Home >> Part 2 >> Building Electrical Systems

Building Electrical Systems

This chapter is current as of November 14, 2012


Three issues dominate the design, construction and operation of electrical systems in educational facilities:

  1. Fire safety.   The modern way of fighting fires is with electrical power present so parallel and auxiliary systems are conceived to ensure occupant egress and to provide power to fire fighters.  One of the leading practice documents is The Life Safety Code© (NFPA 101) which, along with documents prepared by the International Code Council (ICC), are either adopted as a whole or used as resource documents for variants in the building codes promulgated by individual states.   Both NFPA and ICC documents refer to the National Electrical Code© (The NEC, or NFPA 70) for prescriptive requirements for actually wiring the emergency lighting systems, elevator and fire pump power.
  2. Energy efficiency.  Many organizations are addressing energy standards; the American Society of Heating, Refrigeration and Air-Conditioning Engineers (ASHRAE) is one of the leaders establishing national standards to meet federal energy efficiency objectives asserted by the Environmental Protection Agency and the Department of Energy.  ASHRAE has traditionally prepared leading practice documents for the energy efficiency of installed HVAC mechanical systems but in recent years has moved into lighting efficiency.  Since the NEC is principally a fire safety document — with wiring rules that have evolved for the safe use of electricity in the built environment — many have observed that the NEC is out of step with the national effort to conserve energy.  Wire sizing rules — and particularly the rules for sizing transformers — may result in the production of waste heat that must be removed by the systems classically governed by ASHRAE leading practice documents.
  3. Electrician's safety. The rise of leading practices in limiting the effect of electrical arc flash originated in the early conception of the Occupational Safety and Health Administration (OSHA).  The NFPA prepared a document entitled, NFPA 70E - Electrical Safety in the Workplace, for use by OSHA in promulgating rules for reducing fatalities among electrical professionals working on live equipment.  In recent years, many in the electrical industry have tried to translate the rules of NFPA 70E — an operation and maintenance document — into NFPA 70 — a design and installation document.  The effort has not been successful so far.  One of the reasons, which facility managers may appreciate, is that design and construction budgets are rather different from operations and maintenance budgets.

In this chapter we shall illustrate the broad contours of these issues with a numerical example of a typical building power system.  The numerical example should put into relief the parts of a power system design that is focused on fire safety; how electrical engineers conform to the thermal limits of wiring systems.  The numbers shall show how the fire safety requirements of the NEC may not agree with the energy efficiency objectives of ASHRAE.  Finally, the numbers will be run for a typical flash hazard calculation to show how the least expensive moment to meet OSHA’s workplace safety requirements is when a building’s internal electrical distribution system is commissioned and accepted. 


Understanding Electric Power

The bulk electrical distribution grid of most college and university buildings is 13.2-kV, 4,160-V, or 2,400-V three-phase alternating current (AC).  When the system is not part of what is considered to be a premises wiring system it is governed by the safety rules of the National Electric Safety Code (NESC) developed by the Institute of Electronic and Electrical Engineers (IEEE).   The NESC asserts leading safety practices for organizations that work with voltages ranging from 2.4kV to 768kV; typically public power or investor-owned utilities.  When these higher voltages are present on a college or university campus then Article 490 of the NEC is the leading practice document.  The sample building system of this chapter covers low-voltage (less than 600 V) building distribution systems.

For building electrical systems, usually one transformer is used that has 480-V three-phase secondary windings for motor control centers and 277-V single-phase windings for fluorescent lights. A second transformer is used that has 208-V three-phase windings for the three-phase 208-V loads and 120-V single-phase windings for general purpose power outlets.

From the transformer, the power goes to the main distribution panels, sometimes known as circuit breaker panels. Standard distribution panels have four, six, eight, twelve, sixteen, eighteen, or twenty-four circuits, each equipped with circuit breakers or fuses. A main circuit breaker is required for the main distribution panels. The power from the main panel is distributed to secondary panels and eventually to the loads.

A good design allows for at least 25 percent system growth in distribution panels and includes a directory of the circuits and the levels served. If panels are in public areas, they should be inaccessible to all but authorized personnel.


Electrical Safety and Flash Hazard Analysis

Section 110.16 of the National Electrical Code (NFPA 70) requires that all equipment likely to be serviced while energized receive a mark that indicates flash hazard.  This is a requirement for installation only and may be a pre-condition for a certification of occupancy by the electrical inspector.   NFPA 70B Electrical Safety in the Workplace indicates that the incident energy level be marked on all equipment likely to be serviced while energized.  If occupational safety workgroups enforce NFPA 70E as a matter of workplace safety then extensive system short-circuit studies will be required.  The least expensive method of getting the system analyzed for fault current is during the design and construction phase in which all of the building system is modeled. 

While an increasing proportion of end-use equipment runs on direct current, fluorescent lighting, HVAC equipment controlled by variable frequency drives, computers and telecommunications equipment, power is delivered in AC form to transformation equipment; forcing engineers to switch between DC and AC mathematics.

The formula, Power (watts) = amperes x volts, or P = V x I applies to direct current circuits. It also works on some AC loads such as incandescent light bulbs, quartz heaters, electric range heating elements, and other equipment of this general nature. However, when the loads involve a characteristic called inductance, the formula has to be altered to include a new term called “power factor”?.  

The formula for power in three-phase AC circuits becomes:


P = 1.732 x VL x IL x PF                          Eq.  26.1


Where P = total three phase power, in watts

VL = line-to-line voltage in volts

IL = line current in amperes

PF = power factor of the system, the cosine of the phase shift between voltage and current

The term, power factor, is always involved in applications where AC power is used and inductive magnetic elements exist in the circuit. Inductive elements are magnetic devices such as solenoid coils, motor windings, transformer windings, fluorescent lamp ballasts, and similar equipment that have magnetic components as part of their design.

Looking at the electrical flow into this type of device, we would find that there are, in essence, two components. One portion is absorbed and utilized to do useful work. This portion is called the real power. The second portion is literally borrowed from the power company and used to magnetize the magnetic portion of the circuit, called the reactive power. Due to the reversing nature of AC power, this borrowed power is subsequently returned to the power system when the AC cycle reverses. This borrowing and returning occurs on a continuous basis.


Electric Bill Showing Watts and Vars


Power factor bill.jpg


Power factor then becomes a measurement of the amount of real power that is used, divided by the total amount of power, both borrowed and used. Values for power factor will range from zero to 1.0. If all the power is borrowed and returned with none being used, the power factor would be zero. If on the other hand, all of the power drawn from the power line is utilized and none is returned, the power factor becomes 1.0. In the case of electric heating elements, incandescent light bulbs, etc., the power factor is 1.0. In the case of electric motors, the power factor is variable and changes with the amount of load that is applied to the motor. Thus, a motor running on a work bench, with no load applied to the shaft, will have a low power factor (perhaps .1 or 10%), and a motor running at full load, connected to a pump or a fan might have a relatively high power factor (perhaps .88 or 88%). Between the no load point and the full load point, the power factor increases steadily with the horsepower loading that is applied to the motor. These trends can be seen on the typical motor performance data plots which are shown in Figure 1.

Figure 1: Typical Motor Performance Data Plots


In other words, line voltage multiplied by line current results in apparent power, or VA. Apparent power can be graphically broken into two components: real power (watts) and reactive power (or VARs) (Figure 2). Notice that for the same apparent power, real power varies inversely with the size of the angle. Moreover if the angle is positive, it is called  leading power factor; and if the angle is negative, it is called  lagging power factor. In most power systems, the current lags relative to the voltage principally because of motor load, but also, to some extent, because of other loads. Therefore, more current is required to provide a given amount of real power if the power factor is less than unity.  A low power factor implies lower system efficiency. Most electrical utilities include a power factor penalty in their rate structure to discourage customers with low power factors. The penalties usually start when a power factor is less than 0.95.

Figure 2: Real Power vs. Reactive Power



In an electric bill, commercial, industrial or residential, the basic unit of measurement is the kilowatt hour. This is a measure of the amount of energy that is delivered. In many respects, the kilowatt hour could be compared to a ton of coal, a cubic foot of natural gas, or a gallon of gasoline, in that it is a basic energy unit. The kilowatt hour is not directly related to amperes, and at no place on an electric bill will you find any reference to the amperes that have been utilized. It is vitally important to note this distinction. You are billed for kilowatt hours: you do not necessarily pay for amperes.In addition, a close-to-unity power factor is important to reduce system loss, transformer size, cable size, and power cost and to help stabilize the system voltage.

Power factor improvement is accomplished in two ways: by operating equipment at unity power factor and by using auxiliary devices to supply the magnetizing power, or kiloVARs, needed by the load. Equipment that operates at unity power factor are incandescent lamps, resistance heaters and unity power factor synchronous motors.

Capacitors are the best choice in most applications because they have no moving parts and the losses are less than I percent. Capacitors can be located at the loads, in the substations, or on the line. Power factor correction on the line is generally less costly, but it is better to position the power factor correction capacitor close to the device causing the low power factor. The capacity needed to improve power factor is calculated by the general formula shown in Figure 3.

Figure 3: Formula For Imporving Power Factor


However, it is usually not necessary to go through laborious calculations. Multiplication factors can be determined by using a power factor correction table.

To illustrate the system efficiency improvement produced by a power factor improvement, consider a 100-HP, three-phase, 208-V motor that has a power factor of 80 percent. If capacitors are added to improve the power factor to 85 percent, 90 percent, 95 percent, or 100 percent, line losses will drop by 12 percent, 21 percent, 30 percent, and 35 percent, respectively. If the motor is partially loaded, the power factor is even lower, and adding corrective capacitors will reduce line losses even more. Overcompensating for power factor should be avoided because in addition to wasting funds, it results in higher-than-system localized voltages, which can damage certain equipment.

For more information, the reader is referred to the recommended bibliography at the end of this chapter. 


Power System Modeling

Among electrical professionals, whether an electrical system operates at low, medium or high voltage it is understood in context differently by design practitioners versus operations practitioners.  Many have asked for these terms to be harmonized.  To do this would be to bring together two bodies of expertise that have grown in parallel but whom practitioners keep at a distance.  Worldwide IEC standards complicate the situation.  For practitioners, use of the terms becomes  idiomatic. 

Much of the cost of making power systems safe lies in keeping up to date circuit diagrams.  Calls for standard methods between the utility industry and the building industry have been a long conversation and are beyond the scope of this chapter. There is universal acceptance of a normalization method of analyzing power circuits, the "per-unit" method. In electrical engineering, in the field of power transmission, a “per-unit” system is the expression of system quantities as fractions of a defined base unit quantity.  This type of modeling technique can be used in building power system models also.

Calculations are simplified because quantities expressed as per-unit are the same regardless of the voltage level. Similar types of apparatus will have impedance, voltage drop and losses that are the same when expressed as a per-unit fraction of the equipment rating, even if the unit size varies widely. Conversion of per-unit quantities to volts, ohms, or amperes requires a knowledge of the base to which the per-unit quantities were referenced.  A per-unit system provides units for power, voltage, current, impedance, and admittance. Only two of these are independent, usually power and voltage. All quantities are specified as multiples of selected base values.

For example, the base power might be the rated power of a transformer, or perhaps an arbitrarily selected power which makes power quantities in the system more convenient. The base voltage might be the nominal voltage of a bus. Different types of quantities are labeled with the same symbol (p.u.); it should be clear from context whether the quantity is a voltage, current, etc.

Per-unit is used primarily in distribution and transmission studies but the technique can also be applied to building power systems.  Because parameters of transformers and machines (electric motors and electrical generators) are often specified in terms of per-unit, it is important for all power engineers to be familiar with the concept.

The Per Unit System illustrated here is a normalization procedure which provides a mathematical basis for analyzing power networks with relative ease and convenience. In addition, when various quantities are expressed in per unit (pu) or percent values, they usually convey a message. For example, if a bus voltage is 0.98 pu, it means that this value is 98 % of the nominal or base value which could be at any level in the network. It also immediately conveys a message that the value is an acceptable one. On the contrary, if the voltage value is 1.08, then it immediately conveys that the value is higher than the acceptable level of 1.05 pu. Similar conclusions can be drawn for other quantities such as current, power and impedance. The idea here is to express various variables as a fraction of their corresponding base (fixed) variables.

For example, by arbitrarily assuming the source is 100%, instead of, say 100 volts, we mean that the source voltage is at 100% of its rated value, whether it is 120V or 120 kV.  Similarly, we can designate the total impedance of the system under full load conditions as 100% (instead of 100 ohms).   With 1.0 per unit voltage applied and with 1.0 per unit impedance, the current is 1.0 per unit.  The value of the single impedance that, if inserted in the system with rated voltage applied, allows rated current to flow is called the base impedance.  Therefore,

Base impedance  =             
Eq. 26.2

Per unit value  =                                 
Eq. 26.3

The short circuit current from Ohm’s Law is:

Per unit ISC =                                                           Eq. 26.4

Where Z is the equivalent  system per unit impedance between the source and the point of fault.  The actual fault current in amperes will then be:

ISC(A) = (rated current) x (per unit ISC)                                    Eq. 26.5

If, for example, you have a 100V source, with a 5 ohm impedance which results in a current of 100/100 = 1.0 ampere then the short circuit current is:

ISC =   = 20.0 p.u.

This result agrees with the non-per unit method.  The complexity of fault current calculations comes in determining the equivalent system impedance Zs between the source and the point of the fault. 

When short circuit analysis is the core of an electrical safety program, it means that building distribution models must contain significant information about the size and path of the building distribution system.  Determining cable lengths and information about over-current devices means that all the cables have to be measured in length and known in their size.  This field work involves a great deal of time and can put electricians at risk while they investigate live circuits. 

An experienced electrical power engineer can make approximations of fault currents and make judgments about protective measures that should be implemented.  Figure 4 shows, diagrammatically, a typical building power distribution system.

Figure 4: Typical Building Power Distribution



Electrical Wiring

A conductor is a material that easily conducts electricity, requiring little electrical stimulation to induce an electric current. Copper and aluminum are common conductors. The current-carrying capacity of a conductor is determined by its physical size and the ambient temperatures. Therefore, in the design of a circuit, it is important to ensure that the rate at which heat is dissipated in the conductor is equal to or greater than the rate at which heat is generated in the conductor.

All conductors have resistance, although this resistance is small. When current passes through a conductor, heat is generated. The amount of heat generated varies directly with the square of the current, and the amount of heat dissipated is a function of the circumference of the wire and the ambient conditions. To ensure a proper heat dissipation rate, wire size must be increased at a higher rate than wire capacity.

For example, to maintain a higher heat dissipation rate than heat generation rate, doubling the conductor size results in a current-carrying capacity that is less than doubled. Doubling the wire cross-section area increases the circumference by 41.4 percent. However, doubling the current increases the amount of heat generated by four times. Moreover, lower heat dissipation at higher temperatures requires de-rating the current capacity of the wire.

Wire Sizes

Wire size is measured in circular mils (CM). A mil is one thousandth of an inch, and a circular mil is the area of a l-mil-diameter circle. The American Wire Gauge (AWG) is the most commonly used measure in the United States. However, AWG should not be confused with the gauge used to measure steel wire for non-electrical applications. The AWG scale consists of wire numbers, which are even, starting with the number 40, representing a wire diameter of 0.003 in. The smaller the wire number, the larger the cross-sectional area.

Building wiring applications commonly use wire sizes of 14, 12, 10, 8, 6, 4, and 2. Wires larger than 2 are called 1/0 (one-naught), 2/0, 3/0, and 4/0. A wire larger than 4/0 is not designated by a numerical size, but rather by its cross-sectional area in mil circular rail (MCM). One MCM is equal to 1,000 CM; sample designations are 250 MCM, 500 MCM, or 750 MCM. These size designations are used for building wiring. Odd-number size designations are commonly used for magnetic wires, such as those used for motors and transformers. Number 8 or smaller wire can be solid or stranded, but number 6 and larger wire must be stranded to achieve the desired flexibility.  Figure 5 show the different wire sizes for different temperatures, ampacities and wire types.

Figure 5: Wire Sizes and Types

NEC Table 310-16.jpg

Wire Insulation

Indoor building wire is suitable for voltages up to 600 V. The most common wire insulation, thermoplastic, comes in many different types. Moisture resistant thermoplastic (type TW) and moisture and heat-resistant thermoplastic (type THW) are the most common and can be used in wet or dry applications. When type TW is used, wire temperatures should not exceed 140.5°F; when type THW is used, temperatures should not exceed 167.50oF Heat-resistant rubber (RHH) and heat-resistant thermoplastic (THHN) types are used only in dry locations, and the wire temperature should not exceed 194.5°F. THHN and heat and moisture-resistant thermoplastic (THWH) types have an oil-resistant final insulation layer that adds strength and greater insulating capacity. Moisture and heat-resistant cross-lined synthetic polymer (XHHW) type has a cross-linked synthetic polymer, which also adds strength and even higher quality insulation.

Wire Size Selection

In addition to ampacity, an important factor in wire size selection is voltage drop. The voltage drop on branch circuits should be kept below 2 percent, and that on the feeder and branch circuit, below 3 percent. It is important to keep the voltage drop to a minimum, because in addition to the energy loss, voltage drop has a negative effect on electrical appliances. For instance, a 5 percent voltage drop on an electric motor means a 10 percent drop in output power. Similarly, for incandescent lamps, a 5 percent voltage drop means a 16 percent drop in light output; for fluorescent lamps, a 10 percent voltage drop means a 3 percent drop in light output.


Wires in Parallel

In some instances, two or more wires can be used in parallel instead of using single wire.  “Paralleling” means that the conductors of each phase are electrically joined at both ends to effectively form a single conductor.   Starting with conductors 1/0 and above, paralleling is permitted in Section 310.4 of the NEC.  It is important to note, the NEC does not permit paralleling conductors smaller than 1/0; neither is it permitted to tap into only one of the parallel conductors.  Tapping into only one of the parallel conductors will result in unbalanced distribution of tap load current between parallel conductors.

The ampacity of conductors is not a linear function of their size: that is, the doubling of the cross-sectional area of the conductor does not result in the doubling of its ampacity rating.  Another reason to avoid using large conductors is the difficulty of pulling them  into a raceway.   A large-sized conduit is required , which is cumbersome to handle during installation.  

Wire Splices and Terminations

The weak links in building wiring systems are usually wire splices and terminations. It is important to ensure that splices and terminations are electrically and mechanically correct. Screw-type terminals should have more than two-thirds wrap in the clockwise direction, with no overlap. Copper wires or copper-clad aluminum wires commonly require solderless connections, since aluminum rapidly forms aluminum oxide, a poor conductor when exposed to air. The aluminum connectors should be able to penetrate the oxide layer. Note that soldering connections is seldom done for building wiring.


Conduit Systems

A conduit wiring system provides a high level of mechanical protection for electrical circuits. This system reduces the probability of fire from overloaded or short-circuited conductors. With a conduit system, the circuit wires are easily replaced and easily removed, and new circuits are easily pulled if there is space. Conduits may be buried in walls or surface mounted. The ambient conditions determine the type of conduit, the type of coating, and the type of fitting. Dust-tight, vapor-tight, or water-tight conduits are available in 10 ft. lengths. The size is determined by the internal diameter in inches. The standard sizes are 1/2 in., 3/4 in., 1 in., 1 1/2 in., 2 in., 3 1/2 in., 4 in., 4 1/2 in., 5 in., and 6 in. The most common types of conduits are rigid galvanized conduit, intermediate metal conduit, electric conduit, rigid polyvinyl chloride (PVC) conduit, and flexible conduit.

Rigid Galvanized Conduit

Rigid galvanized conduit (RGC) provides the highest level of mechanical protection. RGC is made of heavy-wall steel that is either hot-dipped galvanized or electro-galvanized to reduce the damaging effects of corrosive chemicals found in insulation. RGC differs from wafer-type conduits in that the interior surfaces are prepared so that wires can be easily pulled. The wall is approximately 0.109 in. thick. The disadvantages of RGC are its high cost, its heavy weight, and its difficulty of installation (i.e., cutting and bending).

Electric Metal Tubing

Electric metal tubing (EMT) has a wall thickness that is about 40 percent less than that of RGC, making it lighter and less expensive. EMT is used mostly for branch circuits above suspended ceilings. Unlike RGC, EMT is not threaded into a fitting or box but mostly uses compression or set-screw fitting joints. EMT can be jacketed with PVC to make it resistant to corrosive chemicals. Proper care must be taken to prevent damage to the PVC jacket during cutting or bending.

Rigid PVC Conduit

PVC conduit is lightweight and works well even in highly corrosive areas or places where moisture and condensation are a problem. Two advantages of PVC conduit are that it has no voltage limitation, and it resists aging from ozone and sunlight exposure. Because PVC is not conductive, a grounding conductor may also be required.  

Flexible Conduit

Flexible conduit should be used when a connection is needed with vibrating or moving parts, such as motors, or when rigid conduit cannot be formed to a required contour. Flexible conduit normally is used for short distances of no more than 60 ft.. PVC-jacketed, liquid-tight, flexible conduit is used for damp locations.

Design engineers need to be mindful that the inner diameter of each of these raceway types differs from its nominal value and that additional care is required specifying the same number of wires in EMT or rigid steel. 


Electric Power

Solved Problem: Building Wiring Design

Some of the broad quantitative principles are shown here.  The NEC is a fire safety code and nothing prohibits a designer from exceeding code minimums other than, say, project economics.

Situation.  Each floor of a four floor 12,000 dormitory has 2000 sqft of that is corridor, closet and stairway areas.  480V feeders from the substation run up to each floor and transform power to 208/120V power distribution panels.

There are 120 general use receptacles on each floor. 

Requirements.  A. Calculate the minimum ampacity rating for the feeder conductors to the panel.  Assume that the entire lighting load is continuous and the actual lighting load is less than the computed lighting load. 

Solution.  We obtain the minimum lighting VA allowable for this type of facility from Figure 26.XX NEC Table 220-3(b)


For the feeder taps to each panel:


(2000 sqft @ 0.5  VA/ft2)                  = 1,000 VA

(12,000 2000) ft2 x 2.0 VA/ft2       = 20,000 VA

Receptacle load: 120 x 180 VA          = 21,600 VA

Total computed load                           = 42,600 VA  

Net computed load using demand factors from NEC XX

42,600 VA at 40% (50,000 VA or less) = 17,040 VA


Minimum ampacity =   = 47A


For the main vertical riser:

Total computed load for 3 panels = 3 x 42,600 = 127,800 VA

Using the demand factors from NEC:

First 50,000 VA @ 40%                                 = 20,000 VA

Remainder (127,800 - 50,000) @ 20%           = 15,560 VA

Net computed load                                         = 35,560 VA


Minimum ampacity =   = 98.8 ~ 99A


Remarks.  We have selected the NEC requirements for VA lighting levels because of local code fire safety requirements.   The NEC VA requirements in Article 210 through 230 are coming under scrutiny for energy conservation.  See ASHRAE/IESNA for power density limits.


Electric Power Quality

Currently the electrical distribution system is cluttered with a wide variety of nonlinear devices that generate power disturbances and interference. These disturbances, commonly referred to as electrical noise, are the result of transients and harmonics. Both can have a damaging effect to sensitive electronic equipment. Power quality problems cost the industry about $2 million a day. 

There are two types of noise: normal-mode and common-mode. Normal-mode noise exists when a noise voltage appears equally in each line to-line and line-to-neutral connection. Common-mode noise occurs between line-to-ground or neutral-to-ground connections. Both can negatively impact sensitive electronic devices, resulting in system failure, component damage, database corruption, disk head crashes, and logic errors, among others.


A transient is a high-amplitude, short-duration electric pulse that is superimposed on the normal sinusoidal AC voltage. The duration of a transient can range from 0.5 to 200 ms, and the voltage can rise thousands of volts per microsecond. In other words, a transient is a high-amplitude noise that can result in severe hardware fatigue and failure, in addition to other problems mentioned earlier. There are two principal sources of transients: lightning and switching surges. The impact of transients can be minimized with surge suppressors and active power line conditioners.


Ideally, AC electricity is a pure sinusoidal wave of one single frequency and can be referred to as clean power. More specifically, in the United States this frequency is 60 cycles per second, or 60 Hertz (Hz). In a power distribution system, as long as the circuit devices consist of linear elements, (resistors, unsaturated inductor, and capacitors), the AC power wave shape will remain the same. However, as soon as nonlinear equipment (switching devices, asymmetrical devices, saturated inductor) are introduced into the distribution system, the wave shape of the AC power will be distorted.

The distortion wave can be mathematically analyzed as the summation of integer multiples of the fundamental AC power, known as harmonics. In other words, for a 60-Hz system, the frequencies for the second, third, and fourth harmonics are 120 Hz, 180 Hz, and 240 Hz, respectively.

One of the common measures of calculating the influence of harmonics is total harmonic distortion (THD), which is the summation of the root mean square of all harmonics as a relative percentage of the fundamental frequency; there are many devices available to analyze power harmonics. There are three major classes of nonlinear devices that are sources of harmonics in power distribution systems: switching devices (e.g., variable frequency drives, synchronous generators, converters), ferromagnetic devices (e.g., transformers, motors, reactors), and arcing devices (e.g., fluorescent, mercury vapor, and sodium vapor lighting and arc furnaces). Approaches that can reduce the impact of harmonics are discussed in the sections that follow.

De-rating Distribution Transformers

In the absence of any harmonics, transformers can be fully loaded to their rated value, under normal ambient conditions, without any problems. If there are harmonics in the system, then the transformer must be de-rated accordingly. For instance, if the third and fifth harmonics are 20 percent each, then the transformer should be de-rated by 8 percent. If the third and fifth harmonics are about 70 percent each, then the transformer will be overloaded even when connected to a small load.

Over-sizing the Neutral Conductor

In a three-phase wye system, the neutral conductor carries only the unbalanced current. Therefore, if the system is balanced, the current flowing in the neutral wire is zero. However, this is only true in the case of the fundamental frequency current. If there is any third harmonic (or any other triple) current in the three-phase system, the neutral current, instead of canceling out for the fundamental frequency, will add up. The neutral current can potentially be as high as 1.73 times the phase current. Because the neutral circuit is not normally protected by an over-load device, such an overload will likely result in burned wires and electrical fires, especially at the connectors and splices. Therefore, facilities personnel should avoid the use of shared neutral wire when supplying a single-phase nonlinear load, or double the size of neutral wire when shared neutral must be used.

Loading Circuit Breakers

Harmonics can result in nuisance trips of circuit breakers operating near their design trip point because of the peak current heating of the contacts and the vibrations induced by the higher harmonic currents. It is recommended that when serving nonlinear loads, including computers, the panel circuit breakers should not be loaded above 80 percent of their continuous loading capacity.

Power Factor Considerations

The presence of harmonics reduces the system power factor to a lower value. Adding more capacitors could potentially cause resonant conditions, attracting high-frequency currents, and result in overheating or failure of the capacitors. To avoid such problems, a harmonic trap, which is a series LC circuit tuned to the lowest harmonic, can be installed.

Finally, to the degree possible, use low-impedance distribution transformers connected in a delta-wye configuration. This way the third harmonic will be trapped in the delta winding.

Power Quality Considerations

As facilities personnel experience further proliferation of sophisticated electronic equipment, the problem of power quality will become more prevalent. In most situations, determining the source of the problem is the challenging part. The first question is determining whether the problem is on the line side or the load side. A simultaneous increase in current and a decrease in voltage indicate that the problem is downstream. However, if there is no change in the current, the problem is upstream. If the voltage THD is more than 6 percent for branch circuits or above 4 percent at service entrance, then all THD is from one harmonic; if THD is concentrated at higher harmonics, there should be a cause for concern.

Lack of attention to power quality problems can result in equipment downtime, premature equipment failure, and high service cost. There is no standard solution that can work in every situation. This is yet another challenge that today's facilities manager must overcome.


Electric Motors

Electric motors are rotating machines that convert electrical energy into mechanical energy. The two main elements in motors are the stationary elements (the starter, brushes, yoke, armature winding, and motor housing) and the rotating elements (the field winding rotor and the slip rings).

Most motors can be classified as either synchronous-type or induction-type motors. Synchronous motors, rarely found in modern building systems, have been in use since 1890 and are almost identical to synchronous generators.  Induction motors, on the other hand, are almost universal in building power systems.  Because of the magnetic effects of the deliberate  short-circuiting of their windings they are often regarded as "transformers on a stick." 

The canonical equation for motor power analysis is:

1 horsepower = 746 watts = 0.746 kW                       Eq 26.6

Electric Motors and Energy Management

Electric motors consume about 2/3rds of the electrical energy annually and about 60 percent of the electric motor energy consumption occurs in commercial and industrial applications. Thus, for example, a 30-horsepower motor with an efficiency improvement of 3 percent that works 10 hours a day for 250 days a year means a savings of $167 per year, or a total of $3,340 over the average 20-year life of the motor (at $0.06/kWh).


Efficiency is a measure of a motor's effectiveness in converting electrical energy to mechanical energy. A motor's efficiency is expressed as the ratio of the motor's output to its input. Efficiency is also expressed as the ratio of a motor's output power to its output power plus losses. The lower the losses, the higher the motor's efficiency. Also, the higher the output load, the higher the motor's efficiency, which means the highest motor efficiency is achieved at rated load, and the efficiency is lower for partial loads. Generally speaking, large motors have higher efficiency than small motors. Also, high-speed motors have greater efficiency than low-speed motors. Therefore, a higher speed motor with a reducing gear system might be more energy efficient than a lower speed motor, despite the mechanical energy losses created by the reducing gear.

In induction motors, efficiency is also a function of the motor slip. The higher the slip, the higher the motor energy losses, which causes heating of the windings and reduces the useful life of the motors. Multiple-speed motors have a lower efficiency than single-speed motors. Single-winding multiple speed motors are more efficient than two-winding multiple speed motors.

A small percentage increase in motor efficiency greatly reduces motor losses. For instance, an increase in motor efficiency from 85 percent to 88 percent translates into a 20 percent reduction in losses and a longer operating life for the motor.

Proper care increases the useful life of motors and ensures maintenance of good motor efficiency. A basic maintenance program should include periodic inspections and correction of unsatisfactory conditions. Inspections should incorporate checking lubrication, alignment of motor and load, belts, sheaves, couplings, tightness of the belts, ventilation, presence of dirt, input voltage, percentage of unbalance, and any changes in load conditions. Dust buildup on fans, misalignment of gears and belts, and insufficient lubrication increase motor friction, thus reducing the efficiency and life of the motor.

Example Problem: Motor Efficiency

Situation: A 20 hp motor is to operate 4000 hours per year at 100% rated load.  Any energy-efficient motor has an efficiency of 91.5% as compared to 89.1% for the standard motor.  Assume that the energy-efficient motor costs $1000, and the standard motor costs $750, and energy costs $0.10 per kWhr. 

Requirements.  Calculate the payback period for the energy-efficient motor. 


20 hp x 0.746 kW/hp = 14.92 kW
Standard motor input power =  = 16.75 kW
Energy efficient motor input power =  = 16.31 kW

The cost of energy consumed is:

Standard motor = 16.75 x 4000 x $0.10 = $6700

Energy efficient motor = 16.31 x 4000 x $0.10 = $6524

Savings for the energy efficient motor = $176

Difference in the cost of the motors = $250
Simple payback period =   = 1.42 years

Variable-Frequency Drives

Most industrial processes require variable-speed motors. The speed of an AC motor is determined by the line frequency of 60 Hz. To effectively vary motor speed, a variable-frequency drive (VFD) must be used. A VFD consists of a DC rectifier, a filter circuit, and an invertor (Figure 26-5). Sixty Hertz (60 Hz) line voltage is converted to DC voltage and then to AC voltage. The control circuit determines the output frequency and, thus, the speed of the motor.

Figure 6 :  Variable Frequency Drive


The invertor circuit changes DC voltage into three-phase variable-voltage/variable-frequency AC output. The invertor consists of at least six thyristors, each conducting 180 degrees per cycle. The switching sequence produces a three-phase output voltage.  The output frequency is determined by control circuitry.

The three common types of variable-frequency drives are adjustable voltage input (AVI), current source invertors (CSIs), and pulse width modulation (PWM). The voltage in AVI is controlled by the DC input to the invertor and requires the use of either a silicone-controlled rectifier (SCR) or diode rectifiers in conjunction with a chopper to carry input voltage. AVI is the simplest of the three invertor types in terms of control circuitry and also has regeneration capability.

CSIs receive DC input voltage from an SCR bridge in series with a large inductor, creating the current source. In a PWM invertor, the output voltage wave form has a constant amplitude with periodically reversing polarity that provides the output frequency. The output voltage also is varied by changing the pulse width. PWM has the most complicated control logic and the lowest efficiency, but it also has fewer harmonic problems compared with the other two types. PWM is becoming popular for many applications.




Lighting deals with providing visibility so that humans can visually experience their surroundings. Natural lighting is provided by the sun, and artificial lighting is predominantly provided by converting electrical energy into lighting energy in the frequency range visible for humans. There are many reasons why lighting is needed for an area; these reasons can be divided into three general categories: functional lighting, safety and security lighting, and ornamental or architectural lighting.

Functional lighting is needed to complete a particular human task, such as working in an office, walking along a sidewalk, or driving a car in a parking lot. Lighting level and the quality of light necessary for people to perform the required task have been the subjects of many management studies since World War II.  The reason for the high interest in this topic is that labor costs are a large percentage of most industrial production costs, so even a small percentage gain in productivity can have a significant impact on the net revenue of an enterprise. Because lighting interacts with the work environment, lighting designers must be sensitive to the needs of the people who are affected.

Safety and security is the next major category of lighting needs. Many individuals associate lighting level with safety. This perception has been the source of long debates. Both the quantity of light and lighting quality are important. In outdoor applications, especially in parking garages, improper lighting can result in a variety of problems such as shadow zones, reduced visibility, a loss of direction, a feeling of claustrophobia, and a sense of insecurity. Both horizontal and vertical illumination levels must be considered for these applications. Another important factor for safety and security lighting is the location of lights in relationship to landscaping.  Sometimes large quantities of light may be blocked by tree leaves, large bushes, or other vegetation, and therefore lighting locations should be properly coordinated with the site landscaping.

The third general category of lighting deals with architectural lighting. The main purpose of this category of lighting is aesthetic lighting, which primarily deals with a particular mood or visual impact the designer is trying to create. Because of the ornamental nature of this type of lighting, functionality is not a major concern.

Elements of Lighting

Electric lighting is most commonly generated when electrical current flows through either a resistance filament or a gas. In the first case, the filament is heated to such high temperatures that it starts glowing; in the second case, the gas molecules' excited atoms emit radiant energy. There are basically three types of lighting: incandescent, fluorescent, and high-intensity discharge (HID) (e.g., mercury, metal halide, high-pressure sodium, low-pressure sodium and inductance). All gaseous lamps, such as fluorescent and HID lamps, need a ballast to start the lamp and regulate the current. The entire lighting assembly, lamp, ballast, lens, and associate housing, is referred to as a luminaire.

The quantity of light in an area is related to luminous flux, the amount of light emitted by a lamp, and illuminance, the amount of light reaching the surface area. The initial lumen output of a new lamp will be high, but after the lamp operates for about 100 hours, the light output stabilizes for the useful life of the lamp. When the lamp gets old, the light output drops rapidly over time. Other factors that affect the light output level of a lamp are the amount of dirt and dust accumulated on the fixture, the reflectance of the room surfaces, room proportions, and surface materials.

In addition to the quantity of light, the color of light is important. Color rendering refers to how accurately colors and color shades are rendered and is measured using a color rendering index (CRI) ranging from 0 to 100, with the incandescent lamp defined as a CRI = 100. Typical fluorescent lamps have a CRI rating of about 65.

Lighting efficiency is measured in lumens per watt. The most common types of lamps, incandescent, fluorescent, metal halide, high-pressure sodium, and low-pressure sodium, are discussed in the sections that follow.

Incandescent Lamps  Incandescent lamps are the oldest and simplest type of lamps that consist of a tungsten alloy filament. When the filament is raised to a high temperature, it glows, and light is emitted. Because of their ease and simplicity of construction, incandescent lamps are the simplest and least expensive lighting system. Their high CRI rating makes them desirable for many indoor functions where color rendition is important, such as art galleries and painting studios. Incandescent lamps have an efficiency of 10 to 15 lumens per watt, which is the lowest of all commercially available lighting systems. Moreover, the average useful life of a lamp is about 1,000 hours. For these reasons the use of incandescent lamps has continually decreased for commercial and industrial applications. Many retrofit kits are available to convert existing incandescent lighting systems to more efficient systems with many lamps now available that fit directly into incandescent lighting fixtures.

Fluorescent Lamps  The fluorescent lamp has been the most widely used type of lamp for commercial applications. The main reasons are its relatively high efficiency, its moderate cost, and the maturity of the technology, which has been around for many decades. Moreover, unlike other gas-filled lamps (e.g., HID), fluorescent lamps can be turned on and off frequently and will reach their full brightness rapidly. The typical efficiency of a fluorescent lamp ranges from 60 to 75 lumens per watt, and the average lamp life is between 8,000 and 10,000 hours. However, color rendition is not as good as that of incandescent lamps. The typical CRI of a fluorescent lamp is 65, which is acceptable for most applications, except where color rendition is particularly important. Fluorescent lighting needs a ballast and starter for the fluorescent lamp to operate.

Metal Halide Lamps  The metal halide lamp is an HID lamp that is becoming popular for indoor applications. Such lamps are commonly sized from 32 to 1,000 W and produce reasonable results for color rendering applications with a CRI of 65. Their high efficiency and long life make them a good choice for both indoor and outdoor applications. The efficiency of these lamps can range from 78 to 110 lumens per watt and the lamp life ranges from 10,000 to 20,000 hours. Generally, the larger wattage lamps have a higher efficiency.

There is a safety precaution that must be remembered with metal halide lamps to avoid possible shattering and early failure of the lamp. These lamps are constructed of an outer bulb with an internal arc tube made of quartz. The arc tube operates under high pressure and at approximately 1,100°C. The arc and outer bulb may unexpectedly rupture as a result of system failure or misapplication. To safeguard against any hazards, the following three precautions must be taken:

  1. Lamps must be operated only in fixtures with lenses or diffusers that can contain fragments of hot quartz or glass up to the above mentioned temperature.
  2. If a lamp is supposed to operate continuously, it should be turned off at least once a week for at least 15 minutes. Otherwise, the risk of rupture will increase greatly.
  3. Certain types of metal halide lamps will automatically extinguish if the outer skin is punctured. This is a desirable feature, because if a lamp with a broken outer bulb is allowed to operate, the ultraviolet light emitted from the lamp may cause serious skin burns and eye inflammation.

High-Pressure Sodium Lamps  The high pressure sodium (HPS) lamp is a high-intensity discharge lamp that can be used both for indoor and outdoor applications. Nominal lamp sizes range from 35 to 1,000 W. HPS has a high efficiency (typically 64 to 140 lumens per watt) and relatively long life (10,000 to 24,000 hours). The main objection to HPS lighting is its low CRI rating, which is about 22. With color correction this factor can be increased to 65, but as would be expected with better color rendition, there is an efficiency drop and reduced life as well as a higher cost. Despite their poor color rendition, these lamps are desirable because of relatively higher efficiency and long life. HPS lamps can be used for parking garages, gymnasiums, swimming pools, transportation centers, malls, walkways, parks, and general security lighting.

Low-Pressure Sodium Lamps  The low-pressure sodium (LPS) lamp has the highest efficiency compared with other types of lamps. The typical lumens per watt ranges from 100 for 18-watt lamps to 163 for 135-watt lamps, and the average lamp life is 16,000 hours. The major shortcoming of LPS lighting is its poor color rendition (the CRI rating for LPS lighting is 0), and for this reason these lights are not desirable in areas used by people; such light can be used only in areas where color rendition is not an issue. LPS lighting is commonly used for extremely large areas such as warehouses and general security lighting.  Because of the color rendition issue the industry has pretty much moved away from using LPS.

Inductance Lamps   The inductance lamp is another form of HID.  The typical lamp size is from 40 to 200 watts.  It is an efficient lamp with approximately 85 Lumens per Watt and a long life of approximately 100,000 hours.  It's coloring is good with a CRI of 80.  The main advantage of this lamp is its long life, which is nearly 5 times longer then other HID lamps.  However, this longer life is offset by the cost, which is approximately three times as great for other HID lamps.  A life cycle cost analysis based on initial costs and operating costs will help with the decision on whether to go with the inductance lamp.

Light Emitting Diode (LED)  LED technology is advancing quickly as it moves from small applications in exit signs and decorative lighting to exterior lighting fixtures for pedestrian, street and parking lot use.  Color rendering can be controlled to a wide range of colors and life of LEDs is almost infinite.  The limiting factor of  LED life, on large systems, is the life of the supporting components.  However, the costs are higher but as more manufactures develop LED technology the costs will come down.


Lighting - Part 2

Lighting Application

Placement of lighting deserves careful planning and only the broad concepts will be discussed here.

The lighting levels provided should be uniform, with the maximum ratio between the highest and lowest level being less than 2:1 to prevent glare and undesirable shadows.

Indoor lighting can be direct or indirect. Indirect lighting is softer to the human eye and more desirable for many indoor commercial and residential applications. However, it requires surfaces with good reflectance, is more costly and less efficient than direct lighting. For these reasons, even in most commercial applications and all industrial applications, direct lighting is used.  However, educational institutions are using indirect lighting more and more, especially in classrooms and teaching labs, as it creates higher quality learning environment.

For outdoor lighting, an important phenomenon that must be considered is lighting spill to surrounding areas, which could become a source of nuisance for neighboring customers.  Careful design, proper aiming diagrams and the installation of fixture baffles will help to reduce light spill.

To maintain an acceptable lighting level, lamps should be replaced when they reach their average useful life, even if they are still functioning. Group re-lamping should be considered when access to fixtures is difficult such as in high ceiling atriums, gymnasiums, etc. When lamps can be easily changed out, such as in an office, then single lamp changes may be more cost effective. Another important practice is to clean the lighting fixtures periodically. As dirt and dust accumulate in a fixture, the quantity of light delivered depreciates significantly. This is a function of the ambient conditions. In a clean environment the light loss is about 10 percent, whereas in a dirty environment it can be as high as 50 percent. Cleaning light fixtures on a regular schedule is recommended.  

In the past, lighting designers added large safety margins to compensate for lamp dirt depreciation and lamp aging factors. Currently most lighting systems are designed with relatively smaller safety margins. Therefore, if a few fixtures are burned out or the fixture becomes dirty, the drop in lighting level will be significant. Therefore, it is important for facilities managers to be sensitive to this fact and, where possible, to institute a light fixture cleaning program.  Studies have shown that re-lamping and cleaning fixtures in an area where that maintenance has not been done will increase the lighting levels by 100% or more.

Lighting Design Calculations

Two definitions are helpful in understanding lighting design calculations:

  • Lumen: The quantity of light striking an area of 1 sq. ft., all points of which are 1 ft. from a 1-candlepower lighting source.
  • Foot-candle: A unit of measurement representing the intensity of illumination on a surface that is 1 ft. from a l-candlepower light source and at right angles to the light rays from that light source. One footcandle (FC) is equal to 1 lumen per square foot.

Some sample lighting levels recommended by the Illuminating Engineering Society of North American (IESNA) for typical campus spaces:

Figure 7: Example Recommended Lighting Levels

Space Type Footcandles
Office Space 20 - 50
Classrooms 50 - 100
Conference Rooms 20 - 50
Laboratories 50 - 100
Libraries 20 - 50
Lobbies 10 - 20
Dining Rooms 5 - 10
Outdoors 1 - 3


There is a range of footcandles for the various spaces because of such things as the quality of the material being viewed, the accuracy of the task and the age of the person.

For comparison, the illumination levels of some natural light sources are as follows:

  •          Starlight: 0.0002 FC
  •          Moonlight: 0.02 FC
  •          Daylight: 100-1,000 FC
  •          Direct sunlight: 500-1,000 FC

Several methods exist for calculating the number of fixtures required for a particular area. One of these methods, the lumen method, is summarized as follows:

1.       Determine the required lighting level (IESNA standards)

2.       Calculate the room cavity ratio (CR):

CR = 5 x Room Height x (Room Length + Room Width)/(Room Length x Room Width)    Eq. 26.7

Using the lighting tables, the cavity ratio, combined with wall reflectance and ceiling cavity reflectance, gives the coefficient of utilization (CU). The CU is a function of the ceiling and wall reflectance, as well as of room geometry; it is the measure of the ratio of the lumens reaching the working surface to the total lumens generated by the lamp. The higher and narrower a room, the larger the percentage of light absorbed by the walls, and the lower the CU value.

3.       Determine the illumination loss factors. From the time a luminaire is installed and energized, a number of factors contribute to loss of illumination. The major factors contributing to loss are age, dirt depreciation, and voltage-to-luminaire variation. Variation in voltage to luminaire is a bigger problem for incandescent lamps than for fluorescent lamps. For an incandescent lamp, a 1 percent change in line voltage causes a 3 percent change in lumen output. For a fluorescent lamp, a 2.5 percent change in line voltage causes a 1 percent change in lumen output.

4.       Calculate the number of fixtures required:

Number of fixtures =

5.       Determine the location of lights based on the general architecture, task, and furniture arrangement.

The quality of light is just as important as the quantity of light. Quality of light involves proper light color, proper light distribution, and lack of glare. A relatively bright area within a relatively dark or poorly lighted area causes glare.

Lighting geometry also plays an important role in the quality of the lighting.  Veiling reflections (ex. off of computer screens or glossy magazines) can make it difficult to see material.

Lighting design software from fixture suppliers is available.

In practice, lighting should be one voltage throughout a building.  Larger, newer facilities built since 1960 will usually have 277V lighting systems.  Smaller, older facilities built before 1960 will have lighting circuits run at 120V.   In residences, lighting circuits are restricted to 120V. 

Lighting Control

A pole switch (on-off switch) is the simplest form of light control. A dimmer switch supplies illumination control and can be surface mounted, flush mounted, or controlled by a pull chain. For safety, the control switch should be mounted on the hot wire rather than the neutral wire. If light control is needed from two locations, then three-way switches should be used. If additional controlling points are required, then two three-way and four-way switches should be used. The "on/off" designation is not found on three-way and four-way switches, because the lights can be on or off depending on the position of all switches.

A lighting contactor should be used for controlling large banks of light. The circuitry is similar to that used for a direct-start motor control center. Here the control voltage can be different from the lighting voltage, and the lights can be controlled remotely.

Lighting design is a subtle art, and a short discussion cannot cover the topic. However, it should be mentioned that lighting consumes nearly 35 percent of the nation's electricity, which is why improving lighting efficiency was greatly emphasized in the 1992 Energy Policy Act. The references at the end of the chapter provide a more comprehensive treatment of this topic.



Grounding eliminates the possibility of shocks and minimizes lightning damage; it is a required safety item for secondary distribution. There are two distinct grounding categories.

First is system grounding, which means grounding one of the current-carrying wires to avoid a floating system.

Second is equipment grounding, which consists of placing a non-current-carrying wire between the metallic frame of electrical equipment and the conduits, or armor of a ground rod.

Therefore, a reference to "grounded wire" usually refers to a system ground, in which case, if the circuit is on, the wire is hot. However, "grounding wire" refers to the equipment wire that has no current under normal conditions.

Whenever the topic of electric shock is discussed, the question of danger normally arises. Shock danger is not related to the high voltage of the equipment, but to the current flowing through the human body. A current of one milliampere (mA) is not perceptible. A current of 1-8 mA causes mild to strong sensations. A current of 8-15 mA is unpleasant, but the shock victim can release the object producing the electrical current. A current above 15 mA causes a muscular freeze, preventing the victim from releasing the object producing the current, and a current of more than 75 mA is fatal.

Ground Fault Protection

The National Electric Code (NEC) requires ground fault equipment protection. Ground fault interrupters (GFIs) protect equipment from low-grade faults that are not large enough to be interrupted by conventional over-current protection devices, such as fuses or circuit breakers. The function of GFIs is often misunderstood. It is easier to state what a GFI is not. GFIs do not protect people or prevent shocks. They neither prevent ground faults nor protect equipment from a high-grade fault. GFIs only protect equipment from low-grade faults. The principle behind GFIs is simple: With the use of a current transformer surrounding the live wires, the current remains within the wires, and the net current is zero. However, if leakage occurs, the net current is not zero, and the GFI will trip the circuit breaker.

A GFI is required if the line-to-ground voltage in a grounded wye service is between 150 V and 600 V and service disconnect is rated at 1,000 A or more. For health care facilities, at least one additional layer of ground fault protection should be provided downstream toward the load. Like conventional current protection systems, GFIs should be coordinated to ensure selectivity. As mentioned earlier, a GFI does not prevent ground fault. To eliminate the possible causes of ground fault, required maintenance is recommended, which should include cleaning insulators, replacing cracked insulators, and tightening loose connections.

Solved Problem: Building Power System Design

In this section we illustrate some of discussion with a numerical example that is representative of the engineering effort by designers.  It illustrates some of the first cost issues as well as the operational cost of ensuring electrician safety.

The situation is a university hospital facility.   Four floors with one power distribution panel on each floor for a sum total of 1000 kVA demand.  The mechanical penthouse has chilling tower with the largest motor 150 hp for the entire building.   The MCC is likely to remain energized while in service.

Largest motor is an 150 hp, three-phase 480V squirrel cage induction motor which is the largest motor fed from the MCC. 

Motor Control Center 1000A bus with the sum of all the motors at 425 amperes

With all the motors running, the power factor is 85 percent.

Utility has 500,000 kVA fault current available at the 13.8 kV supply point.

Requirements.   Specify wire sizes, over-current devices and incident energy at the MCC.  Compute flash hazard at MCC

Solution.  Set up Per-Unit Notation:

Let Base kVA = 1500 kVA (the substation’s kVA rating)

Secondary Voltage = 480/277V


Full load amperes =  = 1804 A


ZU =  = 0.003 p.u.


IT =  = 0.0605


 = 16.53 p.u.


IF = Total Fault Current = 1804 x 16.53 = 29,800 amperes


Select the motor control center feeder to meet all of the ampacity, short-circuit and limits on voltage drop.  NEC Article 220 covers branch circuit and feeder calculations.  Specifically, Section 220-10(b) deals with Continuous and Non-Continuous loads. 

From 430-150, the FLA of the chiller motor is 180A

From Figure 26-XX, the minimum ampacity of the motor feeder shall be 125% of the largest motor + 100% of the remainder.

= (1.25 x 180) + 425 = 650A

From NEC Table 310-16 the minimum size is 1750 kcmil which, because of size, will require phase conductors to be wired in parallel; common in circuits that require more than 400 amperes at any voltage.   We want to drive down to less than 400 amperes, the current running through one conductor by splitting the 650A into two conducting paths.

Minimum ampacity per conductor = 650/2 A

Minimum size of each conductor (per NEC 310.16) is 400 kcmil, THW which permits up to 335A per conductor.

Minimum size required for short circuit current.  Since Isc = 37,000 amperes symmetrical,

From the cable manufacturer’s table K0 is 1.3 and the clearing timem is 2 cycles.

IASYM = 1.3 x 37,000 = 48,100 amperes

Minimum size required for voltage drop

Current per conductor = 325

Ampere-feet = 325 amperes x 200 feet = 65,000 = 65 x 1000

Maximum allowable volts drop = 2% of 277 = 5.54 V

Maximum volts  drop/1000 ampere feet = 5.54/65 = 0.085

From the voltage drop Table ##  the minimum size of #4/0 for each conductor by interpolating between 80% and 90% power factor.   Since we have met the short circuit and voltage drop criterion the size of the MCC feeder is determined by ampacity. 

Calculate Available Fault Current at the MCC

Per unit resistance and reactance of the system up to the main secondary bus:

From Eq 26.XX

Isc = 37,000 A              with rated current = 1804 A


Isc =  = 20.5 p.u.


Zs =       = 0.049 = Rs + jXs


Where     = 4 and ? = 76 degrees


Rs = 0.049 cos 76 deg°

Xs = 0.049 sin 76°

Per unit resistance and reactance of the MCC feeder

Table ?? of the NEC for 400 kcmil copper in EMT

RAC = 0.035 ?/1000 ft             XL = 0.049 ?/1000 ft


For 200 ft:


RAC =  = 0.070 ?


XL =  = 0.0098 ?


Base impedance =  = 0.1535


RF MCC feeder =  = 0.046 p.u.           XF MCC feeder =  = 0.064 p.u.


The foregoing impedances are values for each conductor.  Since we have two conductors in parallel:


RF MCC feeder =  = 0.023 p.u.              XF MCC feeder =  = 0.032 p.u.


Total impedance up to MCC:


ZT =     (0.012 + 0.023) = j(0.048 + 0.032)

            = 0.035 + j(0.080) = 0.087 p.u.


ISC =      = 11.5 p.u. = 1804 x 11.5

            = 20,700 A symmetrical


Ratings of MCC main bus

This is a good point to reflect upon how much data needs to be accumulated to calculate short circuit.  The source and load but be identified, the length of the conductor, and the impedance calculated.  Getting this information correct is a dominant factor in performing short circuit studies that are the basis for flash hazard calculations.

Select Overload Protection

From NEC 430-150 Chiller Motor FLA = 180 (Nameplate the same as NEC requirement)  From the NEMA Starter Size table a Size 5 starter is required.   From a manufacturer’s Overload Heater Element Table heater type W39 listed for 180 amperes.  Note that the 180 amperes is the equivalent rating as the heater is connected through a current transformer.  Rated current of overload relay = 1.15 x 180 = 207A Select the motor branch breaker D

Maximum rating of trip unit = 250% of 180 = 450 A

From a manufacturer’s catalog (Frame Sizes and Typical Ratings for Industrial Type Molded-Case Circuit Breakers) the nearest standard trip rating is 450 amperes.  From above,  the available fault current is 20,700 amperes symmetrical.  From the Table, both 400 and 600 ampere standard frames have interrupting ratings of 30,000 amperes symmetrical at 480 volts.

Select the feeder breaker C.  From  Step above, the available fault current is 37,000 amperes symmetrical

Maximum setting of trip unit = (250% of 180) + 425 = 875 A

From the step above, the minimum ampacity is 650 A.  From Table XX, select 1600A frame with interrupting rating of 50,000 symmetrical at 480V (note that the 800 ampere frame does not have sufficient interrupting capacity).  Select a 1200 ampere sensor (next rating above 875 amperes)   See Step XX for coordinated trip unit settings.

Select The Main Secondary Breaker and Main Bus

Transformers are capable of sustaining from 15% to 25% overloads.  Section 450-3 of the NEC limits the maximum setting of the secondary overcurrent  device to 125% of rated current

Rated secondary current = 1804

Maximum setting of trip unit = 1.25 x 1804 = 2255 A

From Table XX, select a 3000 ampere frame with interrupting ratingof 65,000 amperes symmetrical at 480V.  Select a 2400A sensor.  See XX for coordination

Ratings of substation secondary bus are 42,000, Continuous current 2400 A

Select the primary fuse

Rated primary current =  = 63 A


Section 450-3 of the NEC limits primary fusing to a maximum of 300% of rated current of the transformer

Maximum rating of fuse unit = 3 x 63 = 189 A

Short circuit kVA capability (as given by manufacturer) = 500,000 kVA

From Standard Ratings of General Use Fusible Switches p223, select a 400 ampere fuse assembly with equivalent three-phase symmetrical interrupting capacity of 600 MVA.  Note that the 200 ampere fuse assembly has insufficient interrupting capacity even though it is large enough for the fuse unit.  See Step XXX for the final selection of the rating of the fusible element as required for coordination. 

Calculate Incident Energy at the MCC: 

[Use a generic and proprietary calculator to  show that PPE-2 is the most common level of protection required by electricians.  Alternatively, policies that permit electricians to de-energize equipment are the most effective but this policy may conflict with occupant power requirements.]



This is a short description of the broad contours of issues in building electrical system design.  The objective was to show the merging of objectives in the current trends in electrical system design and operation as discussed in the first part of this chapter.  Going forward we shall see more aggressive steps taken by the federal government to reduce energy consumption and it will be the task of safety and energy standards writers to come to a common understanding for the safe and efficient use of electricity in buildings.

Refer to updates on the APPA web site www.appa.org



References and Resources

Anthony, Michael A., Electrical Power System Protection and Coordination, McGraw-Hill Book Company, 1994

Anthony, Michael A,Municipal Power Security?, CHP and Sustainability (Lucus Hyman, Editor) McGraw Hill Book Company, November 2009

Clidero, Robert K. Applied Electrical Systems for Construction. New York: Van Nostrand Reinhold Co., 1982.

Earl, John T. Electrical Wiring Design and Application. Englewood Cliffs, New Jersey: Prentice Hall, 1986.

IEEE Standard 242, Industrial and Commercial Power System Design,

Lighting Handbook. Bloomfield, New Jersey: North American Philips Lighting Corporation, 1984.

Hughes, David, Building Electrical Systems, Prentice-Hall, 1984

Richter, Herbert P., and Creighton W. Schwan. Practical Electrical Wiring. New York: McGraw-Hill, 1987.

Williams, Dan R. Energy Policy of 1992. Englewood Cliffs, New Jersey: Prentice Hall, 1994.

William D. Stevenson, Jr. Elements of Power System Analysis Third Edition,McGraw-Hill, New York

Additional APPA Resources:

Facilities Performance Indicators Survey and Report     APPA's Bookstore     Facilities Manager Magazine     Credentialing      CFAR     Training     Glossary of Terms
Except as permitted under copyright law, no part of this chapter may be reproduced, stored in a retrieval system, distributed, or transmitted in any form or by any means - electronic, mechanical, photocopying, recording, or otherwise - without the prior written permission of APPA.
Please use the Print PDF button to print this Chapter.
You must be a logged into the APPA site and an APPA member to print or create a PDF of this BOK chapter.