3]:

Table 10-2: NFPA 110 emergency power system classes (essentially the same as [3] table 4.1(B))

The Level of an emergency power system refers to the level of equipment installation, performance, andmaintenance requirements. The Levels for emergency power systems are shown in table 10-3 [3]:

Table 10-3: NFPA emergency power system levels

Type Power restoration time

U Basically Uninterruptible (UPS Systems)

10 10 sec

60 60 sec

120 120 sec

M Manual stationary or nonautomatic – no time limit

Class Power restoration time

0.083 0.083 hr. (5 min.)

0.25 0.25 hr. (15 min.)

2 2 hr.

6 6 hr.

48 48 hr.

X Other time, in hours, as required by the application, code, or user.

Level When Installed

1 When failure of the equipment to perform could result in loss of human life or serious injuries

2When 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

Page 138: @Electrical

3

E.) NFPA 101 NFPA 101 [4], Life Safety Code, addresses those construction, protection, and occupancy features necessary tominimize danger to life from fire, including smoke, fumes, or panic. It defines the requirements for what systemsthe Emergency Power System will supply.

F.) NFPA 99NFPA 99 defines establishes criteria to minimize the hazards of fire, explosion, and electricity in health carefacilities. It defines several specific features of electric power systems for these facilities.

Reasons for applicationEmergency and standby power systems are generally designed into the over-all electrical system for one of thefollowing two reasons:

� Legal Requirements – As required by the NEC [2] NFPA 101 [4], NFPA 99 [5], and other local, state, and federalcodes 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 electricalpower 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 strategyare not discussed in this section.

Power sourcesGenerators are by far the most prevalent source of power for emergency and standby power systems. For mostcommercial and industrial power systems these will be engine-generator sets, with the prime-mover and thegenerator built into a single unit. For reciprocating engines, diesel engines are the most popular choice of prime-mover for generators, due to the cost of the diesel engines as compared to other forms of power and the relativeease of application. Gasoline engine generator sets are also available and are generally less expensive thandiesel 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 formaximum 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-turbinegenerator sets are generally lighter in weight than diesel engine-generator sets, run more quietly, and generallyrequire less cooling and combustion air, leading to lower installation costs. However, gas-turbine generator setsare more expensive than diesel engine-generator sets, and require more starting time (normally around 30 scompared 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 orgasoline engines. Further, the fuel supply must be considered; building code and insurance considerations mayforce the fuel storage tank to be well removed from the generator(s), usually forcing the addition of a fuel transfertank near the generator(s).

Care must be taken when sizing engine-generator sets for a given application since several ratings exist for theoutput capability of a given machine. The continuous rating is typically the output rating of the engine-generatorset on a continuous basis with a non-varying load. The prime power rating is typically the continuous outputrating with varying load. The standby rating is typically the output rating for a limited period of time with varyingload. 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, theprocurement of a second utility source which is sufficiently independent from the normal service may beeconomically infeasible.

Page 139: @Electrical

4

Solid-state converters that invert DC voltage from a battery system are another alternative, although they can bedifficult to apply and generally are not available in the larger sizes that may be needed for a medium to largeemergency or standby system.

Because motor starting and block loading can have a big effect on the output voltage and frequency of a smallgenerator such as the engine-generator sets described above, and also because power is not available during theengine starting period, a buffer between the generators and sensitive load equipment is generally required. TheUninterruptible Power Supply (UPS) is usually the buffer of choice for these applications. UPS’s are available inseveral different topologies, but the operational goals are the same regardless of topology: The supply ofuninterrupted power to sensitive, critical loads. The most popular topology for a UPS is the double conversiontopology, as shown in figure 10-1:

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. Othertopologies exist, including the line interactive, double-conversion rotary, hybrid rotary, and line-interactiverotary topologies, each with advantages and disadvantages of application. UPS systems do not alleviate the needfor a generator or second utility service power source, but they do serve to buffer critical loads from the effects ofgenerator starting time and voltage and frequency variations.

Switching devicesA 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 conductorconnections from one power source to another” [1]. The automatic transfer switch is the most common means oftransferring critical loads to the emergency/standby power supply. An automatic transfer switch consists of aswitching means and a control system capable of sensing the normal supply voltage and switching over to thealternate 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 generallypreferred. An automatic transfer switch is generally an open-transition device that will not allow paralleling of thetwo sources. Manual versions of transfer switches are also available. A one-line representation of an automatictransfer switch is shown in figure 10-2.

Figure 10-1: Double-conversion UPS topology

Figure 10-2: Automatic transfer switch one-line diagram representation

Page 140: @Electrical

5

Other options for transferring devices include electrically-operated circuit breakers, as described in “Systemprotection” section (section7 in this guide). For medium voltage transfers, medium voltage circuit breakers aregenerally 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 otherswitching means) and connect the source directly to the load and allow isolation of the transfer switch formaintenance. Figure 10-3 shows a typical bypass/isolation switch arrangement along with the transfer switch:

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 withoutdisconnecting the load.

Static Transfer Switches are typically used when high-speed (~4ms) operation is required. The most commonapplication is to bypass a UPS so that a UPS failure will not result in interruption of service to the load.

System arrangementsVarious ways of arranging emergency and standby power systems exist. The most common arrangements are given here.

A.) Basic arrangement – radial systemThe most basic arrangement for an emergency or standby power system is shown in figure 10-4. This can berecognized as an extension of the single-source radial system from “System arrangements” section (section 5 ofthis 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 inherentweaknesses of the radial system to the emergency system, since a single failure of one piece of equipment canresult in loss of service to the emergency/standby loads. Note that the single generator shown may be severalengine-generator sets operating in parallel, if necessary. This simple system may be expanded to the othersystem types in Section 5.

Figure 10-3: Bypass/isolation switch application

Page 141: @Electrical

6

B.) More complex systemsThe 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 systemas shown in figure 10-5:

In figure 10-5, the emergency/standby load at the bottom of the figure will always be supplied by one of thenormal sources if possible, and by the generator(s) if not. This will avoid the generator starting time for this load ifone utility source were to fail. The two emergency/standby loads in the middle of the figure will be supplied bytheir respective switchboard busses or by the emergency source.

Emergency/standby systems are not limited to the low voltage level. For example, the primary selective/primaryloop/secondary selective system shown in figure 5-14 can be expanded to include an emergency system, asshown in figure 10-6:

Figure 10-4: Simple emergency/standby system arrangement

Figure 10-5: Example of a more complex emergency/standby system arrangement

Page 142: @Electrical

7

In figure 10-6 there is a great deal of flexibility in the system operation. However, instead of automatic transferswitches metal-clad switchgear is used increasing the complexity of the system.

C.) Hospital arrangementsNFPA 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 sourcesof power and all connected distribution systems and ancillary equipment, designed to ensure continuity ofelectrical power to designated areas and functions of a health care facility during disruption of disruption ofnormal power sources,” and the emergency system itself, which is “a system of circuits and equipment intended tosupply 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 10-6: Medium voltage emergency/standby system implementation

Figure 10-7: Minimum requirement per NEC [2] and NFPA 99 [5] for essential electrical system forhospitals 150 kVA or Less (same as [2] FPN figure 517.30 No.2)

Page 143: @Electrical

8

The essential electrical system supplies the equipment system, defined as “a system of circuits and equipmentarranged for delayed, automatic, or manual connection to the alternate power source and that serves primarily 3-phase 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 branchcircuits…intended to provide adequate power needs to insure safety to patients and personnel” [2]. Theemergency system also supplies the critical branch, which is “a subsystem of the emergency system consisting offeeders and branch circuits supplying energy to task illumination, special power circuits, and selected receptaclesserving 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 theequipment system above 150 kVA is required to be delayed.

NEC requirementsThe following are highlights from the NEC [2] requirements for emergency and standby power systems. This is notintended to list all NEC requirements for these systems, but to illustrate the major points that apply in the mostcommon installations and affect the power system design. For the full text of the complete NEC requirements forthese 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 requirementsapply to those systems meeting this definition:

� Witness Test: The authority having jurisdiction must conduct or witness a test of the complete system andperiodically afterward. [700.4 (A)]

� Emergency systems must be tested periodically on a schedule acceptable to the authority having jurisdiction toensure 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 mustbe provided. [700.4 (E)]

� The alternate power source is required to be sized to supply all emergency loads simultaneously. [700.5 (A)]

Figure 10-7: Minimum Requirement per NEC [2] and NFPA 99 [5] for Essential Electrical system forHospitals over 150 kVA (same as [2] FNP figure 517.30 No. 1)

Page 144: @Electrical

9

� The alternate power source is permitted to supply emergency, legally required standby, and optional standbysystem loads where the source has adequate capacity or where automatic selective load pickup or loadshedding is provided to insure adequate power to the emergency, legally required standby, and optional standbysystem loads. If these requirements are met the system may also be used for peak load shaving. Peak loadshaving operation may satisfy the requirement for periodic testing if acceptable to the authority havingjurisdiction. A portable or temporary alternate source must be available if the emergency generator is out ofservice 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, thatthe battery is carrying load, that the battery is not functioning, and to indicate a ground fault in solidly-groundedwye systems of more than 150 V to ground and over 1000 A. The sensor for ground-fault indication must belocated 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 emergencypower sources. A sign is also required where the grounded circuit conductor connected to the emergencysource 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 readilyidentified 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 loadsmust be kept entirely independent of all other wiring and equipment. Exceptions apply where load equipmentmust 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.) bea listed electrical circuit protective system with a 1-hour fire rating, or 3.) be protected by a listed thermal barriersystem for electrical system components, or 4.) be protected by a fire-rated assembly listed to achieve aminimum fire rating of 1 hour, or 5.) be embedded in not less than 50mm of concrete. Feeder circuit equipmentmust 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, emergencylighting, power, or both, must be available within the time required by the application but not to exceed 10seconds.

� The alternate source of power must be a storage battery, generator set, UPS, separate service, or fuel cellsystem, 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 onehours, without the voltage applied to the load falling below 87% of nominal. The battery charging means must beautomatic. [700.12 (A)]

� Generator sets must have a prime-mover acceptable to the authority having jurisdiction, and means ofautomatically starting the prime mover on failure of the normal service. If the prime-mover is an internalcombustion engine, an on-premises fuel supply must be provided to allow not less than 2 hours full-demandoperation of the system. If power is required for operation of fuel transfer pump to deliver fuel to a generator setday tank, this pump must be connected to the emergency power system. Generator sets must not be solelydependent on a public utility gas system for their fuel supply for a municipal water supply for their coolingsystems. If dual supplies for these are used, means must be provided to automatically transfer from one supplyto the other. If a storage battery is used for control or signal power or as the means of starting the prime mover, itmust be equipped with an automatic charging means independent of the generator set. Where power is requiredfor the operation of the dampers used to ventilate the generator set, the dampers must be connected to theemergency system. [700.12 (B) (1), (2), (3), and (4)]

Page 145: @Electrical

10

� If a generator set requires more than 10 seconds to develop power, an auxiliary power supply that energizes theemergency 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 conductorsserve or pass through the building or structure, so long as they are equipped with a readily-accessibledisconnecting means located within sight of the building or structure supplied. [700.12 (B)(6)]

� UPS’s used to provide power for emergency systems must comply with the applicable provisions for batterysystems and generators.

� An additional utility service is permitted to be the power source for the emergency system, if acceptable to theauthority having jurisdiction. A separate service drop or service lateral and service conductors sufficiently remoteelectrically and physically from other service conductors to minimize the possibility of simultaneous interruptionof 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 full-demand operation. Fuel cell systems must meet the requirements of Parts II through VIII of Article 692 (Fuel CellSystems). A single fuel cell that serves as the normal source for the building or group of buildings concernedcannot serve as the alternate source. [700.12(E)]

� Individual unit equipment for emergency illumination must have a rechargeable battery, a battery chargingmeans, provisions for one or more lamps mounted on the equipment or terminals for remote lamps, and arelaying 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 initialillumination level. [700.12 (F)]

� Individual unit equipment for emergency illumination must be fixed in place. Flexible cord-and-plug installation ispermitted if the cord is no more than 3ft. in length. The branch circuit feeding the unit equipment must be thesame as that serving normal lighting in the area and connected ahead of any local switches, and must be clearlyidentified at the distribution panel. Alternatively, for areas with at least three normal lighting branch circuits theemergency 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 otherlights specified as necessary to provide required illumination. Failure of any individual lighting element must notleave in total darkness any space that requires emergency illumination. If HID lighting is used as emergencyillumination, it must operate until normal illumination has been restored. [700.16]

� Emergency lighting must have either an emergency lighting supply, with provisions for automatically transferringthe emergency lights upon the event of failure of the general lighting system supply, or two or more separateand complete systems with independent power supplies, each providing sufficient current for emergency lightingpurposes. If two systems are used, means must be provided for automatically energizing either system uponfailure 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 supplysource 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 emergencylighting. 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 assemblyoccupancies 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 sufficientdaylight may be controlled by an automatic light-actuated device. [700.22]

Page 146: @Electrical

11

� 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 overcurrentprotective devices. [700.27]

B.) Legally required standby systems (Article 701)� 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 standbysystem and optional standby system loads, provided that it either has enough capacity to handle all connectedloads or that automatic selective load pickup and load shedding is provided that will ensure adequate power tothe legally required standby circuits. [701.6]

� Requirements for transfer equipment are essentially the same as for emergency systems, except that norestriction is placed upon the use of transfer equipment use for other systems in addition to the legally requiredstandby system. [701.7]

� Audible and visual signal devices must be provided for indication of derangement of the standby source, that thestandby 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, andcabinets 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, connectionahead 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, UPS’s, separate utility service, and fuels cells as thestandby power source are essentially the same as for emergency systems, except the requirements for fueltransfer 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 powersource. This connection ahead of the normal service must be sufficiently separated from the normal mainservice 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 sameas for emergency illumination individual unit equipment. [701.11 (G)]

� Legally-required standby system overcurrent protection requirements are essentially the same as for emergencysystems, except that ground-fault indication is not required. [701.15, 701.17, 701.18]

C.) Optional standby systems (Article 702)� 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 supplyconductors. [702.6]

� Audible and visual signal devices must be provided for indication of derangement of the standby source and toindicate that the optional standby source is carrying load. [702.7]

Page 147: @Electrical

12

� 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 withother general wiring. [702.9]

� Where a portable optional standby source is used as a separately derived system, it must be grounded to agrounding electrode in accordance with Article 250.30. Where a portable optional standby source is used as anon-seperately derived system, the equipment grounding conductor must be bonded to the system groundingelectrode. [702.10]

� Outdoor generator sets do not require an additional disconnecting means where the ungrounded conductorsserve or pass through the building or structure, so long as they are equipped with a readily-accessibledisconnecting means located within sight of the building or structure supplied. [702.11]

D.) Health care facility essential electrical systems (Article 517 part III)� 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 isinterrupted 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 asamended by Article 517. [517.26]

� Hospitals (Articles 517.30 – 517.35)

� The essential electrical systems for hospitals must comprised of two separate systems: The emergencysystem and the equipment system. The emergency system must be limited to circuits essential to lifesafety and to critical patient care, designated as the life safety branch and the critical branch. Theequipment system must supply major electrical equipment necessary for patient care and basic hospitaloperation. [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 maximumdemand 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 thetransfer will overload the equipment, and they must be automatically shed upon generating equipmentoverloading. [517.30 (B)(5)]

� The life safety and critical branch circuit wiring must be kept independent of all of other wiring andequipment 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 theequipment system is permitted to occupy the same raceways boxes, or cabinets of other circuits that arenot part of the emergency system. [517.30 (C)]

� All wiring of the emergency system must be mechanically protected. Nonflexible metal raceways, type MIcable, or Schedule 80 rigid nonmetallic conduit are permitted, except that nonmetallic raceways cannotbe used for branch circuits that supply patient care areas. Schedule 40 rigid nonmetallic conduit, flexiblenonmetallic or jacketed metallic raceways, or jacketed metallic cable assemblies listed for installation inconcrete may be used if encased in no less than 2 in. of concrete. Listed flexible metal raceways andlisted metal sheathed cable assemblies may be used under certain conditions. Flexible power cords ofappliances or other utilization equipment and secondary circuits of Class 2 or Class 3 communications orsignaling 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 operatingsimultaneously 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)]

Page 148: @Electrical

13

� The life safety branch is permitted to supply only illumination of egress means, exit signs, alarm andalerting systems, communications systems used during emergency conditions, task illumination at the generator set location, elevator cab lighting, control, communications, and signal systems, andautomatic 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 selectedpatient care areas, general care beds, selected labs, additional patient care task illumination andreceptacles as needed, nurse call systems, blood, bone, and tissue banks, telephone equipment roomsand 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 hoodsupply 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 heatingequipment, selected elevators, hyperbaric and hypobaric facilities, automatically operated doors, selectedelectrically-heated autoclaving equipment, and other selected equipment. [complete list given in NEC textand NFPA 99:4.2.2.2.3.5(9)] [517.34 (B)]

� Generator accessories, such as the transfer fuel pump, electrically operated louvers, and otheraccessories essential for generator operation, must be arranged for non-delayed automatic connection tothe alternate power source via the equipment system.

� A minimum of two sources of power are required, one normal, one alternate. The alternate source may begenerator(s) on the premises, an external utility service if the normal service is a generator(s) on thepremises, or a battery system.

� Nursing homes and limited care facilities (Article 517.40 – 517.44)

� Applicability depends upon the type of care given at the facility. Specific exceptions are listed for certaintypes of facilities (see NEC text for details). If a nursing home provides inpatient hospital care, it mustconform to the requirements for hospitals. Nursing homes and limited care facilities that are contiguous orlocated on the same site with a hospital are permitted to have their essential electrical systems suppliedby the hospital. [517.40]

� The essential electrical system must be comprised of two separate branches: The life safety branch andthe 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 notenter the same raceways, boxes, or cabinets with other wiring. Exceptions apply where transfer or loadequipment 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 secondsafter interruption of the normal source. [517.42]

� The life safety branch must supply only illumination of means of egress, exit signs, alarm and alertingsystems, 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 selectedreceptacles in selected patient care areas, sump pumps and other equipment required to operate for thesafety of major apparatus and associated control systems and alarms, smoke control and stairpressurization systems, kitchen hood supply and/or exhaust systems if required to operate during a fire inor under the hood, and supply, return, and exhaust ventilating systems for airborne infections isolationrooms (complete list given in NEC text). [517.43(A)]

Page 149: @Electrical

14

� Delayed automatic connection to the critical branch must be provided for heating equipment to provideheating 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 sourceis a generator(s) on the premises, in which case the alternate source may be either another generator setor external utility service. In certain cases a battery system may be used (see NEC text). [517.44 (B)]

� Other health care facilities (Article 517.45)

� 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[1] IEEE Recommended Practice for Emergency and Standby Power Systems for Industrial and Commercial

Applications, IEEE Std. 446-1995, December 1995.

[2] The National Electrical Code, NFPA 70, The National Fire Protection Association, Inc., 2005 Edition.

[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.

Page 150: @Electrical

1

Section 11: Power Quality ConsiderationsBill Brown, P.E., Square D Engineering Services

IntroductionThe term power quality may take on any one of several definitions. The strict definition of power quality is “theconcept of powering and grounding electronic equipment in a manner that is suitable to the operation of thatequipment 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 format the nominal voltage level. Deviation from this sinusoidal norm therefore denotes a power quality issue. Strictlyspeaking, this deviation is actually a power disturbance, defined as “any deviation from the nominal value (orfrom some selected thresholds based upon tolerance) of the AC input power characteristics” [1]. The mostcommon 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. Typicalvalues 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 atime period between .5 cycles and 3 seconds). A temporary interruption is a voltage loss (<10% of nominal) for atime period between 3 seconds and 1 min. A sustained interruption is the complete loss of voltage for a timeperiod 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 ofthe 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 thewaveform. 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 visualobservation 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 controlsystems in which they occur. Noise may be further categorized as transverse-mode noise, which is measurablebetween phase conductors but not phase-to-ground, and common-mode noise, which is measurable phase-to-ground 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:

Page 151: @Electrical

2

Table 11-1: Common power disturbance causes

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 causemicroprocessor-based equipment such as computers to lose data. In extreme conditions, such as for voltagesurges caused by direct lightning strikes, both power equipment and utilization equipment may be subject tofailure. With the high reliability requirements imposed upon power systems, it is imperative that power systemdisturbances, or potential disturbances, be mitigated to avoid down-time, equipment failure, and risk to human life.

Power quality metricsThere are various methods for categorizing the severity of power disturbances. The most typical indices formeasuring power quality disturbances are:

Distortion Factor: The ratio of the root square value of the harmonic content to the root square value of thefundamental quantity, expressed as a percentage of the fundamental, also known as total harmonic distortion [1].

(11-1)

where

Vh is the RMS harmonic voltage (or current) value at a frequency of n times the fundamental frequency

V1 is the RMS fundamental-frequency voltage or current

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 non-sinusoidal 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

Page 152: @Electrical

3

Alternate forms for the distortion factor are given in [2] as percentages of the nominal voltage or demand loadcurrent for the system under consideration, for use in evaluation of the harmonic content of the system voltage orcurrent. These are referred to as Total Harmonic Distortion (THDVn) and Total Demand Distortion (TDD), definedas follows:

(11-2)

(11-3)

where

Vh is the RMS value of the nth harmonic component of the voltage

Vn is the RMS nominal fundamental voltage value

Ih is the RMS value of the nth harmonic component of the current

IL is the maximum demand load current, typically the average maximum monthly demand over a 12-month period

Crest Factor: The ratio of the peak value of a periodic function to the RMS value, i.e.:

(11-4)

where

ypeak is the peak value of a periodic function

yrms is the RMS value of the function

Because power system voltages and currents are nominally sinusoidal, the nominal crest factor for these wouldbe √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]:

The notch area for the notch as illustrated in figure 11-1 is defined as:

(11-5)

where

An is the notch area in volt-microseconds

t is the notch time duration in microseconds

d is the notch depth in volts

Figure 11-1: Voltage (or current) notch illustration

Page 153: @Electrical

4

Recovery time: This is the time needed for the output voltage or current to return to a value within the regulationspecification after a step load or line change.

Displacement Power Factor: The ratio of the active power of the fundamental wave, in watts, to the apparentpower 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 theeffects of harmonics.

K Factor: A measure of a transformer’s ability to serve non-sinusoidal loads. The K factor is defined as:

(11-6)

where

Ih is the harmonic component at h times the fundamental frequency

h is the harmonic order of Ih in multiples of the fundamental frequency

hmax is maximum harmonic order present

Voltage surgesThe 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 themost common cause of switching surges, while lightning is the most common cause of environmentally-inducedvoltage surges. Both can cause severe damage to unprotected power system components, with the potential forlightning 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 voltagesystems 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 interruptionsVoltage sags, swells and interruptions have many causes. Remote switching or lightning strikes can cause voltageswells, as can the recovery of the system voltage after a fault. Voltage sags can be caused due to transformer ormotor inrush or large step loads, especially on systems without large amounts of available fault current. Voltageinterruptions are generally caused due to protective device operation.

Protection of sensitive equipment against voltage sags and swells can be difficult. Fast-acting voltage regulatorsoffer the one means of defense against these phenomena, although any voltage regulator must be properlyapplied to avoid worsening the problem. Fast-acting voltage regulators can generally be classified as tap-switching, buck-boost, or ferroresonant (also known as CVT “constant voltage transformer”) types [1]. New solid-state tap switching technologies for voltage regulators provide faster response than older, electromechanicalswitching technologies. Other devices, such as “power line conditioners” which combine some TVSS functionswith 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 supplyor UPS. This device is available in several different topologies and is crucial where microprocessor-based devicesare to be powered. UPSs are discussed in more detail in a later section of this guide.

Harmonic distortionHarmonic 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 employpower 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.

Page 154: @Electrical

5

IEEE Standard 519-1992 [2] gives recommended limits for current distortion due to consumer loads and voltagedistortion in the utility supply voltage. Both are referenced at the point on the utility system where multiplecustomers can be served, referred to as the Point of Common Coupling (PCC). The requirements from [2] forcurrent distortion limits on general distribution systems 120 V - 69 kV are given in table 11-2. Table 11-3 showsthe 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, andthat as the ratio Isc/IL increases the limits also increase. The reason for this is that the current distortion limits aredesigned to limit the voltage distortion at the PCC, and the voltage distortion for a given current distortion worsenswith 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)

Table 11-3: IEEE 519-1992 Harmonic voltage distortion limits (essentially same as [2] table 11-1)

Mitigation of harmonic distortion is generally accomplished by one of the following means:

� 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 oneharmonic frequency. They are generally tuned to a value below the harmonic frequency to be attenuated to avoida resonance condition at that frequency. These are custom-engineered solutions that must be designedspecifically for the circuit in question. Passive filters are also used for power factor correction. However, there is

Maximum harmonic current distortion in percent of IL

Individual harmonic order (Odd harmonics)

Isc/IL <11 11<h<17 17≤h<23 23≤h<35 35≤h 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 = maximum short-circuit current at PCC

IL = maximum demand load current (fundamental frequency component) at PCC

Bus voltage at PCC 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

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

Page 155: @Electrical

6

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 correctionconsiderations.

Phase multiplication operates on the principle that if m six-pulse rectifiers are shifted 60/m degrees from eachother, are controlled by the same delay angle, and are loaded equally, the only harmonics present will be:

(11-7)

where

h is a harmonic order present

q = 6m and is known as the pulse number of the circuit

k is any integer

Thus, for standard 6-pulse rectifiers the harmonic orders present will be 5, 7, 11, 13,…, etc. 18-pulse rectifiers arethe 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 ofphase-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 theharmonic orders that do not fit equation (11-7) for the q of the predominate non-linear load are multiplied by afactor of 0.25. Phase-shifting transformer connections are used to achieve the 60/m degree phase shift between6-pulse rectifier units.

Active filtering technology is a still-evolving art. Current state-of-the-art designs measure the current, filter out thefundamental frequency of the measured current, and inject current that is the negative of the result into thesystem to cancel the harmonics up to a given harmonic order. These systems are generally used in existinginstallations that have existing 6-pulse drives where replacing the drives is not a cost-effective solution, or wheremultiple smaller 6-pulse drives are utilized since phase multiplication for a drive below 100hp is generally not cost-effective. State-of-the-art units can also dynamically correct the power factor, and are advantageous vs. passivefilters both in their effectiveness and their flexibility in power factor correction.

Power quality monitoringPower quality monitoring is vital when sensitive equipment is to be powered, and also for the over-all reliability ofthe system. Microprocessor-based technology allows the most common power-quality instrumentation to becombined into a single monitoring device which incorporates wave-form capture, measures of the power qualitymetric values per the above discussion, and conventional current, voltage, power, and energy measurements, withmin/max logging capabilities. These devices are typically true RMS-reading instruments, with measurements up toa given harmonic (typically 31st harmonic or higher).

The inclusion of power monitoring equipment in the initial power system design will make diagnosis of anysubsequent power quality issues, should they arise, much easier and more efficient. Reference [3] contains muchinformation 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.

Page 156: @Electrical

1

Section 12: Arc Flash Hazard ConsiderationsBill Brown, P.E., Square D Engineering Services

IntroductionThe consideration of arc flash hazards is a relatively new concern for power system design. However, it is aconcern that is rapidly gaining momentum due to increasingly strict worker safety standards and system reliabilityrequirements that demand work on live electrical equipment.

BackgroundElectrical arcs form when a medium that is normally an insulator, such as air, is subjected to an electric fieldstrong enough to cause it to become ionized. This ionization causes the medium to become a conductor whichcan carry current. The phenomenon of electrical arcing is as old as the world itself. Lightning is a natural form ofelectrical arc. Man-made electrical arcs exist in devices such as arc furnaces. However, utilization of electricalenergy 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:

� Heat from arc can cause severe flash burns many feet away (temperatures can reach 20,000 K, four times thetemperature 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 andcause 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 amountof damage to the equipment shown.

Electrical safety has traditionally been concerned only with electric shock hazards. The recognition of arc flashhazards 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 energyincident 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.

At the time of publication, there are two basic standards which establish requirements for arc flash hazards. Thefirst is NFPA 70E, Standard for Electrical Safety in the Workplace [1], which defines the basic practices to befollowed for electrical safety, including protective clothing levels which must be worn for given levels of arc flashincident energy and what steps must be taken prior to live work on electrical equipment. The second is the IEEEGuide for Performing Arc-Flash Hazard Calculations, IEEE 1584-2002 [2] which gives the engineer the methodsfor calculating the severity of arc flash incident energy levels. The NEC [3] requires only that certain equipment

Figure 12-1: Example of arcing damage to equipment

Page 157: @Electrical

2

(switchboards, panelboards, industrial control panels, meter socket enclosures, and motor control centers in otherthan dwelling occupancies and likely to require examination, adjustment, servicing, or maintenance while live) befield marked to warn qualified persons of potential electric arc flash hazards.

NFPA 70E requirements for flash hazardsNFPA 70E [1] is divided into four chapters: Safety Related Work Practices (Chapter 1), Safety RelatedMaintenance Requirements (Chapter 2), Safety Requirements for Special Equipment (Chapter 3), and InstallationSafety 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, generatedduring 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 worker’s potential exposure to arc-flash energy, conducted forthe 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 safedistance 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 orapproach to live parts.

Flash Protection Boundary: An approach limit at a distance from exposed live parts within which a person couldreceive 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 shockhazard exists.

Restricted Approach Boundary: An approach limit at a distance from an exposed live part within which there isan increased risk of shock, due to electrical arc over combined with inadvertent movement, for personnel workingin close proximity to the live part.

Prohibited Approach Boundary: An approach limit at a distance from an exposed live part within which work isconsidered 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 electricalequipment 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 hasbeen disconnected from energized parts, locked/tagged in accordance with established standards, tested toensure the absence of voltage, and grounded if determined necessary.

NFPA 70E [1] chapter 1 covers personnel responsibilities (both the employer and the worker have specificresponsibilities for safety), training requirements, the establishment of an electrical safety program, and theestablishment of an electrically safe working condition. These will not be discussed in detail here, but the reader isstrongly 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 basicrequirement is that live parts over 50 V to ground to which an employee might be exposed should be put into an

Page 158: @Electrical

3

electrically safe work condition prior to working on or near them, unless the employer can demonstrate that de-energizing introduces additional or increased hazards or is infeasible due to equipment design or operationallimitations. In this case live work requires an Energized Electrical Work Permit, for which the requirements aregiven in Article 130.1 (A) (2). Some exemptions are given to the requirement for an electrical work permit, such astesting, 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 ofboundaries from an exposed, energized electrical conductor(s) or circuit part(s). The requirements for crossingthese 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 70Etable 130.2 (C) [1]. Qualified persons can approach live parts 50V or higher up to the restricted approachboundary, and can only cross this boundary if they are insulated or guarded and no uninsulated part of the bodycrosses the prohibited approach boundary, if the person is insulated from any other conductive object, or if the livepart is insulated from the person and from any other conductive objects at a different potential. Unqualifiedpersons 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 arcflash. This analysis must set the flash protection boundary, which for voltages below 600 V is equal to 4 ft. basedupon a clearing time of 0.1 second and a bolted fault current of 50 kA (5000 Ampere-seconds) or, where theclearing time x bolted fault is greater than 5000 ampere seconds or under engineering supervision, may becalculated with the equations given in the NFPA 70E text. For voltages over 600V, the flash protection boundary isdefined as the distance from the potential arc which has an incident energy of 1.2 cal/cm2, or 1.5 cal/cm2 if theclearing time is 0.1 second or faster. The means of calculating the arc flash protection boundary for voltages 600Vor less is based upon the theoretical “Lee” method developed in [5]. The method for calculating the arc flashincident energy for a given working distance from live parts is not specified in NFPA 70E code text itself; severalmethods are given in Annex D of NFPA 70E. The preferred methods for performing these calculations are given inIEEE 1584 [2], as detailed below. The option is also given to use pre-prepared tables given in NFPA 70E basedupon given levels of fault current and protective device clearing time to select personal protective equipment inlieu of a formal arc flash study.

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, whichmust 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 thereis 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 orflashes 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 energylevels above 1.2 cal/cm2; an exception allows Category 0 clothing to be worn for exposures 2 cal/cm2 or lower. Anexample of a full flash suit is shown in figure 12-3.

Figure 12-2: Approach boundaries, from [1]

Page 159: @Electrical

4

Table 12-1: Protective clothing characteristics (essentially the same as [1] table 130.7 (c) (11))

IEEE 1584IEEE 1584 [2] is the guide for determining arc flash incident energy levels and protection boundaries. It containsan 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 wherethe 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 spatterand pressure-wave effects.

A.) IEEE 1584 empirical methodThis 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

Hazard/risk category Clothing description Required minimum arc rating of PPE(cal/cm2)

0 Non-melting, flammable materials (i.e.,untreated cotton, wool, rayon, or silk, orblends of these materials) with a fabric weightof at least 4.5 oz/yd2

N/A

1 FR shirt and FR pants or FR coverall 4

2 Cotton underwear – conventional short sleeveand brief/shorts, plus FR shirt and FR pants

8

3 Cotton underwear plus FR shirt and FR pantsplus FR coverall, or cotton underwear plus twoFR coveralls

25

4 Cotton underwear plus FR shirt and FR pantsplus multilayer flash suit

40

Figure 12-3: Example of a full flash suit

Page 160: @Electrical

5

� 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 willescalate into a phase-to-phase fault

The first step in this method is to determine the predicted arcing fault current using the following equation forsystem 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 is the arcing fault current in kA

K = -0.153 for open configurations and -0.097 for box configurations

Ibf is the bolted fault current for three-phase faults in kA

V is the system voltage in kV

G 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 thefault. Care must be taken to identify which device actually clears the fault. The clearing time then becomes thearcing 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 is the normalized incident energy

K1 = -0.792 for open configurations and -0.555 for box configurations

K2 = 0 for ungrounded and high-resistance grounded systems and -0.113 for grounded systems.

G 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 = 1.0 for voltages above 1 kV, and 1.5 for voltages below 1 kV

En is the incident energy in cal/cm2

t is the arcing time per above

D is the working distance in mm

x is a distance exponent from [2] table 4

Page 161: @Electrical

6

Table 4 in [2] gives the distance exponents, along with typical gaps between conductors, for different voltagelevels and equipment types.

The flash protection boundary may be found using the following equation [2]:

(12-5)

where

DB is the boundary distance in mm

EB 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, inwhich case it may be increased to 1.5 cal/cm2. However, equation (12-5) may be used to calculate the boundaryfor any incident energy level, for example, to calculate the boundaries where different categories of PPE per table11-1 may be worn. Note that the larger of the boundaries as calculated from IEEE 1584 or NFPA 70E should beused 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 devicetime-current characteristic and the arcing current level. Because overcurrent protective device tripping times arelower for larger currents due to inverse time-current characteristics, this is an important point. Larger bolted faultcurrents lead to larger predicted arcing fault currents, which lead to generally lower values of arc flash incidentenergy. Lower bolted fault currents lead smaller predicted arcing fault currents, which lead to generally highervalues 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 overcurrentprotective device time-current characteristic is horizontal, such as for the instantaneous characteristic of anelectronic-trip circuit breaker, the two values will be equal since the arcing time will not change.

B.) “Lee” methodWhere the IEEE 1584 empirical method cannot be used due to being outside the limits of applicability as definedabove, the theoretically-derived “Lee” method per [4] may be used. This is based upon maximum power transferand is very conservative above 15 kV. To calculate the incident energy with this method, the following equationsare used [2]:

(12-6)

(12-7)

C.) Simplified device equationsFurther 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 faultcurrent ranges given in [2]. The equations for circuit breakers will yield conservative results and should only beused when they are within the ranges of applicability given in [2] and where nothing else about a particular circuitbreaker 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 moreaccurately model the arc-flash performance of a given device.

Page 162: @Electrical

7

Application guidelines

A.) Arc flash calculationsThe following guidelines are helpful when performing arc flash calculations [3]:

� When choosing a calculation method, be sure the system conditions fall into the calculation method’s range of applicability.

� Use the newest methods given in IEEE 1584-2002. Older methods given in previously-published papers aresuperseded 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 valuestypically 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 determiningwhich 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 forswitchboards, MCC’s, etc.

� Quantify the variables. The working distance, bus gap, equipment configuration, and system grounding are alldependent 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 flashprotection boundaries where multiple sources, such as generation and motor contribution, must be taken intoaccount. 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 designArc flash hazard analysis is typically performed after the system design process, including the time-currentcoordination 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:

Figure 12-4: Example electrical panel

Page 163: @Electrical

8

� Use a dedicated main overcurrent device at transformer secondaries. The secondary of a transformer is one ofthe most difficult places to achieve acceptable arc flash hazard levels. If multiple mains are used for transformersecondaries, the arc flash hazard level downstream from the main but ahead of the feeders must be calculatedusing the transformer primary device timing characteristics, significantly increasing the incident energy. If thesecondary main and feeders are in the same switchboard or panel, this will usually not be applicable due to arcfault propagation to the line side of the main device as described above. For ANSI low voltage switchgear perANSI C37.20.1, however, this can be of real benefit, as well as in cases where the secondary overcurrent deviceis remote from the feeders.

� Closely coordinate devices where possible. The lower the clearing time for the predicted arcing current, thelower the arc flash incident energy.

� Use high-performance devices, such as low-arc-flash circuit breakers, where possible. These will significantlyreduce the arc flash incident energy.

� Use bus differential protection and/or zone selective interlocking where possible. This is high-speed protectionthat 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] The National Electrical Code, NFPA 70, The National Fire Protection Association, Inc., 2005 Edition.

[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.

Page 164: @Electrical

1

Section 13: Utility Interface ConsiderationsBill Brown, P.E., Square D Engineering Services

IntroductionThe vast majority of industrial and commercial facilities are served from public utilities. However, the utilityinterface 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 thesystem. This can result in unexpected costs and project delays. These issues should be addressed during thesystem design stage, where the impacts to system reliability and cost can be adequately managed; only byknowing the utility’s requirements is this possible.

The utility’s jurisdictionBecause utilities must serve multiple consumers, they must take the steps they consider necessary to ensurereliable service over their entire system. Because of this, most utilities impose requirements on the design of thesystems to which they supply power.

Those elements of the system design over which the utility has jurisdiction vary from utility to utility. The utilityalways dictates which service voltages are available for a given size of service. The utility usually has somejurisdiction 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 somecases the utility will require jurisdiction over the entire service equipment, and can impose requirements uponsystem protection, equipment control power, and other parts of the system design. In some cases, the over-allarrangement 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 isusually not feasible. Therefore, knowledge of the utility’s requirements is vital to successful, on-time, on-budgetsystem design and construction.

Utility service requirements standardsEach 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-titledpublication. The format of the standards, and the standards themselves, vary from utility to utility. This can bechallenging 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 formedin 1983, combining southern-California-based PUSERC and northern-California-based WUESSC, which wereolder organizations formed in 1947 and 1950, respectively. The purposes of EUSERC are to promote uniformelectric service requirements among its member utilities, to publish existing utility service requirements for electricservice equipment, and to provide direction for development of future metering technology. EUSERC publishes amanual [1] which delineates requirements for electric service equipment through 34.5 kV. At the time ofpublication, 80 utilities from 12 states are involved with EUSERC. While EUSERC does not eliminate the needfor individual utility requirements, it does help a great deal in making electrical service equipment morestandardized and less costly.

System topology snf protectionRequirements for the system topology are designed to increase both the reliability of the over-all utility system andwith the reliability of service to the installation in question. These requirements typically take the following forms:

� 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

Page 165: @Electrical

2

� 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 utilitysubstation 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 rateagreement for the service in question.

Requirements for restricted access to service cable termination and service disconnect compartments in theservice switchgear are another common. In some cases these must be in a dedicated switchgear or switchboardsection, increasing the service equipment footprint. In many cases grounding means must be provided with theequipment to allow the utility’s preferred safety grounding equipment to be installed. In some cases, requirementsmay be imposed on the entire service switchgear, such as electrical racking for circuit breakers or barriers that arenot 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 utility’s specifications.

Additional protective relaying may be required to prevent abnormal conditions which, although not harmful to thesystem being served, affect the reliability of the utility system. In some cases the makes and models of protectiverelays for the service overcurrent protection are restricted to those the utility has approved.

Revenue metering requirementsOften the utility’s revenue metering requirements can have an effect the over-all system topology. There are twobasic 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 polefor overhead services, or in a dedicated utility-supplied metering compartment outside the facility to be meteredfor underground services. In these cases, the effect of the utility’s instrument transformers and meters on theover-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 theinstrument transformers. The design requirements for these compartments are often detailed, and are present toinsure that no tampering occurs with the instrument transformers or meters. These compartments typically take anentire section, or part of a section, of the service switchgear or a switchboard, increasing the footprint of thisequipment. In some cases, the service equipment must provide housing for the meters as well, along withconvenient access for the utility’s personnel. The utility typically provides and installs the instrument transformersand meters, although a few utilities require the end-user or equipment manufacturer to install these. In extremecases the end-user must supply the instrument transformers and send them to the utility for testing. Identifying therequirements 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 currenttransformers 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 aredesigned for metering, with the current transformers typically being bar or wound-primary type. For services over600 V, both voltage and current transformers are required, either two or three of each depending upon the systemconfiguration. In some cases the utility will not allow the voltage transformers to be fused.

Page 166: @Electrical

3

Additional regulatory requirementsIn some cases there may be additional state regulatory requirements which apply. These are typically concerndistributed generation and may severely restrict or otherwise impact the system design. These requirements mustbe fully understood before the system design is begun to avoid expensive changes later in the process. ThePublic Service Commission or similar governmental regulatory agency for the region in question typically controlsthese requirements.

Utility information required for system designIn designing the power system for any commercial or industrial facility the following information is crucial toadequate system design:

� Nominal service voltage.

� Maximum available fault current and associated X/R ratio.

� Minimum available fault current.

� Data on the utility’s nearest upstream protective device (device type and ratings, relay type and settings if applicable).

� Latest edition of the utility’s service handbook or similar publication.

� Latest edition of additional state regulatory requirements, if applicable.

� Contact information for utility’s 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 theregional Public Service Commission or similar governmental regulatory agency. Item 8 may not be available at theoutset, but should be taken into consideration as soon as it becomes available.

References[1] EUSERC Manual, Electric Utility Service Equipment Requirements Committee, 2005 Edition

Page 167: @Electrical

1

Section 14: Electrical Energy ManagementBill Brown, P.E., Square D Engineering Services

IntroductionElectricity 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 facility’s cost of production.

This energy management primer is intended to introduce some electricity billing fundamentals, especially focusingon the two major aspects of the electric bill, demand and energy. This section also highlights key aspects ofidentifying energy-saving opportunities among major industrial processes and equipment.

Electricity billing basicsMost 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 utility’s cost structure, determine pricescustomers 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 iscommon for the electric utility to allow a facility to choose the rate schedule within its class that offers the lowest price.

� Electricity metering: Electric utilities meter both the real and reactive power consumption of a facility. The realpower consumption, and its integral – energy, usually comprise the largest portion of the electric bill. Reactivepower 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 overthe course of a day as facility loads change. While instantaneous power fluctuations can be significant, electricutilities have found that average power consumption over a time interval of 15, 30, or 60 minutes is a betterindicator of the “demand” on electrical distribution equipment.

Transformers, for example, can be selected based on average power requirements of the load. Short-durationfluctuations in load current may cause corresponding drops in load voltage, but these drops are within thenormal 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 tothe necessary construction of new generating stations, transmission lines, and other utility capital projects.Demand charges often represent 40% or more of an industrial customer’s monthly bill.

Page 168: @Electrical

2

� Energy: The other major component of an electric bill is energy. The same metering equipment that measurespower demand also records customer energy consumption. Energy consumption is reported in kilowatt-hours ormegawatt-hours. Unlike power demand with its capacity relationship, customer energy consumption issometimes 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 powergeneration is based.

� Load factor – Demand/energy relationship: One useful parameter to calculate each month is the ratio ofthe average demand to the peak demand. This unit-less number is a useful parameter that tracks theeffectiveness 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 mostfacilities. But a facility can calculate its historical load factor, and seek to improve it by reducing usage at peaktimes, moving batch processes to times of lower demand, and so forth. Load factor can be calculated fromvalues 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 setduring the billing period, and days is the number of billing days in the month (typically 28-32). “24,” of course isthe 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 on-peak 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 threeshift, 24/7 facilities may reach load factors of 70-90%.

“Demand” is the average instantaneous power consumption over a set time interval, usually 15, 30, or 60 minutes.

TIME

DEMAND

Area under curve = ENERGY (KWh)Area under line = ENERGY (KWh)

DEMAND

(For the interval defined)

Demand for each interval = Average Power over that interval

DEMANDINTERVAL

DEMANDINTERVAL

DEMANDINTERVAL

DEMANDINTERVAL

PO

WE

R (K

ILO

WA

TTS

)

INSTANTANEOUSPOWER

Page 169: @Electrical

3

� Power factor: The relationship of real, reactive, and total power has been introduced previously, and describedas the “power triangle.” For effective electricity cost reduction, it is important to understand how the customer’selectric utility recoups its costs associated with reactive power requirements of its system. Many utilities includepower factor billing provisions in rate schedules, either directly in the form of penalties, or indirectly in the form ofreal-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 acustomer may find it economical to install equipment to improve power factor within its facility, thereby reducingthe 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.� Reduce power factor penalties

� Release capacity of an existing circuit

� Reduce heating losses associated with power distribution (often called I2R losses)

� Improve voltage regulation

Graphical comparison of facilities with dramatically different load factors. The three shift facility pro-duces an average demand that is nearly equal to its peak demand, while the average and peakdemand for the one shift facility is much less than one.

Dem

and,

kW

Dem

and,

kW

Equal EnergyUnequal Demand

T hree S hifts One S hift

Load Factor: 30% 50% 70%

Peak Demand, kW 1142 685 489

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 AsPercent of Total

53% 41% 33%

Page 170: @Electrical

4

� Typical energy auditing process:� Evaluate the current rate schedule

� 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 – Where’s the energy going?An important initial step in evaluating energy saving opportunities is to estimate both:

� 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 havethe most savings potential. This Profile also may identify batch processes or discretionary loads that may bescheduled at times of low demand for the rest of the facility, or during times of off-peak utility prices.

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 building’scentrifugal water chillers.

The Facility Energy Profile identifies the major energy consuming processes and equipment in the facility.

Circuit Monitors

Actual power monitoring data from existing circuit monitors measuring the power consumptionof individual feeders is the best basis for establishing the Facility Energy Profile.

Production Equipment

9%

Packaging Lines8%

HVAC10%

Cooling Tower Fans3%

Chilled Water Pumps

9%Condenser Water

Pumps3%

Chillers33%

Miscellaneous6%

Utility Systems3%

Compressed Air8%

Lighting Systems8%

Page 171: @Electrical

5

Demand analysis techniquesDemand analysis is the methodology used to determine if there are opportunities for a given facility to reducepeak demand charges. Demand analysis involves manipulation of historical demand interval data to determinewhich major processes or loads are operating at times of highest demand; how “steep” or “flat” the facility’s loadprofile appears; and what times of day these peaks are occurring. Armed with this information, the energy auditorcan better evaluate the potential for a variety of demand reduction techniques.

The demand sort is produced by rearranging individual integrated demand readings for a givenbilling period. Meters record demand readings chronologically, 3000 or so readings for a 30-daybilling 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 plant’s peak electrical demand exceeded certain levels.

Using the demand sort table, the engineer is able to determine that a reduction in peak demandto 2200 kW at this example facility would have required a demand reduction of 122 kW for 2515-minute intervals, or 6.25 hours, in August of the sample year.

Page 172: @Electrical

6

Demand controlDemand controls systems are available that perform these basic functions:

� 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 toexceed target kW

These demand controls systems are intended to reduce peak demand for a facility to some predetermined level.

The design engineer’s 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 facility’s 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 andprocesses 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 withmultiple 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 humiditycontrol within the conditioned space, or lack of demand savings.

The reason for this paradox is twofold. One, natural diversity among multiple air conditioning compressorsensures that all compressors are not operating at full load at the same time. Strangely, this fact is oftenhighlighted in the demand control system sales pitch: “Not all compressors are running at the same time, so youshould turn some off for short periods of time.”

Secondly, basic thermodynamic principles of moist air and vapor-compression refrigeration systems requirecompressor power consumption to reduce air temperature and condense moisture. This process is controlled bythermostats and humidistats within the facility. When cooling or dehumidification is removed or reduced at timeswhen 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 thedemand profile for the overall facility. A battery-charging system for forklift vehicles in an automotive facility was

8000

9000

10000

11000

12000

13000

14000

15000

16000

17000

15 100

145

230

315

400

445

530

615

700

745

830

915

1000

1045

1130

1215

1300

1345

1430

1515

1600

1645

1730

1815

1900

1945

2030

2115

2200

2245

2330

Interval Ending Time (Pacific Standard Time)

Dem

and,

kW

Shoulder-Peak

Off-Peak

On-Peak

Shoulder-Peak

Off-Peak

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.

Page 173: @Electrical

7

capable of producing real demand savings during peak times. Finally, a large induction furnace melting scrapmetal proved to be an effective candidate for the rolling mill at a steel plant.

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 costof 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 generatoroperation. That is, the onsite generator must be operated the absolute minimum time necessary to reduce peakdemand the maximum amount. Because the overall average unit price of electricity is not necessarily equivalentto the effective price of electricity at the plant’s peak.

For example, the facility that pays an overall average unit price of 8 ¢/kWh probably pays only about 3-4 ¢/kWhfor actual energy consumption, yet an additional $10-$20/kW for demand. At the end of the month, the total billingamount 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 facilitywith a sharp demand peak, when the peak for the month is set in a few hours or less and the remainder of thetime demand is low, peak-shaving at 12 ¢/kWh can be preferable to paying 20 ¢/kWh.

� Costs of generated power: Onsite generators typically utilize natural gas, wood, fuel oil, or steam derivedfrom a fossil fuel or as a part of a production process. Unit fuel costs for fossil fuels are usually calculated basedon the fuel’s 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 agenerator efficiency of 25%, the fuel cost/kWh is:

Cost/kWh = $2.00 * 3413/108,000 BTU/gal/0.25Cost/kWh = 25 ¢/kWh.

Obviously, peak-shaving is much less attractive at a fuel cost of $2.00/gal, unless required generator operationcan 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 toimplementing peak shaving or cogeneration opportunities. Some utilities have special interconnection andprotective relaying requirements to ensure that onsite generation does not pose a safety hazard for utilityworkers. In addition, many utility rate schedules impose standby charges for onsite generation.

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.

Page 174: @Electrical

8

These charges are intended to recoup the utility’s investment in transformers and other equipment necessary toserve the facility’s entire load when the onsite generation equipment is not operating. Without this standbyequipment, utilities often reserve the right to replace service equipment with smaller facilities, at risk to thefacility of overloading the smaller equipment when onsite generation is not operating.

5 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0

Plan

t Dem

and,

kW On-Peak

Period

Plant Total Power Requirement

Purchased Power

Facilities with onsite generation may be able to operate this equipment to reduce purchased powerrequirements during periods of high demand, or high utility prices.

-18%

-16%

-14%

-12%

-10%

-8%

-6%

-4%

-2%

0%

2%

4%

6000 5800 5600 5400 5200 5000 4800

Generator Setpoint, kW

$0.60/gal$0.80/gal$1.00/gal

Savings – or losses – associated with operation of peak-shaving generators is dependent on fuelprices, 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.

ProcessSteam

CondensateReturn

Turbines Electricity

Condenser

Boilers

Electricity generation and peak shaving can also be accomplished with steam cogeneration systems typical of paper mills, refineries, and other large industrial processes.

Page 175: @Electrical

9

Lighting controlLighting systems in industrial facilities can represent an attractive savings opportunity, especially if lightingsystems have not been upgraded or maintained in the past five years. The most cost-effective approach forlighting energy savings is to address the following three issues, in order:

� 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 ofNorth America (www.iesna.org) has established recommended light levels for different types of work tasks andarea usage types. In addition, it offers design guidance in laying out lighting systems, estimating light levels byzonal cavity and point-by-point lighting design methodologies.

These light level recommendations are typically described as ranges of footcandles, the footcandle being aquantity of light measured at a horizontal or vertical surface. Light output of a fixture is usually published inlumens. Many manufacturers of lamps and lighting systems offer software tools to aid in designing new systems,or in evaluating changes to existing systems.

� Some lighting essentials� Lighting controls work better than people

While “turn-off-the-light” programs have been widely utilized in all types of facilities, sophisticated lightingcontrol systems have proven to be much more cost-effective. Certainly, it’s cheaper to have a worker turn offa light, but workers forget, workers may not have access to circuit breakers controlling large banks ofindustrial 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 circuitbreakers are much more effective. These devices can be programmed to accommodate complicated shiftconfigurations, including nights, weekends, and holidays. They also include simple over-ride features fortemporary or unusual work schedules. In addition, these systems can be monitored and controlled remotelyusing standard web-browser software packages, and they can interface with other control devices such asmotion sensors or photocells.

� Light levels decline with age of the lighting system.Several factors contribute to this decline. Lamps, including fluorescent and high-intensity discharge sourceslike high-pressure sodium and metal halide, experience Lamp Lumen Depreciation, or LLD. The LLD istypically less than 1.0, indicating that average lamp light output at some point in the future is less than lightoutput of a new lamp.

Power Supply

Microprocessor

Remote-OperatedCircuit 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.

Page 176: @Electrical

10

Light levels are also adversely affect by dirt and the accumulation of dust on the light fixture. Luminaire DirtDepreciation, or LDD, also a factor less than 1.0, is a function of the type of light fixture as well as theenvironment 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 ofthe 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 toestimate the number of fixtures required to provide the desired light level – not at initial installation, rather atsome designated point in the future. For example,

# fixtures = total required lumens/initial lumens/fixture/(LLD * LDD * BF).

� Lighting designers need to know the facility’s lamp replacement practicesManufacturers publish the “rated life” expectancy of a given lamp. This value, usually given in thousands ofhours, is not a guarantee that every lamp will extinguish at the same rated-life time. In fact, the “rated life” isa 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 facility’s 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 ain’t broke, don’t fix it. That’s whyspot replacement is the most common practice by far. There is, however, a sound reason for considering thegroup-replacement strategy: Economics.

If the lighting designer knows, for example, that a facility will adopt the practice of group replacement, thedesigner can utilize fewer light fixtures at the outset. That’s because the lamps replaced before their end oflife produce considerably more lumens than those allowed to burn to failure. The designer can use a higherLLD 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 building’s airconditioning system. Labor costs have also been shown to be lower for group replacement as compared tospot replacement. Group replacement can be scheduled to occur during unoccupied times; set up and takedown costs are reduced; the cost per lamp itself can be lower with large-quantity purchases.

Electric motorsThree-phase squirrel-cage induction motors comprise a considerable percentage of the electrical load in theUnited States. Design, operation, and maintenance of these machines is well described in other references; thisdocument focuses on their energy efficiency aspects.

Induction motors typically range in full load efficiency from about 87% to 94%. This efficiency is very difficult tomeasure accurately in the field, requiring a dynamometer and other specialized equipment. Fortunately, energysaving 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 thanrewinding – 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 canbe costly. A standard efficiency 20 hp motor operating 8000 hours annually, for example, costs about $7000 per

Page 177: @Electrical

11

year to operate at an average electricity rate of 7 ¢/kWh. Once this motor fails, the least-cost option for returning itto service is typically rewinding.

The incremental cost of replacing this failed motor with an energy-efficient motor, however, is only $430. Thisamount 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 inductionmotors. Since the programs are based on replacement at failure, the full savings potential is realized after threeyears or more.

HP Rewound

EfficiencyStandard Efficiency

Energy Efficient

Efficiency

1 69.7% 70.7% 82.6%2 79.5% 80.5% 83.4%3 79.4% 80.4% 86.6%5 81.4% 82.4% 88.3%8 83.1% 84.1% 90.0%10 85.1% 86.1% 91.1%15 85.5% 86.5% 92.0%20 87.3% 88.3% 92.9%25 88.0% 89.0% 93.5%30 88.1% 89.1% 93.7%40 88.7% 89.7% 94.2%50 90.0% 91.0% 94.4%60 89.9% 90.9% 94.7%75 90.4% 91.4% 94.9%100 90.4% 91.4% 95.4%125 90.6% 91.6% 95.3%150 91.5% 92.5% 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.

Page 178: @Electrical

12

Variable-speed drivesThere 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-frequencycontrol devices designed to reduce energy consumption and improve operation of three-phase AC inductionmotors. See www.squared.com for technical publications that describe these devices in greater detail.

AC motor loads are typically grouped in four major categories:

Energy-saving opportunities commonly focus on the variable-torque category, because the energy saving potentialis 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 orfan 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 isdetermined by the number of motor poles, AC frequency, and motor slip, varying the speed of a motor requires anexternal device. This external device is commonly referred to as an adjustable-speed drive, variable-frequencydrive, inverter, vector drive, or adjustable-frequency controller.

Type of Load Typical Examples

Variable torque Centrifugal pumps and fans

Constant torque Reciprocating pumps, conveyors, hoists

Constant horsepower Grinders

Impact Punch press

Variable-torque loads, such as centrifugal pumps and fans, exhibit a cubic relationship between brake horsepower and speed.

Page 179: @Electrical

13

Compressed airCompressed 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 motorsrepresenting hundreds, thousands, or even ten-thousands of horsepower in capacity. One chemical plantproviding raw materials for synthetic textile manufacturing operated one 22,000 hp, and two 8,000 hpcompressors in a portion of its process.

While the 22,000 hp compressor is rare, significant energy reduction opportunities associated with compressedare available.

These opportunities may include:

� Reduce outlet pressure – compressor discharge pressure in some facilities is set too high. Since the pressurerise across a compressor is a key factor in its power consumption, reducing outlet pressure can offer significantsavings. Some reasons for excessive pressure may be straightforward; eg, production equipment with lowerrequirements has replaced older machines without a corresponding reduction in compressed air setpoint. Otherreasons may be more complex; eg, piping system losses or leaks may force higher setpoints at the compressorsin order to provide adequate air pressure at production equipment.

� Reduce air volume (CFM) requirements – compressed air leaks usually offer the most attractive opportunity forreducing compressed air volume. Some facilities have ignored leaks to the point that one compressor iseffectively 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 aboveindicates, reducing inlet air temperature can reduce the work associated with a compressor. The usual methodof reducing air temperature is to provide outside air intakes for the compressor, rather than allowing thecompressor to utilize air from a hot equipment room.

� Increase inlet pressure – It’s common to assume that inlet pressure to a compressor is fixed at atmosphericpressure, but this is a misconception. Air compressor inlet systems, especially air filters, need to be kept cleanand free of obstructions. Pressure drop across dirty or blocked intakes serves to reduce the pressure at thecompressor and increase power consumption.

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).

Multi-stage air compressors, equipped with inter- and after-coolers to optimize efficiency and provide heat recovery, are common in industrial facilities.

Page 180: @Electrical

14

Centrifugal water chillersCentrifugal 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 requiredenergy input. Centrifugal water systems were the focus of cholorfluorocarbon (CFC) legislation in the 1980 thatdrove the replacement or reconditioning of many of these machines. Opportunities still exist, however, for chilleroptimization.

One opportunity is to change the operational strategy of multiple chillers operating on a common chilled waterheader. 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 notrepresentative of plant conditions. Cooling load on a typical industrial water chiller systems is often influenced byprocess 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.

Other successful strategies for chiller optimization include:

� 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 automaticchiller 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 condenserwater temperature serves to reduce the compressor power requirements. Condenser water temperaturereduction of one degree can reduce compressor power consumption by 0.5%-1%.

0.5

0.550.6

0.650.7

0.750.8

0.850.9

0.95

20% 30% 40% 50% 60% 70% 80% 90% 100%

Part Load (%)

Dem

and

per

Ton

(kW

/ton)

Chiller Efficiency

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.

0.0%

20.0%40.0%60.0%

80.0%100.0%120.0%140.0%

3/6/

98 7

:58

3/8/

98 5

:31

3/10

/98

15:1

7

3/13

/98

9:57

3/16

/98

16:5

2

3/19

/98

23:1

9

3/21

/98

7:16

3/23

/98

5:01

3/25

/98

2:46

3/27

/98

0:31

3/28

/98

22:1

6

3/30

/98

20:0

1Perc

ent o

f Ful

l Loa

d kW

(%)

%loaded CHILLER-1_KWD

%loaded CHILLER-2_KWD

%loaded CHILLER-5_KWD

Operating three chillers at partial loads is less efficient that operating two chillers at or near their rated capacity.

Page 181: @Electrical

15

� Monitor and maintain chiller approach temperatures – chiller condensers and evaporators are shell-and-tubeheat exchangers that require periodic maintenance to maintain optimum heat transfer characteristics. Sincewater travels through the condenser and evaporator tubes, solids have a tendency to accumulate on internaltube surfaces, requiring annual “rodding” to remove the scale and restore heat transfer coefficients.

Heating, ventilating, and air conditioning systemsHVAC systems should be the focus of a targeted energy study, with similar objectives as the lighting analysis:

� 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

WAGESWAGES 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 theseparameters accurately, and of posting this information in a common, preferably web-based, format for use by thelocal site and by remote engineers and managers.

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

100%

105%

110%

115%

120%

125%

130%

135%

140%

145%

Clean 0.001 0.002 0.003 0.004

Condenser Fouling Factor

Rela

tive

HP p

er T

on

Source: Carrier System Design Manual, Carrier Air Conditioning Co, 1963.

137% Additional Power Required!

Est'd FF = 0.0034

Increase in condenser tube fouling can have a significant adverse effect on compressor power consumption.

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.

Page 182: @Electrical

16

Energy survey checklist

Lighting1. Lighting operating more hours than needed?

� Reduce operating hours with lighting control system.

2. Areas over lit for task performed?� Reduce light levels by disconnecting or replacing lamps or fixtures.

3. Incandescent or quartz lamps operating more than 2,000 hours per year?� Convert to fluorescent or other energy efficient source.

4. Mercury vapor lamps.� Convert to energy saving fluorescent, metal halide, or high-pressure sodium.

5. Standard fluorescent lamps operating one shift.� Convert to energy saving fluorescent lamps and ballasts.

6. Standard fluorescent lamps operating two or three shifts.� Convert to energy saving fluorescent lamps and ballasts.

7. Fluorescent at 18-feet or higher mounting heights.� Convert to high pressure sodium.

8. VHO fluorescent fixtures.� Convert to energy saving fluorescent, metal halide, or high pressure sodium.

9. Standard fluorescent ballasts.� Replace with energy savings electronic ballasts at failure.

Induction motors1. Motors operating 75%+ full load, more than 6,000 hours per year.

� Replace with energy efficient motors at failure.

2. Standard V-belts on pumps or fans.� Convert to cog V-belts.

3. Fans or pumps that are throttled with dampers or control valves.� Consider variable speed drives.

Demand management1. Sharp demand peaks of short duration (low load factor)?

� Identify loads to shed or reschedule to off-peak.

2. Batch processes? � Shift to off-peak.

3. Consider Time-of-Use savings opportunities.

Exhaust, ventilation, and pneumatic conveying1. Transport velocities or exhaust flows higher than minimum required?

� Consider changing belts and sheaves to reduce air velocity.

2. Consider variable speed or inlet vane control.

Page 183: @Electrical

17

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 units1. Consider air side economizers.

2. Considered chilled water reset.

3. Consider water side economizer.

Centrifugal water chillers1. Multiple chillers operating on a common header.

� 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.� Clean condenser and evaporator tubes.

6. Adding cooling load or chillers?� Consider thermal energy storage.

Cooling towers1. Consider variable speed drives for fan motors.

2. Consider PVC fill to replace wood fill material.

3. Consider velocity recovery stacks.

Boilers1. Stack gas temperature > 400 F? (Ideal temperature: 100 degrees plus saturation temperature of the steam)

� Consider economizer to preheat feedwater or combustion air.

2. Manual or intermittent blowdown?� Consider automatic blowdown system.

3. Continuous blowdown?� Consider blowdown heat recovery system.

4. Excess air high or unburned combustibles?� Consider boiling tuning.

5. Large amounts of high pressure condensate?� Consider high pressure condensate receiver.

6. Increase amount of condensate returned.

7. Improve boiler chemical treatment.

8. Maintain steam traps.

Page 184: @Electrical

18

Heat recovery1. Waste water streams > 100 F?

� Consider heat exchanger and/or heat pump.

2. Waste air or gas stream > 300 F?� Consider heat exchanger.

Cogeneration1. Boiler rated pressure 100 psi greater than pressure required by process?

2. Concurrent steam and electrical demands?� Consider back-pressure turbine.

Refrigeration1. Consider hot gas heat recovery.

2. Consider thermal storage.

Compressed air1. Provide additional small air compressor for loads.

2. Provide outside air intake.

3. Eliminate air leaks.

Page 185: @Electrical

1

Section 15: Project CoordinationBill Brown, P.E., Square D Engineering Services

IntroductionWith 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 phaseThe following aspects are often overlooked during the initial design phase of a project, and can causeconsiderable time and money to be expended later in the process:

� Coordination with the Serving Utility: Coordination with the serving electric utility is vital if a clear understandingis to be achieved between both parties. Often, additional requirements are uncovered that affect the design ofthe 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 runsmoother later in the process, especially for difficult equipment application situations, since a clearunderstanding can be gained regarding the characteristics of the equipment in question and the bestalternatives can be evaluated.

� Coordination with the installing contractor, if the actual construction is under your purview, can save time andfrustrating delays by making your installation requirements clear.

Evaluating equipment bidsIn evaluating equipment bids, any clarifications or exceptions to the project specifications that the equipmentmanufacturers have submitted must be taken into account. This is the time to request re-quotes based uponrejection of the equipment manufacturer’s 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 processOnce 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 deliveryrequirements. 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 markedclearly. Blanket statements to “adhere to the specifications,” without details, can lead to frustrating project delays.This is also the time to submit the manufacturer’s shop drawings to the serving utility and/or localplanning/codes/regulatory authorities, if required. The equipment manufacturer will typically submit to the servingutility if required, but this should be double-checked to avoid confusion. Also, it is good practice to obtain theequipment submittal markups from the serving utility in order to be aware of any changes they request.

Equipment inspectionsIf you require an inspection of the equipment, be sure that the manufacturer clearly understands yourexpectations. Make sure the manufacturer contacts you well in advance of the equipment availability date to allowfor adequate trip planning time.

Page 186: @Electrical

2

CommissioningWhen the equipment begins to arrive on site, it is a good idea to coordinate frequently with the installingcontractor. Arrange for the local sales representatives for the major equipment to be on-site periodically duringconstruction so that problems can be quickly resolved. If the manufacturer’s service technicians are responsiblefor 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 twoweeks in advance to allow for proper planning.

Final acceptanceOnce system commissioning is complete, arrange a walk-through with the client to show the completedinstallation. Also, obtain all equipment operation and maintenance manuals, including field and factory testreports, and store them in a secure area for future use.

Page 187: @Electrical

Guide to Power System Selective Coordination 600V and Below

Document Number0100DB0603

2. Background2.1. What is Selective

Coordination?

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.

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.

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.

Page 188: @Electrical

Guide to Power System Selective Coordination 600V and Below

2 ©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

Fig. 1: Example System

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).

UTILITY SERVICE

A

B

C

D

E

MAIN SWITCHBOARD

LIGHTING PANEL"LP1"

CB M1

CB F1

CB PM1

CB B1

Table I: Protective Device Operation for System of Fig. 1

Fault location Device that should operate for selective coordination

A Utility protective deviceB CB M1C CB F1D CB PM1E CB B1

2.2. The Nature of Overcurrents

Page 189: @Electrical

Guide to Power System Selective Coordination 600V and Below

�©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

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.

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.

2.�. The Protective Zone Concept

UTILITY SERVICE

CB M1

CB F1

CB PM1

CB B1

CB M1 PRIMARYPROTECTIVEZONE

CB F1 PRIMARYPROTECTIVEZONE

CB PM1 PRIMARY PROTECTIVE ZONE

CB B1 PRIMARY PROTECTIVE ZONE

Fig. 2: Ideal Primary Protective Zones for the Example System of Fig. 1

Page 190: @Electrical

Guide to Power System Selective Coordination 600V and Below

� ©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

Note also that, for an overcurrent condition in CB B1’s primary protective zone, if CB B1 fails to operate CB PM1 should operate as a backup. Thus, CB B1’s 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.

Page 191: @Electrical

Guide to Power System Selective Coordination 600V and Below

�©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

Fig. �: Backup Protective Zones for the Example System of Fig. 1

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.

UTILITY SERVICE

UTILITY PROTECTION BACKUP PROTECTIVE ZONE

CB M1

CB F1

CB PM1

CB B1

CB M1 BACKUPPROTECTIVEZONE

CB F1 BACKUP PROTECTIVE ZONE

CB PM1 BACKUP PROTECTIVE ZONECB PM1 BACKUP

PROTECTIVE ZONE

Page 192: @Electrical

Guide to Power System Selective Coordination 600V and Below

6 ©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

CB M1

CB F1

10

100

1K 10K

100K

10

100

1K 10K

100K

0.01

0.10

1

10

100

1000

0.01

0.10

1

10

100

1000

CURRENT IN AMPERES

TIM

E IN

SE

CO

ND

S

30kA Available Fault

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.

2.�. How is selective coordination achieved?

Fig. �: Typical Time-Current Coordination Plot

Page 193: @Electrical

Guide to Power System Selective Coordination 600V and Below

�©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

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. �: Primary Protective Zones for the System of Fig. 1, Showing the Available Fault Current Referenced in Fig. �

UTILITY SERVICE

30kA Available Fault

OvercurrentCB M1

CB F1

CB PM1

CB B1

CB M1 PRIMARYPROTECTIVEZONE

CB F1 PRIMARYPROTECTIVEZONE

CB PM1 PRIMARY PROTECTIVE ZONE

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

Page 194: @Electrical

Guide to Power System Selective Coordination 600V and Below

� ©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

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

Fig. �: Protective Zones for Time-Current Plot of Fig. 6

30kA Available Fault

25kA Available Fault

21.6 kA Available Fault

2kA Available Fault

UTILITY SERVICE

CB F1

CB M1 PRIMARYPROTECTIVEZONE

CB M1

CB PM1

CB B1

CB F1 PRIMARYPROTECTIVEZONE

CB PM1 PRIMARY PROTECTIVE ZONE

CB B1 PRIMARY PROTECTIVE ZONE

CB F1, CB PM1 Coordinate through 21.6kA

CB PM1, CB B1 Coordinate through 2kA

10 100

1K 10K

100K

10

100 1K

10K

100K

0.01 0.01

0.10 0.10

1 1

10 10

100 100

1000 1000

CURRENT IN AMPERES

TIM

E IN

SE

CO

ND

S

CB M1

CB F1

CB PM1

CB B1

Page 195: @Electrical

Guide to Power System Selective Coordination 600V and Below

�©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

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.

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.�. What about Equipment Protection?

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]:

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. NEC Requirements for Selective Coordination

2.6.1. Coordinated Short-Circuit Protection/Overload Indication Permitted When Orderly Shutdown is Required (NEC 2�0.12)

Page 196: @Electrical

Guide to Power System Selective Coordination 600V and Below

10 ©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

2.6.2. Elevators, Dumbwaiters, Escalators, Moving Walks, Wheelchair Lifts, and Stairway Chair Lifts (NEC 620.62)

2.6.�. Emergency Systems (NEC �00.2�)

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.

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.

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).

2.6.�. Legally Required Standby Systems (NEC �01.1�)

Page 197: @Electrical

Guide to Power System Selective Coordination 600V and Below

11©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

2.6.�. Service Ground-Fault Protection for Equipment (NEC 2�0.��)

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 ground-fault 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.

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

2.6.6. Feeder Ground-Fault Protection for Equipment (NEC 21�.10)

Page 198: @Electrical

Guide to Power System Selective Coordination 600V and Below

12 ©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

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.

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.

2.6.�. Ground-Fault Protection in Heath Care Facilities (NEC �1�.1�)

Page 199: @Electrical

Guide to Power System Selective Coordination 600V and Below

1�©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

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.

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.

2.�. The Coordination Study

Page 200: @Electrical

Guide to Power System Selective Coordination 600V and Below

1� ©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

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 time-current 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 time-current 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

Fig. �: Fuse Timing Illustration

Overcurrent condition initiated.Fuse element begins to melt

Fuse element is melted. Arcing begins.

Overcurrent is cleared.

Melting Time Arcing Time

Total Clearing Time

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.

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 current-limiting range within ½ 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.

�. Protective Device Characteristics

�.1. Fuses

Page 201: @Electrical

Guide to Power System Selective Coordination 600V and Below

1�©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

Fig. �: Typical Low-Voltage Fuse Time-Current Characteristic Band

100 1K 10K

100K

100 1K 10K

100K

0.01

0.10

1

10

100

1000

0.01

0.10

1

10

100

1000

CURRENT IN AMPERES

TIM

E IN

SE

CO

ND

S

Minimum Melting

Characteristic

Total Clearing Characteristic

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.

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.

Page 202: @Electrical

Guide to Power System Selective Coordination 600V and Below

16 ©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

Fig. 10: Fuse Coordination Example

FU1 & FU2 Coordinate to 8200A Above 8200A Coordination must be established by fuse ratio tables

10

10

100

100

1K1K

10K

10K

100K

100K

0.01 0.01

0.10 0.10

1 1

10 10

100 100

1000 1000

CURRENT IN AMPERES

TIM

E IN

SE

CO

ND

S

FU 1

FU 2

UTILITY BUS

FU 1

FU 2

FU 1

FU 2

�.2. Circuit Breakers 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

Page 203: @Electrical

Guide to Power System Selective Coordination 600V and Below

1�©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

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 instantaneous-only 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 manufacturer’s data must be consulted for a given circuit breaker.

Table II: Low-Voltage Circuit Breaker Types1

Circuit Breaker Type

Standard Tripping Type Short-time Withstand Capability2

Molded-Case UL 489

Thermal-magnetic Typically much lower than interrupting rating

Electronic Typically lower than interrupting rating

Electronic (insulated case)3

Often comparable to interrupting rating

Low-Voltage Power

ANSI C37.13 UL 1066

Electronic Typically comparable to interrupting rating

1 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

2 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 ½-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.

3 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.

�.2.1. Thermal-Magnetic Molded-Case Circuit Breakers

Page 204: @Electrical

Guide to Power System Selective Coordination 600V and Below

1� ©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

Fig. 11: Typical Thermal-Magnetic Molded-Case Circuit Breaker Time-Current Characteristic Band

Thermal (long-time) Characteristic

Magnetic (instantaneous) Characteristic

100

1K 10K

100K

TIM

E IN

SE

CO

ND

S

10

100 1K

10K

100K10

0.01 0.01

0.10 0.10

1 1

10 10

100 100

1000 1000

CURRENT IN AMPERES

Fig. 12: Thermal-Magnetic Circuit Breaker Time-Current Characteristic showing adjustable instantaneous characteristic

Magnetic (instantaneous) Characteristic HI setting

Magnetic (instantaneous) Characteristic LO setting

100

100

1K1K

10K

10K

100K

100K

TIM

E IN

SE

CO

ND

S

10

10

0.01 0.01

0.10 0.10

1 1

10 10

100 100

1000 1000

CURRENT IN AMPERES

Page 205: @Electrical

Guide to Power System Selective Coordination 600V and Below

1�©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

�.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 breaker’s 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 non-adjustable. 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.

Page 206: @Electrical

Guide to Power System Selective Coordination 600V and Below

20 ©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

Fig. 1�: Typical Time-current characteristics for electronic-trip circuit breaker (molded-case circuit breaker with “low” short-time withstand shown)

100

100

1K1K

10K

10K

100K

100K

TIM

E IN

SE

CO

ND

S

10

10

0.01 0.01

0.10 0.10

1 1

10 10

100 100

1000 1000

CURRENT IN AMPERES

Instantaneous Override Value (24kA -10%)

Instantaneous Timing is Non-Adjustable

Instantaneous Override Timing is Non-Adjustable

Adjustable Long-Time Pickup

Adjustable Long-Time Delay

Adjustable Short-Time Pickup

Adjustable Short-Time Delay

Adjustable Instantaneous Pickup

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.

�.�. Current-Limiting Circuit Breakers

Page 207: @Electrical

Guide to Power System Selective Coordination 600V and Below

21©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

Fig. 1�: Typical Time-Current Characteristic for Current-Limiting Circuit Breaker

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.

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

0.5 1 10

100

1K 10K

100K

1 10 100

1K 10K

100K

0.01 0.01

0.10 0.10

1 1

10 10

100 100

1000 1000

CURRENT IN AMPERES

TIM

E IN

SE

CO

ND

S

�.�. Circuit Breakers in Series: The Dynamic Impedance Concept

�.�.1. Short-Circuit Coordination Tables

Page 208: @Electrical

Guide to Power System Selective Coordination 600V and Below

22 ©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

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 time-current 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. 1�: 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

CB PM1, CB B1 Coordinate through 2kA

21.6kA

10

100

100

1K1K

10K

10K

100K

100K

TIM

E IN

SE

CO

ND

S

10

0.01 0.01

0.10 0.10

1 1

10 10

100 100

1000 1000

CURRENT IN AMPERES

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 manufacturer’s circuit breakers)

CB M1

CB F1

CB PM1

CB B1

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.

�.�. Ground-Fault Protection of Equipment

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

Page 209: @Electrical

Guide to Power System Selective Coordination 600V and Below

2�©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

Fig. 16: Typical Ground-Fault Protection Characteristic

0.5 1 10

100

1K 10K

100K

1 10 100

1K 10K

100K

0.01 0.01

0.10 0.10

1 1

10 10

100 100

1000 1000

CURRENT IN AMPERES

TIM

E IN

SE

CO

ND

S

GF CHAR (TYP)

�. Tying it All Together – Design Philosophies and Guidelines

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.

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/CT’s, a single zero-sequence sensor/CT, or may be a complex affair with differential connections of the sensors/CT’s, known as a modified-differential ground-fault arrangement. The application of the sensors/CT’s is beyond the scope of this paper but the engineer responsible for coordination should be cognizant of the requirements and potential application issues.

Page 210: @Electrical

Guide to Power System Selective Coordination 600V and Below

2� ©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

Table III: Selective coordination requirements/comments per various industry standards

Standard Requirement/Comment

NFPA 110 Standard for Emergency and Standby Power Systems [6]

6.� Protection 6.�.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 AA.6.�.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 � Application and Coordination of Protective Devices

�.1.� 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.

�.2 Analysis of System Behavior and Protection Needs

�.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 � System Protection and Coordination

�.� Selective Coordination

�.�.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.

�.�.� 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 Recommended Practice for 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 1� Overcurrent coordination

1�.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 Recommended Practice for Emergency and Standby Power Systems for Industrial and Commercial Applications (Orange Book) [7]

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…

Page 211: @Electrical

Guide to Power System Selective Coordination 600V and Below

2�©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

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 industry’s assertion that available fault current values can change over time due to system changes), although this is typically a challenge due to the industry’s reliance on infinite bus calculations.

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

�.1. Consider Selective Coordination Early in the Design Process

�.2. Recognize the Conflicts and Issues with the 200� NEC

�.2.1. Selective Coordination – What is it?

Page 212: @Electrical

Guide to Power System Selective Coordination 600V and Below

26 ©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

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. 1�: Typical Low-Voltage Transformer Protection Scenario

SECONDARY CB

PRIMARY CB

TRANSFORMER

PRIMARY CBPROTECTIVE ZONE

SECONDARY CBPROTECTIVE 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.

Page 213: @Electrical

Guide to Power System Selective Coordination 600V and Below

2�©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

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.

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.

Fig. 1�: 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

PANEL 1

PANEL 2

CB 1

CB 2

G

CB 1

CB 2

ENGINE-GENERATOR SET

SWITCHBOARD

a.) b.)

�.2.2. Ground-Fault Protection in Health-Care Facilities

Page 214: @Electrical

Guide to Power System Selective Coordination 600V and Below

2� ©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

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. 1�: Typical Health-Care Facility Electrical System (Source: NEC 200� FPN Figure �1�.�0)

Automatic switching equipment

Delayed automatic switching equipment

Nonessential loads

Normal source

Alternate power source

Equipment system

Life safety branch

Critical branch

Emergency system

Essential electrical system

Normal 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.

Page 215: @Electrical

Guide to Power System Selective Coordination 600V and Below

2�©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

As mentioned in 2.2 above, the frequency of occurrence of high-magnitude 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.

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.

�.2.�. Is Coordination up to the Available Fault current Justified on a Practical Basis?

�.2.�. Avoid Placing Protective Devices in Series with No Equipment Between Them

Page 216: @Electrical

Guide to Power System Selective Coordination 600V and Below

�0 ©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

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 regulator’s 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.

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

Fig. 20: Examples of Fig. 1� re-designed to eliminate redundant devices

PANEL 1

PANEL 2

CB 1G

CB 1

ENGINE-GENERATOR SET

SWITCHBOARD

a.) b.)

�.�. Recognize the Pitfalls of Generator Protection

Page 217: @Electrical

Guide to Power System Selective Coordination 600V and Below

�1©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

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

G

CB 2

AUTOXFERSW

CB 4

TO NORMAL SOURCE

EN

CB 3

AUTOXFERSW E N AUTOXFER

SW

CB 5

EN

Fig. 22: System of Fig. 21 Expanded to Show Primary Protective Zones

G

CB 1

G

CB 2

AUTOXFERSW

CB 4

TO NORMAL SOURCE

EN

CB 3

AUTOXFERSW E

N AUTOXFERSW

CB 5

EN

CB 6

CB 1 PROTECTIVE ZONE CB 2 PROTECTIVE ZONE

CB 3 PROTECTIVE ZONE

CB 6 PROTECTIVE ZONE

Page 218: @Electrical

Guide to Power System Selective Coordination 600V and Below

�2 ©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

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. 2�: System of Fig. 22 Re-designed for Selective Coordination

G G

AUTOXFERSW

TO NORMAL SOURCE

E NAUTOXFERSW

E N AUTOXFERSW

E N

CB 6

CB 1 PROTECTIVE ZONE

CB 6 PROTECTIVE ZONE

G

CB 1

Fig. 2�: System of Fig. 22 with Higher Settings for CB 1 and Differential Relaying Added

G G

AUTOXFERSW

CB 4

TO NORMAL SOURCE

E NAUTOXFERSW

E N AUTOXFERSW

CB 5

E N

CB 1 CB 2

87 B

CB 3

CB 6

CB 1 PROTECTIVE ZONE CB 2 PROTECTIVE ZONE

87B PROTECTIVE ZONE

CB 3 PROTECTIVE ZONE

CB 6 PROTECTIVE ZONE

Page 219: @Electrical

Guide to Power System Selective Coordination 600V and Below

��©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

In Fig. 24, the differential relay 87B would typically be of the high-impedance 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”.

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.

�.�. Utilize Circuit Breakers with High Short-Time Withstand Capabilities

Page 220: @Electrical

Guide to Power System Selective Coordination 600V and Below

�� ©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

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 low-voltage 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.

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.

�.�. Avoid Multiple Levels of Protective Devices Where Possible

Fig. 2�: Illustration Showing Multiple Levels of Selectivity: a.) Three Levels b.) Same Number of Panels Re-Arranged with Two Levels

PANEL 1

PANEL 2

CB 1

CB 2

a.) b.)

PANEL 3

CB 3

PANEL 1

PANEL 2

CB 1

CB 2

PANEL 3

CB 3

Page 221: @Electrical

Guide to Power System Selective Coordination 600V and Below

��©2006 Schneider Electric. All rights reserved.

Document Number0100DB0603

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.

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.

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.

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, thermal-magnetic circuit breakers with adjustable instantaneous characteristics, ground-fault relays, etc.

�.6. Utilize Step-Down Transformers to Lower Fault Current

�.�. Increase Transformer Sizes Where Necessary

�.�. Zone-Selective Interlocking – The Facts and the Misconceptions

�.�. Don’t Forget On-Site Adjustment Requirements

Page 222: @Electrical

Guide to Power System Selective Coordination 600V and Below

1415 S. Roselle Road Palatine, IL 60067Tel: 847-397-2600Fax: 847-925-7500

Schneider Electric - North American Operating Division

©20

06 S

chne

ider

Ele

ctric

. All

right

s re

serv

ed.

tk

Document Number0100DB0603

�. References [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.

[9] IEEE Recommended Practice for Electric Power Systems in Commercial Buildings, IEEE Std. 241-1990, December 1990.

Page 223: @Electrical

Selectivity Guidelines For Square D® Panelboards

The natural advantages of circuit breakers make them the logical choice forovercurrent protection. New requirements in the National Electrical Code foremergency and legally required standby systems make it advantageous toconsider selective coordination at the beginning of the design process. Thisguide is intended to facilitate the design of selectively coordinated systemswhen using Square D® I-Line®, NF and NQOD panelboards.

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 SelectiveCoordination for Low Voltage Circuit Breakers) is utilized, along with TCCcomparisons where necessary. The result is a set of tables which allow foreasy and efficient selection of Square D panelboards and their overcurrentdevices. Two specifications for “selective coordination” are considered:coordination from 0.1 – 1000s and coordination from 0.01 - 1000s. Thespecification that is used will depend upon the NEC and other coderequirements of the installation and the interpretation of these requirementsby the authority having jurisdiction.

The tables herein may be used to select feeder and branch circuit breakersthat will be selectively coordinated when NF and NQOD panelboards are usedin a configuration as illustrated below:

Introduction

I-LINE (UPSTREAM)PANELBOARD - 480Y/277Vor 208Y/120VV

FEEDERCIRCUITBREAKER

FEEDER BRANCHCIRCUITBREAKER

NF Available Ampacities: 125A,250A, 400A, 600A, 800A

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

NF (480Y/277V208Y/120V) orNQOD (208Y/120V)MLO(DOWNSTREAM)PANELBOARD

Available Ampacities: 100A,225A, 400A, 600A, 800A,1200A

Alternatively, the upstream panelboard may be an I-Line section incorporatedinto a QED-2 switchboard.

Document Number 0100DB0604

Page 224: @Electrical

Listing Of Tables

To find the table which applies to your application: Select a downstreampanelboard type in the left-hand column. Read across the row to find the tablewhich is listed under the appropriate voltage level and selectivity specification.For example, if the downstream panelboard is an NF panelboard, theappropriate table for selectivity from 0.01 – 1000s at 480Y/277V is Table IB.

Assumptions All thermal-magnetic circuit breakers with adjustable instantaneous tripsettings are assumed to have their instantaneous settings at maximum

Electronic-trip circuit breakers are assumed to have the smallestsensor/rating plug size which meets or exceeds the ampacity requirementsof the given circuit. The long-time trip/delay must be set to the appropriatelevel to give the breaker trip setting shown. The instantaneous function isassumed to be turned off, if possible for the breakers under consideration,or otherwise set to maximum. Electronic-trip circuit breakers are assumed tohave a short-time function and the short-time pickup and delay settings areassumed to be set at maximum

All circuit breakers are shown with their maximum available ampacityranges. For most circuit breakers, these apply for 2- or 3-poleconfigurations, although this is not always the case. The availability of agiven circuit breaker ampacity for a given model and configuration must bedouble-checked

How To Use The Tables If feeder size is known:

1. Locate Feeder Size/Upstream Panelboard Circuit Breaker Size inleftmost column

2. Required Downstream Panelboard Ampacity is in next column to right

3. Follow row to right and select the closest Maximum Available FaultCurrent at Upstream Panelboard which is greater than or equal to theavailable fault current at upstream panelboard (adjust available fault currentvalue if necessary due to system X/R ratio - see table explanatory notes)

4. Follow row to right and select an Upstream Panelboard Feeder CircuitBreaker Type

5. Follow row to right to obtain the Downstream Panelboard Branch CircuitBreaker Type and the Largest Possible Branch Circuit Breaker. For“total coordination” tables, the Maximum Available Fault Current atDownstream Panelboard is given also. As long as the circuit breaker typeand maximum size are adhered to (and the available fault current at thedownstream panelboard is less than or equal to the value shown for “totalcoordination” tables), selective coordination will be achieved as per thecoordination parameters for the table

2 © 2006 Schneider Electric. All rights reserved.

Selectivity Guidelines for Square D® Panelboards

Downstream Panelboard Type Upstream Panelboard Type: I-Line®

208Y/120V 480Y/277V0.1 - 1000s 0.01 - 1000s 0.1 - 1000s 0.01 - 1000s

NF Table IIA Table IIB Table IA Table IBNQOD Table IIIA Table IIIB N/A N/A

Document Number 0100DB0604

Page 225: @Electrical

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 FeederCircuit 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 FeederSize/Upstream Panelboard Circuit Breaker Size and repeat steps 1through 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 Breakersizes in rightmost column. Select the first one that is greater than or equalto the desired branch circuit size. For “total coordination” tables, MaximumAvailable Fault Current at Downstream Panelboard must be greaterthan or equal to the actual fault current at the downstream panelboard

2. When the desired branch circuit breaker is found, follow row to left. Makesure that the actual available fault current at the upstream panelboard isless than or equal to the Maximum Available Fault Current at UpstreamPanelboard (adjust available fault current value if necessary due to systemX/R ratio - see table explanatory notes)

3. The required Downstream Panelboard Ampacity and FeederSize/Upstream Panelboard Circuit Breaker Size are as shown. This isthe smallest feeder circuit breaker that will satisfy the coordination criteriafor the table

4. Scan the rightmost column for other instances of the required branch circuitbreaker size and follow steps 1 through 3 again. The feeder circuit/I-Linefeeder circuit breaker size may be larger, but the I-Line circuit breaker maybe less expensive

3© 2006 Schneider Electric. All rights reserved.

Selectivity Guidelines for Square D® PanelboardsDocument Number

0100DB0604

Page 226: @Electrical

FEEDER SIZE /UPSTREAM (I-LINE)

PANELBOARD CIRCUITBREAKER SIZE

(A)

REQUIREDDOWNSTREAM (NF)

PANELBOARDAMPACITY

(A)

MAXIMUM AVAILABLEFAULT CURRENT ATUPSTREAM (I-LINE)

PANELBOARD(kA RMS Sym.)1

UPSTREAM (I-LINE)PANELBOARD

FEEDER CIRCUITBREAKER TYPE

DOWNSTREAM (NF)PANELBOARD BRANCH

CIRCUIT BREAKERTYPE

LARGEST POSSIBLEBRANCH CIRCUIT

BREAKER(A)

Table IAI-Line®/NF Panelboard Selective Coordination At 480Y/277V0.1s – 1000s

4 © 2006 Schneider Electric. All rights reserved.

18 FA, HD, LX ED 30PG ED 50

25 FH EG 30100 125 HG EG 30

35 LX EG 30PG EG 50

65 HJ, LX EJ 30PJ EJ 50

18 HD ED 30110 125 35 HG EG 30

65 HJ EJ 3018 HD, LA, LX ED 30

PG ED 5030 LA EG 30

125 125 35 HG, LH, LX EG 30PG EG 50

65 HJ, LX EJ 30PJ EJ 50HD ED 30

18 JD, LA ED 35LX ED 40PG ED 70

30 LA EG 35HG EG 30

150 250 JG, LH EG 3535 LX EG 40

PG EG 70HJ EJ 30JJ EJ 35

65 LX EJ 40PJ EJ 70

18 JD, LA ED 40LX, PG ED 70

30 LA EG 40175 250 35 JG, LH EG 40

LX, PG EG 7065 JJ EJ 40

LX, PJ EJ 70JD, LA, LX, ED 70

18 PG ED 80LA-MC ED 125

30 LA EG 70200 250 LA-MC EG 125

JG, LH, LX EG 7035 PG EG 80

LH-MC EG 12565 JJ, LX EJ 70

PJ EJ 80JD, LA, LX ED 70

18 PG ED 110LA-MC ED 125

30 LA EG 70LA-MC EG 125

225 250 JG, LH, LX EG 7035 PG EG 110

LH-MC EG 12565 JJ, LX EJ 70

PJ EJ 110JD, LA ED 70

18 LA-MC ED 125LX, PG ED 125

30 LA EG 70250 250 LA-MC EG 125

JG, LH EG 7035 LH-MC EG 125

LX, PG EG 12565 JJ EJ 70

LX, PJ EJ 125

Selectivity Guidelines for Square D® PanelboardsDocument Number

0100DB0604

Page 227: @Electrical

5

X/R Ratio Adjustment:All available fault currents are given in RMS symmetrical amperes. For asystem X/R ratio larger than the test X/R ratio of the circuit breaker inquestion, the available fault current equivalent RMS symmetrical duty forcomparison with the values in the tables must be adjusted by a multiplyingfactor. See IEEE Std. 242-2001 (Buff Book), IEEE Std. 1015-1997 (Blue Book)or NEMA AB 3-2001 for details.

Note that this is a consideration for breaker fault duty rather than for selective coordination.

1 Available fault currents are based upon system X/R ratios less than or equal to the circuit breaker test X/R ratio. Seethe 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 rangePG (ET1.0) = ET1.0 trip unitPG = 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)

© 2006 Schneider Electric. All rights reserved.

FEEDER SIZE /UPSTREAM (I-LINE)

PANELBOARD CIRCUITBREAKER SIZE

(A)

REQUIREDDOWNSTREAM (NF)

PANELBOARDAMPACITY

(A)

MAXIMUM AVAILABLEFAULT CURRENT ATUPSTREAM (I-LINE)

PANELBOARD(kA RMS Sym.)1

UPSTREAM (I-LINE)PANELBOARD

FEEDER CIRCUIT BREAKER TYPE

DOWNSTREAM (NF)PANELBOARD BRANCH

CIRCUIT BREAKERTYPE

LARGEST POSSIBLEBRANCH CIRCUIT

BREAKER

18 LA, MG, LX, PG3 ED 125300 400 30 LA EG 125

35 LH, MG, LX, PG3 EG 12565 LC, MJ, LX, PJ3 EJ 12518 LA, MG, LX ED 125

350 400 30 LA EG 12535 LH, MG, LX EG 12565 LC, MJ, LX EJ 12518 LA, LA-MC, MG, LX, PG ED 125

400 400 30 LA, LA-MC EG 125

35 LH, LH-MC, MG, LX, PG EG 125

65 LC, MJ, LX, PJ EJ 12518 LC, MG, LX, PG3 ED 125

450 600 35 LC, MG, LX, PG3 EG 12565 LC, MJ, LX, PJ3 EJ 12518 LC, MG, LX, PG3 ED 125

500 600 35 LC, MG, LX, PG3 EG 12565 LC, MJ, LX, PJ3 EJ 12518 LC, MG, LX, PG

(ET1.0)2, PG2 ED 125600 600 35 LC, MG, LX, PG

(ET1.0)2, PG2 EG 12565 LC, MJ, LX, PJ (ET1.0)2,

PJ2 EJ 12518 MG, PG2,3 ED 125

700 800 35 MG, PG2,3 EG 12565 MJ, PJ2,3 EJ 12518 MG, PG (ET1.0)2, PG2 ED 125

800 800 35 MG, PG (ET1.0)2, PG2 EG 12565 MJ, PJ (ET1.0)2, PJ2 EJ 125

Selectivity Guidelines for Square D® Panelboards

Molded Case Circuit Breaker Interrupting Rating Test X/RGreater than 20kA 4.910kA - 20kA 3.2Less than 10kA 1.7

Document Number 0100DB0604

Page 228: @Electrical

Table IBI-Line®/NF Panelboard Selective Coordination At 480Y/277V0.01s – 1000s

© 2006 Schneider Electric. All rights reserved.

FEEDER SIZE /UPSTREAM (I-LINE)

PANELBOARD CIRCUITBREAKER SIZE

(A)

REQUIREDDOWNSTREAM (NF)

PANELBOARDAMPACITY

(A)

MAXIMUM AVAILABLEFAULT CURRENT ATUPSTREAM (I-LINE)

PANELBOARD(kA RMS Sym.)1

UPSTREAM (I-LINE)PANELBOARD

FEEDER CIRCUITBREAKER TYPE

DOWNSTREAM (NF)PANELBOARD BRANCH

CIRCUIT BREAKERTYPE

MAXIMUM AVAILABLEFAULT CURRENT ATDOWNSTREAM (NF)

PANELBOARD(kA RMS Sym.)1,3

LARGEST POSSIBLEBRANCH CIRCUIT

BREAKER

18 PG ED 18 50100 125 35 PG EG 35 50

65 PJ EJ 65 5018 PG ED 18 50

125 125 35 PG EG 35 5065 PJ EJ 65 5018 PG ED 18 70

150 250 35 PG EG 35 7065 PJ EJ 65 7018 PG ED 18 70

175 250 35 PG EG 35 7065 PJ EJ 65 70

PG ED 18 8018 LA-MC ED 18 15

LA-MC ED 10 20LA-MC ED 6 100LA-MC EG 18 15

200 250 30 LA-MC EG 10 20LA-MC EG 6 100

PG EG 35 8035 LH-MC EG 18 15

LH-MC EG 10 20LH-MC EG 6 100

65 PJ EJ 65 80PG ED 18 110

LA-MC ED 18 1518 LA-MC ED 14 20

LA-MC ED 8 30LA-MC ED 7 100LA-MC EG 18 15

225 250 30 LA-MC EG 14 20LA-MC EG 8 30LA-MC EG 7 100

PG EG 35 110LH-MC EG 18 15

35 LH-MC EG 14 20LH-MC EG 8 30LH-MC EG 7 100

65 PJ EJ 65 110LA-MC ED 18 30

18 LA-MC ED 10 40LA-MC ED 8 100

PG ED 18 125LA-MC EG 18 30

250 250 30 LA-MC EG 10 40LA-MC EG 8 100LH-MC EG 18 30

35 LH-MC EG 10 40LH-MC EG 8 100

PG EG 35 12565 PJ EJ 65 12518 PG4 ED 18 125

300 300 35 PG4 EG 35 12565 PJ4 EJ 65 125

LA-MC ED 18 10018 LA-MC ED 6 125

PG ED 18 125400 400 30 LA-MC EG 18 100

LA-MC EG 6 125LH-MC EG 18 100

35 LH-MC EG 6 125PG EG 35 125

65 PJ EJ 65 12518 PG4 ED 18 125

450 600 35 PG4 EG 21.6 12565 PJ4 EJ 9 12518 PG4 ED 18 125

500 600 35 PG4 EG 21.6 12565 PJ4 EJ 9 125

600 600 18 PG (ET1.0)2, PG2 ED 18 125

6

Selectivity Guidelines for Square D® PanelboardsDocument Number

0100DB0604

Page 229: @Electrical

7

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 explanatorynote 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 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: 1000Aw/LTPU=0.625, 700A: 1000A w/LTPU= 0.7)

X/R Ratio Adjustment:All available fault currents are given in RMS symmetrical amperes. For asystem X/R ratio larger than the test X/R ratio of the circuit breaker inquestion, the available fault current equivalent RMS symmetrical duty forcomparison with the values in the tables must be adjusted by a multiplyingfactor. See IEEE Std. 242-2001 (Buff Book), IEEE Std. 1015-1997 (Blue Book)or NEMA AB 3-2001 for details.

Note that this is a consideration for breaker fault duty rather than for selective coordination.

© 2006 Schneider Electric. All rights reserved.

FEEDER SIZE /UPSTREAM (I-LINE)

PANELBOARD CIRCUITBREAKER SIZE

(A)

REQUIREDDOWNSTREAM (NF)

PANELBOARDAMPACITY

(A)

MAXIMUM AVAILABLEFAULT CURRENT ATUPSTREAM (I-LINE)

PANELBOARD(kA RMS Sym.)1

UPSTREAM (I-LINE)PANELBOARD

FEEDER CIRCUITBREAKER TYPE

DOWNSTREAM (NF)PANELBOARD BRANCH

CIRCUIT BREAKERTYPE

MAXIMUM AVAILABLEFAULT CURRENT ATDOWNSTREAM (NF)

PANELBOARD(kA RMS Sym.)1,3

LARGEST POSSIBLEBRANCH CIRCUIT

BREAKER(A)

35 PG (ET1.0)2, PG2 EG 35 12565 PJ (ET1.0)2, PJ2 EJ 65 12518 PG2,4 ED 18 12535 PG2,4 EG 21.6 12565 PJ2,4 EJ 9 12518 PG (ET1.0)2, PG2 ED 18 12535 PG (ET1.0)2, PG2 EG 21.6 12565 PJ (ET1.0)2, PJ2 EJ 9 125

700 800

800 800

Selectivity Guidelines for Square D® Panelboards

Molded Case Circuit Breaker Interrupting Rating Test X/RGreater than 20kA 4.910kA - 20kA 3.2Less than 10kA 1.7

Document Number 0100DB0604

Page 230: @Electrical

Table IIAI-Line®/NF Panelboard Selective Coordination At 208Y/120V0.1s – 1000s

© 2006 Schneider Electric. All rights reserved.

Selectivity Guidelines for Square D® Panelboards

FEEDER SIZE /UPSTREAM (I-LINE)

PANELBOARD CIRCUITBREAKER SIZE

(A)

REQUIREDDOWNSTREAM (NF)

PANELBOARDAMPACITY

(A)

MAXIMUM AVAILABLEFAULT CURRENT ATUPSTREAM (I-LINE)

PANELBOARD(kA RMS Sym.)1

UPSTREAM (I-LINE)PANELBOARD

FEEDER CIRCUITBREAKER TYPE

DOWNSTREAM (NF)PANELBOARD BRANCH

CIRCUIT BREAKER TYPE

LARGEST POSSIBLEBRANCH CIRCUIT

BREAKER (A)

25 FA2, HD, LX ED 30PG ED 50

100 125 65 FH, HG, LX EG3 30PG EG3 50

100 HJ, LX EJ3 30PJ EJ3 50

25 HD ED 30110 125 65 HG EG3 30

100 HJ EJ3 3025 HD, LA, LX ED 30

PG ED 5042 LA EG3 30

125 125 65 HG, LH, LX EG3 30PG EG3 50

100 HJ, LX EJ3 30PJ EJ3 50HD ED 30

25 JD, LA ED 35LX ED 40PG ED 70

42 LA EG3 35HG EG3 30

150 250 65 JG, LH EG3 35LX EG3 40PG EG3 70HJ EJ3 30

100 JJ EJ3 35LX EJ3 40PJ EJ3 70

25 JD, LA ED 40LX, PG ED 70

42 LA EG3 40175 250 65 JG, LH EG3 40

LX, PG EG3 70100 JJ EJ3 40

LX, PJ EJ3 70JD, LA, LX, ED 70

25 PG ED 80LA-MC ED 125

42 LA EG3 70200 250 LA-MC EG3 125

JG, LH, LX EG3 7065 PG EG3 80

LH-MC EG3 125100 JJ, LX EJ3 70

PJ EJ3 80JD, LA, LX ED 70

25 PG ED 110LA-MC ED 125

42 LA EG3 70LA-MC EG3 125

225 250 JG, LH, LX EG3 7065 PG EG3 110

LH-MC EG3 125100 JJ, LX EJ3 70

PJ EJ3 110JD, LA ED 70

25 LA-MC ED 125LX, PG ED 125

42 LA EG3 70250 250 LA-MC EG3 125

JG, LH EG3 7065 LH-MC EG3 125

LX, PG EG3 125100 JJ EJ3 70

LX, PJ EJ3 12525 LA, MG, LX, PG5 ED 125

300 400 42 LA EG3 12565 LH, MG, LX, PG5 EG3 125

8

Document Number 0100DB0604

Page 231: @Electrical

9© 2006 Schneider Electric. All rights reserved.

1 Available fault currents are based upon system X/R ratios less than or equal to the circuit breaker test X/R ratio. Seethe explanatory note 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

4 The P-Frame Powerpact® circuit breaker is available with ET1.0 or Micrologic 5.0/6.0 trip units in this size rangePG (ET1.0) = ET1.0 trip unitPG = Micrologic 5.0/6.0 trip unit

5 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 asystem X/R ratio larger than the test X/R ratio of the circuit breaker inquestion, the available fault current equivalent RMS symmetrical duty forcomparison with the values in the tables must be adjusted by a multiplyingfactor. See IEEE Std. 242-2001 (Buff Book), IEEE Std. 1015-1997 (Blue Book)or NEMA AB 3-2001 for details.

Note that this is a consideration for breaker fault duty rather than for selective coordination.

Selectivity Guidelines for Square D® Panelboards

FEEDER SIZE /UPSTREAM (I-LINE)

PANELBOARD CIRCUITBREAKER SIZE

(A)

REQUIREDDOWNSTREAM (NF)

PANELBOARDAMPACITY

(A)

MAXIMUM AVAILABLEFAULT CURRENT ATUPSTREAM (I-LINE)

PANELBOARD(kA RMS Sym.)1

UPSTREAM (I-LINE)PANELBOARD

FEEDER CIRCUIT BREAKER TYPE

DOWNSTREAM (NF)PANELBOARD BRANCH

CIRCUIT BREAKERTYPE

LARGEST POSSIBLEBRANCH CIRCUIT

BREAKER(A)

100 LC, MJ, LX, PJ5 EJ3 12525 LA, MG, LX ED 12542 LA EG3 12565 LH, MG, LX EG3 125

100 LC, MJ, LX EJ3 12525 LA, LA-MC, MG, LX, PG ED 12542 LA, LA-MC EG3 12565 LH, LH-MC, MG, LX, PG EG3 125

100 LC, MJ, LX, PJ EJ3 12525 LC, MG, LX, PG5 ED 12565 LC, MG, LX, PG5 EG3 125

100 LC, MJ, LX, PJ5 EJ3 12525 LC, MG, LX, PG5 ED 12565 LC, MG, LX, PG5 EG3 125

100 LC, MJ, LX, PJ5 EJ3 12525 LC, MG, LX, PG

(ET1.0)4, PG4 ED 12565 LC, MG, LX, PG

(ET1.0)4, PG4 EG3 125100 LC, MJ, LX, PJ

(ET1.0)4, PJ4 EJ3 12525 MG, PG4,5 ED 12565 MG, PG4,5 EG3 125

100 MJ, PJ4,5 EJ3 12525 MG, PG (ET1.0)4, PG4 ED 12565 MG, PG (ET1.0)4, PG4 EG3 125

100 MJ, PJ (ET1.0)4, PJ4 EJ3 125

350 400

400 400

450 600

500 600

600 600

700 800

800 800

Molded Case Circuit Breaker Interrupting Rating Test X/RGreater than 20kA 4.910kA - 20kA 3.2Less than 10kA 1.7

Document Number 0100DB0604

Page 232: @Electrical

© 2006 Schneider Electric. All rights reserved.

TABLE IIBI-Line®/NF Panelboard Selective Coordination At 208Y/120V0.01s – 1000s

Selectivity Guidelines for Square D® Panelboards

FEEDER SIZE /UPSTREAM (I-LINE)

PANELBOARD CIRCUITBREAKER SIZE

(A)

REQUIREDDOWNSTREAM (NF)

PANELBOARDAMPACITY

(A)

MAXIMUM AVAILABLEFAULT CURRENT ATUPSTREAM (I-LINE)

PANELBOARD(kA RMS Sym.)1

UPSTREAM (I-LINE)PANELBOARD

FEEDER CIRCUITBREAKER TYPE

DOWNSTREAM (NF)PANELBOARD BRANCH

CIRCUIT BREAKER TYPE

MAXIMUM AVAILABLEFAULT CURRENT ATDOWNSTREAM (NF)

PANELBOARD(kA RMS Sym.)1,5

LARGEST POSSIBLEBRANCH CIRCUIT

BREAKER(A)

25 PG ED 21.6 50100 125 65 PG EG3 65 50

100 PJ EJ3 100 5025 PG ED 21.6 50

125 125 65 PG EG3 65 50100 PJ EJ3 100 5025 PG ED 21.6 70

150 250 65 PG EG3 65 70100 PJ EJ3 100 7025 PG ED 21.6 70

175 250 65 PG EG3 65 70100 PJ EJ3 100 70

PG ED 21.6 8025 LA-MC ED 18 15

LA-MC ED 10 20LA-MC ED 6 100LA-MC EG3 18 15

200 250 42 LA-MC EG3 10 20LA-MC EG3 6 100

PG EG3 65 8065 LH-MC EG3 18 15

LH-MC EG3 10 20LH-MC EG3 6 100

100 PJ EJ3 100 80PG ED 21.6 110

LA-MC ED 18 1525 LA-MC ED 14 20

LA-MC ED 8 30LA-MC ED 7 100LA-MC EG3 18 15

42 LA-MC EG3 14 20225 250 LA-MC EG3 8 30

LA-MC EG3 7 100PG EG3 65 110

LH-MC EG3 18 1565 LH-MC EG3 14 20

LH-MC EG3 8 30LH-MC EG3 7 100

100 PJ EJ3 100 110LA-MC ED 18 30

25 LA-MC ED 10 40LA-MC ED 8 100

PG ED 21.6 125LA-MC EG3 18 30

250 250 42 LA-MC EG3 10 40LA-MC EG3 8 100LH-MC EG3 18 30

65 LH-MC EG3 10 40LH-MC EG3 8 100

PG EG3 65 125100 PJ EJ3 100 12525 PG6 ED 21.6 125

300 300 65 PG6 EG3 65 125100 PJ6 EJ3 100 125

LA-MC ED 18 10025 LA-MC ED 6 125

PG ED 21.6 12542 LA-MC EG3 18 100

400 400 LA-MC EG3 6 125LH-MC EG3 18 100

65 LH-MC EG3 6 125PG EG3 65 125

100 PJ EJ3 100 12525 PG6 ED 21.6 125

450 600 65 PG6 EG3 21.6 125100 PJ6 EJ3 9 12525 PG6 ED 21.6 125

500 600 65 PG6 EG3 65 125100 PJ6 EJ3 100 125

600 600 25 PG (ET1.0)4, PG4 ED 21.6 125

10

Document Number 0100DB0604

Page 233: @Electrical

11© 2006 Schneider Electric. All rights reserved.

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 notesbelow 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

4 The P-Frame Powerpact®

circuit breaker is available with ET1.0 or Micrologic 5.0/6.0 trip units in this size rangePG (ET1.0) = ET1.0 trip unitPG = Micrologic 5.0/6.0 trip unit

5 Values in red are taken from data bulletin 0100DB0501; all other values in this column generated via TCC comparison

6 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 asystem X/R ratio larger than the test X/R ratio of the circuit breaker inquestion, the available fault current equivalent RMS symmetrical duty forcomparison with the values in the tables must be adjusted by a multiplyingfactor. See IEEE Std. 242-2001 (Buff Book), IEEE Std. 1015-1997 (Blue Book)or NEMA AB 3-2001 for details.

Note that this is a consideration for breaker fault duty rather than for selective coordination.

Selectivity Guidelines for Square D® Panelboards

FEEDER SIZE /UPSTREAM (I-LINE)

PANELBOARD CIRCUITBREAKER SIZE

(A)

REQUIREDDOWNSTREAM (NF)

PANELBOARDAMPACITY

(A)

MAXIMUM AVAILABLEFAULT CURRENT ATUPSTREAM (I-LINE)

PANELBOARD(kA RMS Sym.)1

UPSTREAM (I-LINE)PANELBOARD

FEEDER CIRCUITBREAKER TYPE

DOWNSTREAM (NF)PANELBOARD BRANCH

CIRCUIT BREAKER TYPE

MAXIMUM AVAILABLEFAULT CURRENT ATDOWNSTREAM (NF)

PANELBOARD(kA RMS Sym.)1,5

LARGEST POSSIBLEBRANCH CIRCUIT

BREAKER(A)

65 PG (ET1.0)4, PG4 EG3 65 125100 PJ (ET1.0)4, PJ4 EJ3 100 12525 PG4 6 ED 21.6 12565 PG4,6 EG3 21.6 125

MG EG3 65 125100 PJ4,6 EJ3 9 125

MJ EJ3 100 12525 PG (ET1.0)4, PG4 ED 21.6 12565 MG, PG (ET1.0)4, PG4 EG3 65 125

100 MJ, PJ (ET1.0)4, PJ4 EJ3 100 125

700 800

800 800

Molded Case Circuit Breaker Interrupting Rating Test X/RGreater than 20kA 4.910kA - 20kA 3.2Less than 10kA 1.7

Document Number 0100DB0604

Page 234: @Electrical

TABLE IIIAI-Line®/NQOD Panelboard Selective Coordination At 208Y/120V0.1s – 1000s

FEEDER SIZE /UPSTREAM (I-LINE)

PANELBOARD CIRCUITBREAKER SIZE

(A)

REQUIREDDOWNSTREAM (NF)

PANELBOARDAMPACITY

(A)

MAXIMUM AVAILABLEFAULT CURRENT ATUPSTREAM (I-LINE)

PANELBOARD(kA RMS Sym.)1

UPSTREAM (I-LINE)PANELBOARD

FEEDER CIRCUITBREAKER TYPE

DOWNSTREAM (NF)PANELBOARD

BRANCH CIRCUITBREAKER TYPE

LARGEST POSSIBLEBRANCH CIRCUIT

BREAKER (A)

12 © 2006 Schneider Electric. All rights reserved.

HD QO 2010 FA, LX QO 25

PG QO 40100 100 HD QO-VH 20

22 FA2, LX QO-VH 25PG QO-VH 403

HG QH 2065 FH, LX QH 25

PG QH 3010 HD QO 25

110 225 22 HD QO-VH 2565 HG QH 25

HD, LA QO 2510 LX QO 40

PG QO 70HD, LA QO-VH 25

125 225 22 LX QO-VH 30PG QO-VH 603

42 LA QH 2565 HG, LH QH 25

LX, PG QH 3010 HD QO 25

JD, LA, LX, PG QO 70HD QO-VH 25

150 225 22 JD, LA QO-VH 30LX QO-VH 403

PG QO-VH 703

42 LA QH 3065 HG QH 25

JG, LH, LX, PG QH 3010 JD, LA, LX, PG QO 70

JD, LA QO-VH 403

175 225 22 LX QO-VH 503

PG QO-VH 703

42 LA QH 3065 JG, LH, LX, PG QH 30

JD, LA, QO 7010 LX QO 806

LA-MC, PG QO 1006

200 225 JD, LA, LX, QO-VH 503

22 PG QO-VH 803

LA-MC QO-VH 1003

42 LA, LA-MC QH 3065 JG, LH, LH-MC, LX, PG QH 3010 JD, LA, LX, PG QO 1006

LA-MC QO 1257

JD, LA QO-VH 503

225 225 22 LX QO-VH 603

PG QO-VH 803

LA-MC QO-VH 1253

42 LA, LA-MC QH 3065 JG, LH, LH-MC, LX, PG QH 3010 JD, LA, LX, PG QO 1006

LA-MC QO 1257

JD, LA QO-VH 603

250 400 22 LX QO-VH 803

PG QO-VH 1103

LA-MC QO-VH 1503

42 LA, LA-MC QH 3065 JG, LH, LH-MC, LX, PG QH 3010 LA, MG, LX QO 1006

PG5 QO 1257

LA QO-VH 903

300 400 22 MG, LX QO-VH 1003

PG5 QO-VH 1253

42 LA QH 3065 LH, MG, LX, PG5 QH 30

350 400 10 LA QO 1006

Selectivity Guidelines for Square D® PanelboardsDocument Number

0100DB0604

Page 235: @Electrical

1 Available fault currents are based upon system X/R ratios less than or equal to the circuit breaker test X/R ratio. Seethe 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)

4 The P-Frame Powerpact® circuit breaker is available with ET1.0 or Micrologic 5.0/6.0 trip units in this size rangePG (ET1.0) = ET1.0 trip unitPG = Micrologic 5.0/6.0 trip unit

5 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)

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 coordinatesup to 100A)

X/R Ratio Adjustment:All available fault currents are given in RMS symmetrical amperes. For asystem X/R ratio larger than the test X/R ratio of the circuit breaker inquestion, the available fault current equivalent RMS symmetrical duty forcomparison with the values in the tables must be adjusted by a multiplyingfactor. See IEEE Std. 242-2001 (Buff Book), IEEE Std. 1015-1997 (Blue Book)or NEMA AB 3-2001 for details.

Note that this is a consideration for breaker fault duty rather than for selective coordination.

FEEDER SIZE /UPSTREAM (I-LINE)

PANELBOARD CIRCUITBREAKER SIZE

(A)

REQUIREDDOWNSTREAM (NQOD)

PANELBOARDAMPACITY

(A)

MAXIMUM AVAILABLEFAULT CURRENT ATUPSTREAM (I-LINE)

PANELBOARD(kA RMS Sym.)1

UPSTREAM (I-LINE)PANELBOARD

FEEDER CIRCUIT BREAKER TYPE

DOWNSTREAM (NQOD)PANELBOARD

BRANCH CIRCUITBREAKER TYPE

LARGEST POSSIBLEBRANCH CIRCUIT

BREAKER (A)

MG, LX QO 1257

350 400 22 LA, MG, LX QO-VH 1503

42 LA QH 3065 LH, MG, LX QH 3010 LA, LA-MC, MG, LX, PG QO 1257

400 400 22 LA, LA-MC, MG, LX, PG QO-VH 1503

42 LA, LA-MC QH 3065 LH, LH-MC, MG, LX, PG QH 3010 LC, MG, LX, PG5 QO 1257

450 600 22 LC, MG, LX, PG5 QO-VH 1503

65 LC, MG, LX, PG5 QH 3010 LC, MG, LX, PG5 QO 1257

500 600 22 LC, MG, LX, PG5 QO-VH 1503

65 LC, MG, LX, PG5 QH 3010 LC, MG, LX, PG

(ET1.0)4, PG4 QO 1257

600 600 22 LC, MG, LX, PG (ET1.0)4, PG4 QO-VH 1503

65 LC, MG, LX, PG (ET1.0)4, PG4 QH 30

13© 2006 Schneider Electric. All rights reserved.

Selectivity Guidelines for Square D® Panelboards

Molded Case Circuit Breaker Interrupting Rating Test X/RGreater than 20kA 4.910kA - 20kA 3.2Less than 10kA 1.7

Document Number 0100DB0604

Page 236: @Electrical

FEEDER SIZE /UPSTREAM (I-LINE)

PANELBOARD CIRCUITBREAKER SIZE

(A)

REQUIREDDOWNSTREAM (NQOD)

PANELBOARD AMPACITY(A)

MAXIMUM AVAILABLEFAULT CURRENT ATUPSTREAM (I-LINE)

PANELBOARD(kA RMS Sym.)1

UPSTREAM (I-LINE)PANELBOARD

FEEDER CIRCUITBREAKER TYPE

DOWNSTREAM (NQOD)PANELBOARD

BRANCH CIRCUITBREAKER TYPE

MAXIMUM AVAILABLEFAULT CURRENT AT

DOWNSTREAM (NQOD)PANELBOARD1.8

LARGEST POSSIBLEBRANCH CIRCUIT

BREAKER(A)

10 HD QO 1.3 20PG QO 10 40HD QO-VH 1.3 20

100 100 22 PG QO-VH 21.6 403

PJ QO-VH 22 403

HG QH 1.3 2065 FH QH 0.9 25

PG QH 65 3010 HD QO 1.3 25

110 225 22 HD QO-VH 1.3 2565 HG QH 1.3 2510 HD QO 1.3 25

PG QO 10 70HD QO-VH 1.3 25

125 225 22 LA QO-VH 3.2 (2P ONLY) 25PG QO-VH 21.6 603

PJ QO-VH 22 603

HG QH 1.3 2565 LH QH 65 (1P), 3.2 (2P, 3P) 25

PG QH 65 30HD QO 1.3 25

10 JD QO 2.3 70PG QO 10 70HD QO-VH 1.3 25JD QO-VH 2.3 30

150 225 22 LA QO-VH 3.2 (2P ONLY) 30PG QO-VH 21.6 703

PJ QO-VH 22 703

HG QH 1.3 2565 JG QH 2.4 30

LH QH 65 (1P), 3.2 (2P, 3P) 30PG QH 65 30

10 JD QO 2.3 70PG QO 10 70JD QO-VH 2.3 403

22 LA QO-VH 3.2 (2P ONLY) 403

175 225 PG QO-VH 21.6 703

PJ QO-VH 22 703

JG QH 2.4 3065 LH QH 65 (1P), 3.2 (2P, 3P) 30

PG QH 65 30JD QO 2.3 70

LA-MC QO 18 (1P, 2P), 16 (3P) 15LA-MC QO 18 (1P, 2P), 10 (3P) 20LA-MC QO 7 (1P), 10 (2P), 6.5

10 (3P) 30LA-MC QO 7 (1P, 2P), 6 (3P) 40LA-MC QO 6 (1P, 2P), 5.5 (3P) 50LA-MC QO 5 (1P, 3P), 6 (2P) 70LA-MC QO 5 1006

PG QO 10 1006

JD QO-VH 2.3 503

200 225 LA QO-VH 3.2 (2P ONLY) 503

PG QO-VH 21.6 803

PJ QO-VH 22 803

LA-MC QO-VH 22 (1P, 2P), 16 (3P) 1522 LA-MC QO-VH 22 (1P, 2P), 10 (3P) 20

LA-MC QO-VH 7 (1P), 10 (2P), 6.5(3P) 30

LA-MC QO-VH 7 (2P), 6 (3P) 403

LA-MC QO-VH 6 (2P), 5.5 (3P) 503

LA-MC QO-VH 6 (2P), 5 (3P) 703

LA-MC QO-VH 5 1003

42 LA-MC QH 3.4 30JG QH 2.4 30

65 LH QH 65 (1P), 3.2 (2P, 3P) 30LH-MC QH 3.4 30

PG QH 65 30225 225 10 JD QO 2.3 1006

14 © 2006 Schneider Electric. All rights reserved.

Selectivity Guidelines for Square D® Panelboards

Table IIIBI-Line®/NQOD Panelboard Selective Coordination At 208Y/120V0.01s – 1000s

Document Number 0100DB0604

Page 237: @Electrical

FEEDER SIZE /UPSTREAM (I-LINE)

PANELBOARD CIRCUITBREAKER SIZE

(A)

REQUIREDDOWNSTREAM (NQOD)

PANELBOARD AMPACITY(A)

MAXIMUM AVAILABLEFAULT CURRENT ATUPSTREAM (I-LINE)

PANELBOARD(kA RMS Sym.)1

UPSTREAM (I-LINE)PANELBOARD

FEEDER CIRCUITBREAKER TYPE

DOWNSTREAM (NQOD)PANELBOARD

BRANCH CIRCUITBREAKER TYPE

MAXIMUM AVAILABLEFAULT CURRENT AT

DOWNSTREAM (NQOD)PANELBOARD1,8

LARGEST POSSIBLEBRANCH CIRCUIT

BREAKER(A)

PG QO 10 1006

LA-MC QO 18 15LA-MC QO 18 (1P, 2P), 16 (3P) 20LA-MC QO 11 (1P), 18 (2P), 8

(3P) 30LA-MC QO 10 (1P, 2P), 7.5 (3P) 40

10 LA-MC QO 10 (1P, 2P), 7 (3P) 50LA-MC QO 8 (1P), 10 (2P), 6.5

(3P) 60LA-MC QO 7 (1P), 10 (2P), 6

(3P) 70LA-MC QO 8 (2P), 6 (3P) 806

LA-MC QO 6 1006

LA-MC QO 3.825 1257

JD QO-VH 2.3 503

225 225 LA QO-VH 3.2 (2P ONLY) 503

PG QO-VH 21.6 803

PJ QO-VH 22 803

LA-MC QO-VH 22 (1P, 2P), 18 (3P) 15LA-MC QO-VH 22 (1P, 2P), 16 (3P) 20

22 LA-MC QO-VH 11 (1P), 22 (2P), 8(3P) 30

LA-MC QO-VH 18 (2P), 7.5 (3P) 403

LA-MC QO-VH 18 (2P), 7 (3P) 503

LA-MC QO-VH 13 (2P), 6.5 (3P) 603

LA-MC QO-VH 10 (2P), 6 (3P) 703

LA-MC QO-VH 8 (2P), 6 (3P) 803

LA-MC QO-VH 6 1003

LA-MC QO-VH 3.825 1253

42 LA-MC QH 3.825 30JG QH 2.4 30

65 LH QH 65 (1P), 3.2 (2P, 3P) 30LH-MC QH 3.825 30

PG QH 65 30JD QO 2.3 1006

PG QO 10 1006

LA-MC QO 18 20LA-MC QO 18 (1P, 2P), 14 (3P) 30

10 LA-MC QO 10 40LA-MC QO 10 (1P, 2P), 9 (3P) 50LA-MC QO 10 (1P, 2P), 8 (3P) 60LA-MC QO 10 (1P, 2P), 7.5 (3P) 70LA-MC QO 10 (2P), 7.5 (3P) 1006

LA-MC QO 4.25 1257

JD QO-VH 2.3 603

LA QO-VH 3.2 (2P ONLY) 603

250 400 PG QO-VH 21.6 1103

PJ QO-VH 22 1103

LA-MC QO-VH 22 (1P, 2P), 18 (3P) 2022 LA-MC QO-VH 22 (1P, 2P), 14 (3P) 30

LA-MC QO-VH 18 (2P), 10 (3P) 403

LA-MC QO-VH 18 (2P), 9 (3P) 503

LA-MC QO-VH 13 (2P), 8 (3P) 603

LA-MC QO-VH 11 (2P), 7.5 (3P) 803

LA-MC QO-VH 10 (2P), 7.5 (3P) 1003

LA-MC QO-VH 4.25 1503

42 LA-MC QH 4.25 30JG QH 2.4 30

65 LH QH 65 (1P), 3.2 (2P, 3P) 30LH-MC QH 4.25 30

PG QH 65 3010 PG5 QO 10 1257

LA QO-VH 3.2 (2P ONLY) 903

22 MG QO-VH 3.6 (3P ONLY) 1003

300 400 PG5 QO-VH 21.6 1253

LH QH 65 (1P), 3.2 (2P, 3P) 3065 MG QH 65 (1P), 3.6 (2P, 3P) 30

PG5 QH 65 3022 LA QO-VH 3.2 (2P ONLY) 1503

350 400 MG QO-VH 3.6 (3P ONLY) 1503

65 LH QH 65 (1P), 3.2 (2P, 3P) 30MG QH 65 (1P), 3.6 (2P, 3P) 30

LA-MC QO 18 30400 400 10 LA-MC QO 10 1006

LA-MC QO 6 1257

PG QO 10 1257

Selectivity Guidelines for Square D® Panelboards

15© 2006 Schneider Electric. All rights reserved.

Document Number 0100DB0604

Page 238: @Electrical

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 notebelow for additional information

2 480V-rated3 2 Pole or 3 Pole only. QO-VH 1 Pole is available up to 30A (and coordinates up to 30A)4 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 unitPG = Micrologic 5.0/6.0 trip unit

5 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)

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)8 Values in red are taken from data bulletin 0100DB0501; all other values in this column generated via TCC comparison

X/R Ratio Adjustment:All available fault currents are given in RMS symmetrical amperes. For asystem X/R ratio larger than the test X/R ratio of the circuit breaker inquestion, the available fault current equivalent RMS symmetrical duty forcomparison with the values in the tables must be adjusted by a multiplyingfactor. See IEEE Std. 242-2001 (Buff Book), IEEE Std. 1015-1997 (Blue Book)or NEMA AB 3-2001 for details.

Note that this is a consideration for breaker fault duty rather than for selective coordination.

Selectivity Guidelines for Square D® Panelboards

FEEDER SIZE /UPSTREAM (I-LINE)

PANELBOARD CIRCUITBREAKER SIZE

(A)

REQUIREDDOWNSTREAM (NQOD)

PANELBOARD AMPACITY(A)

MAXIMUM AVAILABLEFAULT CURRENT ATUPSTREAM (I-LINE)

PANELBOARD(kA RMS Sym.)1

UPSTREAM (I-LINE)PANELBOARD

FEEDER CIRCUITBREAKER TYPE

DOWNSTREAM (NQOD)PANELBOARD BRANCH

CIRCUIT BREAKER TYPE

MAXIMUM AVAILABLEFAULT CURRENT AT

DOWNSTREAM (NQOD)PANELBOARD

(kA RMS Sym.)1,8

LARGEST POSSIBLEBRANCH CIRCUIT

BREAKER(A)

LA QO-VH 3.2 (2P ONLY) 1503

LA-MC QO-VH 22 (1P, 2P), 18 (3P) 3022 LA-MC QO-VH 22 (2P), 18 (3P) 1003

LA-MC QO-VH 6 1503

MG QO-VH 3.6 (3P ONLY) 1503

400 400 PG QO-VH 21.6 1503

42 LA-MC QH 6 30LH QH 65 (1P), 3.2 (2P, 3P) 30

65 LH-MC QH 6 30MG QH 65 (1P), 3.6 (2P, 3P) 30PG QH 65 30

10 PG5 QO 10 1257

22 MG QO-VH 3.6 (3P ONLY) 1503

450 600 PG5 QO-VH 21.6 1503

65 MG QH 65 (1P), 3.6 (2P, 3P) 30PG5 QH 65 30

10 PG5 QO 10 1257

22 MG QO-VH 3.6 (3P ONLY) 1503

500 500 PG5 QO-VH 21.6 1503

65 MG QH 65 (1P), 3.6 (2P, 3P) 30PG5 QH 65 30

10 PG (ET1.0)4, PG4 QO 10 1257

22 MG QO-VH 5.4 (3P ONLY) 1503

600 600 PG (ET1.0)4, PG4 QO-VH 21.6 1503

65 MG QH 65 (1P), 5.4 (2P, 3P) 30PG (ET1.0)4, PG4 QH 65 30

16 © 2006 Schneider Electric. All rights reserved.

Molded Case Circuit Breaker Interrupting Rating Test X/RGreater than 20kA 4.910kA - 20kA 3.2Less than 10kA 1.7

Document Number 0100DB0604

Page 239: @Electrical

1

Data Bulletin0600DB0601

2/2007Cedar Rapids, IA, USA

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 simplified-coordination 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 articles—for 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):

fuse—An overcurrent protective device with a circuit-opening fusible part that is heated and severed by the passage of overcurrent through it.

circuit breaker—A 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.

Page 240: @Electrical

A Comparison of Circuit Breakers and Fuses for Low-Voltage Applications 0600DB0601Data Bulletin 2/2007

© 2007 Schneider Electric All Rights Reserved2

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 ratings—10 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 LVPCBs—they 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 question—several 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

Page 241: @Electrical

0600DB0601 A Comparison of Circuit Breakers and Fuses for Low-Voltage Applications2/2007 Data Bulletin

© 2007 Schneider Electric All Rights Reserved 3

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 levels—if 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-to-phase (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 themselves—equipment 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.

Page 242: @Electrical

A Comparison of Circuit Breakers and Fuses for Low-Voltage Applications 0600DB0601Data Bulletin 2/2007

© 2007 Schneider Electric All Rights Reserved4

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 protector—MCP 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 100–125% of the motor's actual load level (not the nameplate rating) to provide better levels of protection against damage resulting from single-phasing [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.

Page 243: @Electrical

0600DB0601 A Comparison of Circuit Breakers and Fuses for Low-Voltage Applications2/2007 Data Bulletin

© 2007 Schneider Electric All Rights Reserved 5

D. Component Protection 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 non-current-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 applied—and 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.

Page 244: @Electrical

A Comparison of Circuit Breakers and Fuses for Low-Voltage Applications 0600DB0601Data Bulletin 2/2007

© 2007 Schneider Electric All Rights Reserved6

E. Arc-Flash Protection 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 factors—the 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 fault—i.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-of-the-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 when possible

• Evaluate devices at the actual system fault current levels

• Evaluate adjustable-trip circuit breakers at their chosen settings

Evaluate Specific Devices The IEEE 1584 standard contains three basic calculation models that can be used to determine arc-flash hazard levels—an 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.

Page 245: @Electrical

0600DB0601 A Comparison of Circuit Breakers and Fuses for Low-Voltage Applications2/2007 Data Bulletin

© 2007 Schneider Electric All Rights Reserved 7

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 “worst-case” arc-flash levels allowed by any example of a circuit breaker within a given class—e.g., 100–400 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 LVPCBs—the 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.

The “LVPCB w/ST” and “LVPCB w/INST” curves are based on the IEEE 1584 simplified equations for low-voltage power circuit breakers with short-time 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.

Figure 1: Incident Energy vs. Bolted Fault Current for 2000 A Circuit Breaker’s Simplified Equations vs. Actual Data

0

30

60

90

120

0 20 40 60 80 100 120

Bolted Fault Current (kA)

Inci

dent

Ene

rgy

(cal

/cm

^2)

NW-LF NW-L LVPCB w/INST LVPCB w/ST

Page 246: @Electrical

A Comparison of Circuit Breakers and Fuses for Low-Voltage Applications 0600DB0601Data Bulletin 2/2007

© 2007 Schneider Electric All Rights Reserved8

What is the system fault current range? 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 protection—allowing 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.

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.

Figure 2: Incident Energy vs. Bolted Fault Current for 400 A Circuit Breakers and 400 A Class RK-1 Fuses.

0.0

0.4

0.8

1.2

1.6

2.0

0 20 40 60 80 100 120

Bolted Fault Current (kA)

Inci

dent

Ene

rgy

(cal

/cm

^2)

400A LH 400A LC 400A RK-1 Fuse

Page 247: @Electrical

0600DB0601 A Comparison of Circuit Breakers and Fuses for Low-Voltage Applications2/2007 Data Bulletin

© 2007 Schneider Electric All Rights Reserved 9

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 arc-flash 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 location—should 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].

Figure 3: Incident Energy Comparison for 1600 A Protective Devices

0

5

10

15

20

25

30

0 20 40 60 80 100 120

Bolted Fault Current (kA)

Inci

dent

Ene

rgy

(cal

/cm

^2)

NW-LF NW-H 1600L Fuse

Page 248: @Electrical

A Comparison of Circuit Breakers and Fuses for Low-Voltage Applications 0600DB0601Data Bulletin 2/2007

© 2007 Schneider Electric All Rights Reserved10

IV. Selective Coordination Selective coordination of overcurrent protective devices is required to ensure that two somewhat mutually-exclusive goals are met—faults 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.

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 system—and 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 breaker—and only the Feeder-C circuit breaker—should 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 system—namely, Panel-A, Panel-B, and the Chiller Motor—and 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 other—that 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

Figure 4: Sample One-Line Diagram

Utility

S

P Utility Transformer

Switchboard

Feeder-A

SwitchboardMain

Feeder-B

Panel-B

Panel-A

Feeder-C Chiller

Panel-C

Chiller Motor

S

PTransformer-A

Panel-A Main

Page 249: @Electrical

0600DB0601 A Comparison of Circuit Breakers and Fuses for Low-Voltage Applications2/2007 Data Bulletin

© 2007 Schneider Electric All Rights Reserved 11

of the power system—namely, 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 legally-required 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 devices—ideally, 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 coordination—for 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.

Page 250: @Electrical

A Comparison of Circuit Breakers and Fuses for Low-Voltage Applications 0600DB0601Data Bulletin 2/2007

© 2007 Schneider Electric All Rights Reserved12

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 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.

Figure 5: Time-Current Characteristics for Feeder-A and Panel-A Main Circuit Breakers.

0.5 1 10 100 1K 10K0.01

0.10

1

10

100

1000

Current in Amperes

Tim

e in

Sec

onds

PANEL 'A' MAIN

FDR 'A'

PANEL 'A' MAIN

FDR 'A'

Page 251: @Electrical

0600DB0601 A Comparison of Circuit Breakers and Fuses for Low-Voltage Applications2/2007 Data Bulletin

© 2007 Schneider Electric All Rights Reserved 13

Figure 6 shows the time-current characteristics of the Feeder-B and Feeder-C circuit breakers in Figure 4.

As shown in the plot, the two devices—a 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 time-current 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.

Figure 6: Time-Current Characteristics for the Feeder-B and Feeder-C Circuit Breakers.

0.5 1 10 100 1K 10K0.01

0.10

1

10

100

1000

Current in Amperes

Tim

e in

Sec

onds

FDR 'C'

FDR 'B'

FDR 'C'

FDR 'B'

Page 252: @Electrical

A Comparison of Circuit Breakers and Fuses for Low-Voltage Applications 0600DB0601Data Bulletin 2/2007

© 2007 Schneider Electric All Rights Reserved14

For example, see Figure 7, which shows the time-current characteristics for two Square D® thermal-magnetic circuit breakers—an 800 A MJ and a 125 A EJB, both at 208 V.

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.

Figure 7: Trip Curves for 800 A MJ and 125A EJB.

0.5 1 10 100 1K 10K0.01

0.10

1

10

100

1000

Current in Amperes

Tim

e in

Sec

onds

800A MJ

125A EJB

800A MJ

125A EJB

Page 253: @Electrical

0600DB0601 A Comparison of Circuit Breakers and Fuses for Low-Voltage Applications2/2007 Data Bulletin

© 2007 Schneider Electric All Rights Reserved 15

V. Reliability 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 40°C. They are actually capable of carrying higher-than-rated currents for ambient levels below 24°C, 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 25°C, 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 (high-frequency 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].

Page 254: @Electrical

A Comparison of Circuit Breakers and Fuses for Low-Voltage Applications 0600DB0601Data Bulletin 2/2007

© 2007 Schneider Electric All Rights Reserved16

VII. Renewability 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” process—the fusible link can no longer carry current and must be replaced. For non-renewable fuses—on low-voltage systems, this encompasses all but certain types of Class H fuses—this 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 re-closed 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 wise—particularly if any damage or stress is seen when the circuit breaker is inspected—to 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 re-energized. (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.

Page 255: @Electrical

0600DB0601 A Comparison of Circuit Breakers and Fuses for Low-Voltage Applications2/2007 Data Bulletin

© 2007 Schneider Electric All Rights Reserved 17

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.

Page 256: @Electrical

A Comparison of Circuit Breakers and Fuses for Low-Voltage Applications 0600DB0601Data Bulletin 2/2007

© 2007 Schneider Electric All Rights Reserved18

VIII. Flexibility A wide variety of circuit breakers are available—from relatively basic molded-case circuit breakers to the “top of the line” low-voltage power circuit breakers—with 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:

— 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.

Page 257: @Electrical

0600DB0601 A Comparison of Circuit Breakers and Fuses for Low-Voltage Applications2/2007 Data Bulletin

© 2007 Schneider Electric All Rights Reserved 19

IX. References [1] NFPA 70-2005, National Electrical Code, National Fire Protection Association, Quincy, MA.

[2] IEEE Std. 1015-1997, IEEE Recommended Practice for Applying Low-Voltage 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] SPD—Selecting 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: Arc-flash 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: http://www.us.squared.com

[10] Square D Data Bulletin 0100DB0403, Enhancing Short Circuit Selective Coordination with Low Voltage Circuit Breakers. Available: http://www.us.squared.com

[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

Page 258: @Electrical

A Comparison of Circuit Breakers and Fuses for Low-Voltage Applications 0600DB0601Data Bulletin 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.

© 2007 Schneider Electric All Rights Reserved

Schneider Electric USA3700 Sixth St SWCedar Rapids, IA 52404 USA1-888-SquareD (1-888-778-2733)www.us.SquareD.com