CFE Schneider Electrical& Power

Electrical
& Power
Fall Edition
The role of NFPA 110 and its

interaction with other codes
When and how to design parallel

generators
Data center design

considerations
Contents
3  The role of NFPA 110 and its interaction with other codes
14  TLC Engineering & Schneider Electric – Partners at Nemours
Children’s Hospital
15  When and how to design parallel generators
25  Creating a life-saving environment through the IoT-enabled
EcoStruxure™ architecture and platform
34  Data center design considerations
Content archive
61  How to design a grounded power supply system
73  Critical power: hospitals and data centers
79  Power-generation systems in high-performance buildings
90  How to design electrical rooms
102  Understanding the changes to NFPA 70-2017
114  Selecting, sizing transformers in nonresidential buildings
126  Designing backup, standby, and emergency power for high-
performance buildings
136  Specifying paralleled generation systems
148  Your questions answered: Emergency, standby, and backup generators
Sponsored by
The role of NFPA 110 and its
interaction with other codes
The coordination of requirements for emergency and standby power  The role of NFPA
110 and its
systems between different code publications has improved over recent
interaction with
years, but variations still exist that can cause confusion. It is important for other codes
engineers to review all the applicable codes and standards to develop a
TLC Engineering &
full picture of the requirements for standby systems. While NFPA 110-2019: Schneider Electric –
Standard for Emergency and Standby Power Systems defines system levels, Partners at Nemours
types, and classifications, it does not determine what situations will require
any given level, type, or class. Design engineers must coordinate the When and how
to design parallel
requirements of NFPA 110 with other sections of the NFPA, International generators
Building Code, Facility Guidelines Institute, and other authority having
Creating a life-saving
jurisdiction requirements as applicable. environment through
the IoT-enabled
D esigners have to tackle myriad codes and standards when designing emergency EcoStruxure™
architecture and
and standby power systems for mission critical facilities, such as hospitals, data platform
centers, and critical operations facilities. In a hospital, for example, designers must re-
Data center design
view NFPA 99-2018: Health Care Facilities Code; NFPA 101-2018: Life Safety Code; NFPA considerations
110-2019: Standard for Emergency and Standby Power Systems; NFPA 70-2017: National
Content archive
Electrical Code (NEC), Articles 445, 517, 700, 701, 702, and 708; and the Facility Guide-
lines Institute (FGI) (or equivalent enforced standard). This list is not exhaustive, as there
are backup system requirements elsewhere in the code and some authorities having ju-
risdiction (AHJ) enforce different codes or have their own amendments impacting back-
up system design.
3
The role of NFPA 110 and its interaction with other codes
Codes do not have standardized

designations for emergency and
standby power systems. System  The role of NFPA
110 and its
designations used in different lo- interaction with
cations of the code include emer- other codes
gency, essential, nonessential, TLC Engineering &
legally required, backup power, Schneider Electric –
Partners at Nemours
optional standby, standby, Lev-
el 1, and Level 2. The following
information will focus primarily on When and how
to design parallel
the requirements for emergency and standby Medium-voltage emergency substations on large
generators
compuses increase the challenge of meeting
power systems as defined in NFPA 110. mandated power within specific time fames.
Image courtesy: WSP USA
environment through
System designations the IoT-enabled
NFPA 110 covers emergency power supplies (EPS) and emergency power supply systems EcoStruxure™
architecture and
(EPSS). The EPS consists of the source of electric power. The EPSS consists of the EPS platform
and all necessary components up to and including the transfer switches. Components
Data center design
include conductors, disconnecting and overcurrent protective devices, transfer switches,
considerations
controls, and supporting devices.
Content archive
System levels
NFPA 110 is applicable to systems designated Level 1 or Level 2 as defined by the code.
A Level 1 system failure to perform could result in the loss of human life or serious injury.
Examples of Level 1 loads include egress illumination, fire alarm and detection systems,
fire pumps, and emergency communication systems. Level 1 systems correspond well to
the requirements of NEC, Article 700: Emergency Systems. 4
A Level 2 system failure to perform could cause hazards or hamper rescue or firefighting
operations. Examples of Level 2 loads include HVAC systems, ventilation, smoke-remov-
al systems, sewage disposal, and industrial processes. Level 2 systems correspond with  The role of NFPA
110 and its
the requirements of NEC, Article 701: Legally Required Standby Systems. The NEC goes interaction with
on to define a third level of standby power systems with Article 702: Optional Standby other codes
Systems. The systems described under NEC, Article 702, do not fall under the purview of TLC Engineering &
NFPA 110 and are not considered Level 1 or Level 2 systems. Schneider Electric –
Partners at Nemours
System classifications
NFPA 110 establishes definitions for the classification of an EPSS. Classification identifies When and how
to design parallel
the amount of time for which an EPSS will operate at its rated load without being refu- generators
eled or recharged. It is important to note that this requirement of the NFPA is based on
rated load, not calculated or actual load, which means that a system designed with 50%
environment through
spare capacity still requires enough fuel to handle the full-rated load of the system for the IoT-enabled
the time determined by its classification. EcoStruxure™
architecture and
platform
The classes established by NFPA 110 range from Class 0.083 (5 minutes) to Class X, which
Data center design
is a designation used to require an amount of time, usually between 48 and 96 hours, as
considerations
defined elsewhere in the code or as determined by the designer based on system re-
quirements. Health care facilities, for example, fall under Class X per NFPA 99. The actual Content archive
time requirement chosen by the designer or AHJ should include an evaluation of past
outages, fuel-delivery challenges, and environmental conditions. It also should include
an evaluation of the facility’s seismic design category, although this recommendation
was moved to the Appendix of NFPA 110 during the 2013 version rather than in the code
body where it was previously.
5
System types
NFPA 110 then defines types of EPSS. The type identifies the maximum amount of time
that an EPSS will allow the load terminals of the transfer switch to be without acceptable
 The role of NFPA
110 and its
power, which is defined as the power quality, capacity, and stability as required by the interaction with
other codes
load to ensure proper system functions. EPSS types range from Type U (basically uninter-
ruptible) to Type M (manual). Examples of system types are the NFPA 99 requirement for TLC Engineering &
Type 10 (10 seconds) systems for health care essential systems, which corresponds to the Schneider Electric –
Partners at Nemours
NEC, Article 517.32(B), requirement for hospital life safety and critical branches, and the Children’s Hospital
NFPA 101 requirement for Type 60 (60 seconds) systems for ventilation of smoke-proof
When and how
enclosures, which corresponds to the NEC 701.12 requirement for legally required stand- to design parallel
by systems. Manual systems (Type M) are further discussed in NEC, Article 702 Optional generators
Standby Systems, and NEC, Article 517.35(B).
environment through
While NFPA 110 defines emergency and standby system class, type, and level and out- the IoT-enabled
EcoStruxure™
lines requirements for those designations, NFPA 110 does not determine which specific
architecture and
systems fall under these categories beyond the general Level 1 and 2 designations out- platform
lined above. Designers must turn to other codes to determine the required system for
Data center design
any given application. considerations
Content archive
Energy sources
NFPA 110 allows the use of only liquefied petroleum, liquefied petroleum gas, or natu-
ral/synthetic gas for EPS energy sources. NFPA 110 does not require onsite storage in
situations where an offsite supply, such as a natural gas line, will reliably provide for the
continuous operation of the EPS for the duration defined by the classification. The ex-
ception to this, which is only for Level 1 systems, is that locations with a high probability
6
of interruption of offsite fuel must have onsite fuel storage of adequate capacity for the
full duration of the class. The offsite fuel still can be the primary source, but provisions
are required for automatic transfer between the sources during a failure. Before using  The role of NFPA
110 and its
an offsite source, it also is crucial that designers review other applicable codes and stan- interaction with
dards. For example, FGI 2.1-8.3.3.1 states that where onsite fuel storage is required by other codes
other codes, a minimum of 24 hours shall be provided for hospitals. TLC Engineering &
Schneider Electric –
Partners at Nemours
When providing onsite fuel storage, NFPA 110, Chapter 5.5.3, requires the main fuel tank
is to have a minimum capacity of 133% of the full-load running of the EPS for the dura-
tion specified by the class. This requirement has three separate aspects that can trip up When and how
to design parallel
designers, as follows: generators
• The 133% requirement is intended to account for fuel that may be used in tests or
environment through
over time and for fuel in the top and bottom of the tank that may be below the us- the IoT-enabled
able level or above the fillable level. EcoStruxure™
architecture and
platform
• The requirement is for the EPS at full load, not design or actual load.
Data center design
considerations
• The requirement applies to the main fuel tank and does not allow for calculation of
day tanks or fuel-pipe volume. It is important for designers to understand that only Content archive
the main tank can contribute to the required fuel volumes. While the code does not
directly address installations with multiple main tanks, it can be assumed that any
primary tank volumes can contribute to the required volume.
While NFPA 110 does allow the use of natural gas for both Level 1 and Level 2 systems,
it is key for designers to realize the limitations of sources and how they interact with the 7
required system type as

defined above. For ex-
ample, employing a large  The role of NFPA
110 and its
natural gas generator for interaction with
a Level 1, Type 10 system other codes
may not result in power TLC Engineering &
availability and trans- Schneider Electric –
Partners at Nemours
fer within the required
10-second time frame.
The start-time delay for When and how
to design parallel
natural gas is primarily generators
due to fuel travel time to
get to the ignition cham-
environment through
ber. This delay is a safety the IoT-enabled
measure to reduce the EcoStruxure™
It can be particularly challenging for small architecture and
risk of gas explosions on start-up. Large natural platform
emergency systems to achieve required
gas generators also experience difficulties meet- coordinaton. Image courtesy: WSP USA
Data center design
ing the requirements from NFPA 110, Chapter
considerations
5.6.3.1.2, to handle the full-rated load in a single step.
Content archive
Other design variables, such as using medium-voltage emergency distribution via sub-
stations upstream of the transfer switches, add additional challenges to meeting a
10-second maximum time frame. In these large systems, the start time associated with
large generators, the energization time of large substation transformers, and the long
run lengths of distribution wiring all combine to increase the difficulty of achieving sta-
ble power in under 10 seconds. In addition to the challenges above, NFPA 110, Chapter 8
6.2.5, requires that a minimum 1-second delay be provided on most systems before ini-
tiating the EPS to reduce nuisance starts of the EPS system. (The time delay is 0.5 sec-
ond for gas turbine units.) This required time delay is counted as part of the overall time  The role of NFPA
110 and its
allowance provided by the EPSS type and effectively reduces the allowed start time from interaction with
10 seconds to 9 seconds. other codes
TLC Engineering &

EPS and EPSS locations Schneider Electric –
Partners at Nemours
When designing Level 1 systems indoors, NFPA 110 requires EPS installations to be in
a 2-hour-rated room. The code allows only EPS, EPSS, and other systems that directly
serve the space to be in the room. When and how
to design parallel
generators
The code further requires that EPSS equipment is not be installed in the same room with
the normal service equipment where the service equipment is rated 1,000 amp or higher
environment through
and more than 150 V to ground. While the EPSS equipment is not required by NFPA 110 the IoT-enabled
to be in a 2-hour-rated room, designers should carefully review the requirements of NEC, EcoStruxure™
architecture and
Article 700.10(D), for high-rise and assembly occupancies as well as the requirements platform
of IBC for smoke control, stair pressurization, and fire/access elevators associated with
Data center design
high-rise construction. The requirements of these other codes may necessitate a rated
considerations
room depending on the project.
Content archive
The EPSS room must also be designed and located to minimize damage from flooding,
sewer backups, and other events. This requirement doesn’t necessarily preclude EPSS
systems from being installed in basements or similar spaces, but it does require that the
design professional consider these elements and take corrective actions as necessary to
reduce environmental risks to the emergency systems. Corrective actions could include
floor drains and raised equipment, for example. This requirement is also not intended to 9
preclude the installa-

tion of fire-suppres-
sion systems in these  The role of NFPA
110 and its
rooms. NFPA 110, interaction with
Chapter 7.11.2, does other codes
prohibit the use of TLC Engineering &
carbon dioxide or Schneider Electric –
Partners at Nemours
halon systems unless
combustion air is
ducted from outside When and how
to design parallel
the structure. It also generators
prohibits automatic
dry-chemical systems unless they are certified by the Generator paralleling-gear installations
environment through
require testing of load shed, load
EPS manufacturer to not impede EPS operation. the IoT-enabled
optimization, and other operational
EcoStruxure™
features. Image courtesy: WSP USA
architecture and
For indoor installations, NFPA 110, Chapter 5.6.5.6, platform
requires the remote manual stop to prohibit accidental actuation and to be located
Data center design
outside of the room containing the EPS. While NFPA 110 does not give specific require-
considerations
ments for the emergency power off (EPO) to be in a secured location, designers should
consider the hazards created by an unauthorized person having access to an EPS EPO. Content archive
Consideration should also be given to the needs of first responders in an emergency.
When designing EPS systems in outdoor enclosures, NFPA 110 requires that enclosures resist
the entrance of snow and rain and allows for EPSS equipment to be mounted within the EPS
enclosure. For outdoor installations, the remote manual stop is required to be located out-
side of the enclosure and of a type to prohibit inadvertent actuation. Again, care should be 10
taken by the designer to limit the access of unqualified personnel to the EPO button.
Coordination requirements  The role of NFPA

110 and its
Chapter 6.5.1 of NFPA 110 requires that the overcurrent protective devices (OCPDs) in interaction with
the EPSS be coordinated. Coordination of OCPDs ensures that the devices closest to other codes
the overcurrent event trip before upstream devices. This ensures, for example, that a 20- TLC Engineering &
amp branch circuit will not cause an upstream feeder circuit breaker to trip, which would Schneider Electric –
Partners at Nemours
result in the loss of emergency loads that were not associated with the overcurrent event.
The Appendix of the code further clarifies that the OCPDs should be coordinated to the
extent practical and mentions that full coordination may not be feasible or practical. When and how
to design parallel
generators
For health care facilities, both NFPA 99, Chapter 6.7.2.2.2, and NEC, Article 517.30(G),
clarify the requirements for the essential electrical system OCPDs to be coordinated for
environment through
the period beyond 0.1 second, but there is no corresponding clarification for other types the IoT-enabled
of facilities. NEC, Articles 700.32, 701.27, and 708.54, all require systems to be selective- EcoStruxure™
architecture and
ly coordinated by an engineer or another qualified person, but it leaves the exact level platform
of coordination up to the discretion of the qualified individual. A strict reading of the
Data center design
NEC, Article 100, definition of selective coordination states that coordination needs to
considerations
be achieved for the full range of available overcurrents and OCPD opening times. This
leaves the door open for coordination requirements less than 0.1 second, depending on Content archive
AHJ interpretation. A requirement for full coordination less than 0.1 second could ef-
fectively reshape an entire electrical system and, in many cases, may not be possible to
achieve due to technological limitations of the OCPDs operating at such high speeds.
Understanding the above coordination requirements and AHJ interpretation is critical

to the design of emergency and standby systems. These requirements have the largest 11
proportional impact on the

smallest systems. For small
facilities, it is overcurrent  The role of NFPA
110 and its
coordination that often will interaction with
determine your overall system other codes
size, and not the load experi- TLC Engineering &
enced by the system. OCPDs Schneider Electric –
Partners at Nemours
have improved dramatically in
recent years, with ever thin-
ning and more precise trip When and how
to design parallel
curves, but it is still difficult to achieve coordination Paralleled generators have new NFPA 100
generators
requirements for initial acceptance and
with small thermal magnetic circuit breakers.
maintenance testing. Image courtesy: Creating a life-saving
WSP USA environment through
Revised requirements the IoT-enabled
The 2016 edition of NFPA 110 includes a few notable modifications as mentioned below. EcoStruxure™
architecture and
A new requirement in Chapter 7.9.13 prohibits the use of automatically actuated valves platform
in the fuel-supply and return lines. This requirement is aimed at reducing the susceptibil-
Data center design
ity of an EPS to single points of failure in the fuel lines. The code does not preclude the
considerations
use of manual valves.
Content archive
The 2016 edition also calls for facilities with multiple automatic transfer switches (ATS) to
rotate the monthly EPS test-initiating ATS (Chapter 8.4.3.1). This requirement helps en-
sure that, over time, all ATS are reaffirmed to be able to successfully initiate the EPS. This
requirement does not include a maximum time that an ATS can go without initiating the
generator, so it could still take years for a facility to rotate through all ATS in very large
campus installations. 12
Specific initial acceptance-testing requirements for parallel systems (Chapter 7.13.4.1.3)

and follow-up maintenance testing (Chapter 8.3.5) also have been added to the code.
The initial testing requirement calls for verifying paralleling and load-shed functionality.  The role of NFPA
110 and its
interaction with
The maintenance-testing requirements call for the same maintenance as transfer switch- other codes
es but add a requirement for verifying that system controls will operate as intended. The TLC Engineering &
Appendix further clarifies that load optimization, load shed, and other operating features Schneider Electric –
Partners at Nemours
should be tested and adjusted over time as ATS load profiles change. This Appendix
note emphasizes important added variables for systems with paralleled generators that
facilities need to be aware of. It also is vital for designers of renovations and EPSS mod- When and how
to design parallel
ifications to understand how revisions may require this load-shedding and optimization generators
strategy to be adjusted even when they aren’t adding new EPSS equipment.
environment through
While the coordination of requirements for emergency and standby power systems the IoT-enabled
between different code publications has improved over recent years, variations still exist EcoStruxure™
architecture and
that can cause confusion. It is important for designers to review all the applicable codes platform
and standards to develop a comprehensive perception of the requirements for standby
Data center design
systems.
considerations
While NFPA 110 defines system levels, types, and classifications, it does not determine Content archive
what situations will require any given level, type, or class. Designers must coordinate
the requirements of NFPA 110 with other sections of the NFPA, IBC, FGI, and other AHJ
requirements as applicable.
Jared Mason is a senior electrical engineer and project manager at WSP USA.
13
TLC Engineering & Schneider Electric – Nemours Children’s Hospital
The role of NFPA 110

and its interaction with
other codes
 TLC Engineering
& Schneider
Electric – Partners
at Nemours
 When and how

to design parallel
generators
environment through
the IoT-enabled
EcoStruxure™
architecture and
platform
TLC Engineering & Schneider Electric – Partners
at Nemours Children’s Hospital Data center design
considerations
With redundancy, resiliency and safety in mind, TLC Engineering
was chartered to design a lifesaving environment for children, Content archive
families, and the medical teams at Nemours Children’s Hospital.
Discover why TLC & Nemours chose to work with Square D by
Schneider Electric as their preferred vendor to provide a safe,
reliable, and connected space.
14
When and how to design
parallel generators
When designing generator systems, electrical engineers must ensure The role of NFPA 110
that generators and the building electrical systems that they support are other codes
appropriate for the specific application. Whether providing standby power
TLC Engineering &
for health care facilities or prime power for processing plants, engineers Schneider Electric –
must make decisions regarding generator sizing and whether generators Partners at Nemours
should be paralleled.
 When and how
A s the demand for reliable electricity grows, so does the need for reliable backup to design parallel
generators
power sources. Factors that affect the decision to provide paralleling capability for
standby generator systems include required or desired redundancy, additional capacity, Creating a life-saving
maintainability, upgrades to existing systems, and future growth. environment through
the IoT-enabled
EcoStruxure™
Building systems are increasing in efficiency, but the use of technology and the desire for architecture and
platform
standby emergency power is steadily increasing. Information technology (IT) equipment,
mechanical cooling equipment, building lighting (required and convenience), and branch Data center design
power loads are all adding to the total emergency power load. considerations
Content archive
It is typical for a building to grow in physical size or increase in electrical demand over its
lifespan. Designing systems for future growth requires cost-vs.-benefit discussions with
the client and usually includes considering 5-, 10-, and 20-year campus master planning
visions and goals.
Systems capable of paralleling generators inherently provide system-growth capability

with minimum system downtime or outages, which can be a high-value investment. The 15
When and how to design parallel generators

other codes
TLC Engineering &

Partners at Nemours
 When and how

to design parallel
generators
environment through
the IoT-enabled
EcoStruxure™
architecture and
Figure 1: Generators are shown in
platform
potential expandability of paralleled generator systems
a revised room in a hspital patient
also offers the owner the ability to purchase additional Data center design
tower, with the fuel-oil trench and
considerations
generator capacity as it is needed. discharge plenum visible. Image
courtesy: IMEG Corp. Content archive
Environmental conditions
Generators are electromechanical machines that require maintenance and experience a
wide range of environmental conditions. Generators with unit-mounted radiators require
a large volume of outside air to be brought into the room and across the generator sur-
faces. Outside air temperatures vary widely, as does humidity. Outside air also can con-
tain moisture, humidity, pollen, dust, insects, and other particulates.
16
Generators themselves generate substantial heat, vibration, and noise. These factors
combine to create adverse environmental conditions that will inevitably affect system
reliability. It is not recommended to provide screens or filters on intake grills and louvers and its interaction with
to alleviate dust and debris, as those methods will become clogged and prevent proper other codes
airflow, which may adversely affect a generator system’s cooling and capacity.
TLC Engineering &
Maintenance Partners at Nemours
When generator systems need to be maintained, they are taken out of service. This pre-
vents them from starting during an outage and necessitates careful planning by the facil-  When and how
ity, which usually results in work performed outside of normal hours that require interim to design parallel
generators
life safety measures.
environment through
The 2016 edition of NFPA 110: Standard for Emergency and Standby Power Systems, the IoT-enabled
Article 7.13, requires that additional sources of standby emergency power be available EcoStruxure™
during generator maintenance. Paralleled generator systems may provide redundancy of architecture and
platform
required systems. This means that regular maintenance can be provided during periods
of convenience, which increases the likelihood that maintenance will be performed on a Data center design
considerations
regular basis.
Content archive
Expansion of existing systems
Projects that replace or add onto existing emergency power supply (EPS) and emergency
power supply systems (EPSS, as defined in NFPA 110) are the most challenging because
they require the design team to consider all impacts to the existing system. Emergency
power system outages are the most difficult aspects to manage for health care, critical
operations, industrial, or mission critical buildings.
17

other codes
TLC Engineering &

Partners at Nemours
 When and how

to design parallel
generators
environment through
the IoT-enabled
EcoStruxure™
architecture and
platform
Taking a standby emergency power system offline for Figure 2: This shows fuel-oil
polishing—along with two day Data center design
any amount of time will require risk assessments with the tanks—where fuel oil routes through considerations
owner. These events usually trigger disaster preparedness the in-floor trench. Image courtesy:
Content archive
IMEG Corp.
teams that outline plans for managing electrical outages
without available backup power. Impacts to IT, phone, radio, lighting and egress, and
hazardous safety will need to be discussed and evaluated by the team. The nature of the
risk also is affected by the proposed duration of the emergency power system outage
and how long it would take to make connections in the event of a utility outage.
Due to these complications, providing a new EPS and EPSS may be easier than trying to
18
add to an existing system. Additionally, when connecting to existing distribution sys-

tems, impacts to selective coordination should be evaluated as they will likely impact
how and where connections will be made. and its interaction with
other codes
Design choices TLC Engineering &
Once the decision is made to go with paralleled generator systems, the design team Schneider Electric –
and owner must make decisions regarding physical location, fuel system requirements, Partners at Nemours
electrical-demand requirements, system voltage requirements, desired system redun-
dancy, and the amount of future growth potential. Determining the operating voltage  When and how
of the EPS and EPSS is sometimes, but not always, done by matching the normal ser- to design parallel
generators
vice voltage.
environment through
Discrepancies occur when EPSS are located long distances from the building(s) they the IoT-enabled
serve, and in these cases, higher voltages—even medium-voltage levels—may be pru- EcoStruxure™
dent. Generator voltages above 208/120 V in the United States will result in the need architecture and
platform
for step-down transformers, which influence the generator load and take up physical
space. Unless long feeder distances justify medium-voltage levels, the EPS voltage Data center design
considerations
usually matches the mechanical equipment, process equipment, and any traction ele-
vator systems. Content archive
Paralleling
Generator sets can be paralleled through traditional switchgear with motor-operated
breakers and paralleling logic housed inside the switchgear or in remote cabinets. An-
other option is to allow the unit-mounted generator controllers to provide paralleling
control, which can reduce total system cost. Onboard paralleling allows the design team
to plan for a modular generator system, which can allocate space for several additional 19
generator packages that can be

added as the load increases.
While stand-alone paralleling gear and its interaction with
usually comes at a higher initial other codes
cost, this equipment can readily
TLC Engineering &
provide an entire system overview, Schneider Electric –
and the breakers and controls can Partners at Nemours
be located in a conditioned space
outside of the EPS room. Due to  When and how
the environmental conditions a to design parallel
generators
generator must encounter, it can be
Design for
prudent to provide physical protec- Creating a life-saving
environment through
tion of sensitive electronics. the IoT-enabled
Connectivity
EcoStruxure™
Conversely, unit-mounted controllers architecture and
platform
offer enhanced redundancy because
Applications for Large and Critical Facilities the loss of one system controller Data center design
considerations
may not eradicate the paralleling
Hospitals, data centers, and industrial facilities all have a critical need for digital
energy management solutions. The Schneider Electric™ Digital Applications function of the remaining genera- Content archive
Design Guide is your resource for designing, specifying, and building digital
Range name here
solutions backed by our EcoStruxure™ Power technology.
tors. Because of the likelihood of
Discover how to: electronic failures, paralleling sys-
1) Address end user needs with the most appropriate applications
tems for critical functions should be
2) Implement applications while accommodating system design constraints
3) Select the required products able to be manually controlled. The
steps required to manually start,
Get the Guide
connect, and parallel generators
© 2019 Schneider Electric. All Rights Reserved. Life Is On Schneider Electric is a trademark and the property of Schneider Electric SE, its subsidiaries and affiliated companies.
should be clearly defined and made 20

readily available in the generator control space. Proper training should be provided to all
approved personnel.
Location other codes
Choosing a physical location for the EPS and EPSS requires a lot of forethought and
TLC Engineering &
consideration of impacts on adjacent systems. Besides the physical requirements of the Schneider Electric –
electrical distribution equipment, generators are noisy, smelly, and large. An acoustical Partners at Nemours
noise study should be performed to determine the maximum permissible noise levels,
and the first step in this process is determining the expected noise levels from the gen-  When and how
erator system. The components of the generator system that are of acoustical concern to design parallel
generators
include the radiator fan, exhaust system, engine and generator surfaces, and system
vibration levels. Creating a life-saving
environment through
the IoT-enabled
Once the noise sources are characterized, the design team must evaluate the permissi- EcoStruxure™
ble noise levels in adjacent rooms, buildings, and properties. Noise-abatement solutions architecture and
platform
include muffler systems with high-insertion-loss values, acoustical louvers with sufficient
free-area and sound-attenuation levels, and building walls with sufficient transmission Data center design
considerations
loss. Vibration isolation of hanging piping can alleviate structure-borne noise transfer.
Content archive
Further, the proximity of generator exhaust to the building’s ventilation-air intakes must
be taken into consideration along with prevailing wind directions. When generator sys-
tems are located within the building they serve, providing isolation of noise, heat, and
exhaust fumes adds significant cost and complexity to the building design.
EPS and EPSS locations should also take flooding into consideration, including efforts
from firefighting as defined in the 2012 edition of NFPA 99: Health Care Facilities Code, 21
6.4.1.1.8. Wind-speed graphs, such as wind rose graphs, can identify possible problemat-
ic building orientations in which high wind speeds and directions could have an adverse
impact on the EPSS. and its interaction with
other codes
Airflow TLC Engineering &
Paralleled generator systems may include unit-mounted radiators or remotely located ra- Schneider Electric –
diators. The airflow requirements of unit-mounted radiator fans are large, and the gener- Partners at Nemours
ator set’s capacity is affected by the number of restrictions to radiator airflow, also known
as pressure drop. The generator set will be designed to accommodate a maximum level  When and how
of pressure drop experienced by the radiator fan. Factors that influence airflow pressure to design parallel
generators
drop include intake-louver free area and design, air-pathway configuration, and dis-
charge-plenum louver-free area and design. Creating a life-saving
environment through
the IoT-enabled
Acoustical louvers further increase system pressure drop and available free area, which EcoStruxure™
can result in louver systems that are several times larger in dimension than the radiator architecture and
platform
itself. When the design allows for future growth, the designers should consider how ad-
ditional air will be brought into the EPS room as generator capacity grows. Data center design
considerations
When generator exhaust systems are required to comply with U.S. Environmental Pro- Content archive
tection Agency (EPA) Tier 4 requirements, there are several additional components that
greatly increase the physical size of the exhaust system. If the owner is considering oper-
ating the generators in a manner that requires EPA Tier 4 compliance, the design team
should allocate space for these additional systems.
Fuel oil
Code requirements for fuel system capacity vary based on the system’s class designation 22
outlined in NFPA 110-2016, Article 4.1, and requirements in the Facility Guidelines Insti-
tute guidelines that define run times for health care facilities. Depending on the nature
of the building and the availability of fuel oil, the owner and design team will need to and its interaction with
determine the desired run time duration. other codes
TLC Engineering &

The maximum amount of fuel oil that can be stored in a building is driven by NFPA 110, Schneider Electric –
Article 7.9.5, which limits capacity to an aggregate total of 660 gal/building. Day tanks Partners at Nemours
are offered with 150%-capacity leak capture and leak detection. Fuel-oil supply pumps
are usually located on the day tank and are configured in a duplex-type system where a  When and how
redundant pump is provided. Achieving long run times (24 hours or greater) will require to design parallel
generators
additional fuel-storage capacity, which can be exterior aboveground or belowground
fuel tanks sized to meet or exceed run time demands. Both concepts have benefits and Creating a life-saving
environment through
risks, which will vary based on location. the IoT-enabled
EcoStruxure™
Additionally, when storing large amounts of fuel for disaster preparedness, the fuel will architecture and
platform
need to be routinely cleaned to maintain fuel integrity. It is common for a fuel-polishing
system to be provided, which will provide active filtering of the entire storage system on Data center design
considerations
a scheduled basis. These systems are not very expensive and they prolong fuel life as
well as the generator’s fuel filter life and reliability. Content archive
When routing fuel lines for paralleling generator systems, the design team must consider
the reliability of the fuel-return piping. The ideal scenario is one in which fuel-oil-return
piping is routed to allow for gravity-only drainage. This means that the fuel storage is
below the level of the generator system and return-fuel piping is routed below the lev-
el of the generator and day tank return connections. This arrangement is benefited by
recessed trenches for fuel-line routing to eliminate trip hazards within the room. 23
When fuel-return piping is routed above the level of the generator and the day tank, a
fuel-oil-return pump is required along with solenoid valves and, sometimes, check valves. and its interaction with
In a properly designed fuel-oil system, the only time fuel oil would be returned to the other codes
main tank is in the event of a control failure where the supply pumps are stuck in the
TLC Engineering &
“on” condition. Schneider Electric –
Partners at Nemours
It is not recommended to install the day tank at a level above the generator, as addition-
al check valves are required to prevent flooding the engine. Fuel-oil-storage systems are  When and how
commonly shared with boilers and act as secondary fuel sources for building heating sys- to design parallel
generators
tems. Designers must ensure that generator fuel systems do not pose negative impacts
on boiler fuel-oil systems. Creating a life-saving
environment through
the IoT-enabled
Richard Vedvik is a senior electrical engineer and acoustics engineer at IMEG Corp. He EcoStruxure™
is a member of the Consulting-Specifying Engineer editorial advisory board. architecture and
platform
Data center design

considerations
Content archive
24
Creating a life-saving environment
through the IoT-enabled
EcoStruxure™ architecture and The role of NFPA 110
platform other codes
TLC Engineering &

An EcoXpert™ partnership helps to deliver a world-class healthcare facility
Partners at Nemours
that boosts efficiency, comfort, safety, and reliability. Children’s Hospital
When and how
N emours Children’s Hospital (NCH) in Orlando, Florida is a renowned, state-of-the-art

hospital that provides care for children with the most-complex illness and diseases.
The hospital is just five years old but already, families seeking world-class pediatric care
to design parallel
generators
have come from more than 70 different countries.

 Creating a
life-saving
environment
through the
NCH, part of the Nemours Chil-
IoT-enabled
dren’s Health System, was de- EcoStruxure™
signed through close consultation architecture and
platform
with families, many of whom had
extensive experience with the Data center design
considerations
pediatric health system. “We see
many patients from all over the Content archive
world,” said Nelson Roque, Di-
rector of Facilities and Operations
for Nemours. “Some of our kids
unfortunately are here for many
weeks and months at a time.”
25
Creating a life-saving environment through the IoT-enabled EcoStruxure™
Goal
To create a world-class power system for a new The role of NFPA 110
hospital that ensures a safe, comfortable envi- and its interaction with
other codes
ronment for staff and patients.
TLC Engineering &
Story Partners at Nemours
Through collaboration and innovation, four Children’s Hospital
Schneider Electric partners, including one From the way rooms are organized to When and how
Master EcoXpert partner, worked with Nemours be as homelike as possible, to the spe- to design parallel
from the early design stages, gaining an inti- cial visits from characters from our local generators
mate understanding of the hospital’s needs and
the unique power requirements of the ad-
theme parks, everything at Nemours is  Creating a
designed to create a calm, comfortable, life-saving
vanced medical equipment that would be used. environment
and efficient place of recovery for the through the
children and their families. IoT-enabled
Solution EcoStruxure™
An integrated system built on the IoTenabled architecture and
This attention to excellence extends to platform
EcoStruxure for Healthcare architecture and the hospital’s world-class medical tech-
platform, enabling critical operational efficiency Data center design
nology, and even to the electrical system considerations
for this world-renowned healthcare facility. that powers this technology and the
Content archive
building itself.
Results
A LEED® Gold facility built on a pure, contin- Building a team
uous source of power that provides Nemours For hospitals, a pure, continuous source
Children’s Hospital with reliable electricity for of power in a healing environment is es-
their equipment to function 24/7. sential. It not only provides reliable elec- 26
tricity for equipment to func-

tion 24/7, it also provides staff,
patients, and parents with the and its interaction with
constant peace-of-mind nec- other codes
essary to maintain a successful
TLC Engineering &
recuperation environment. Schneider Electric –
Electricity is the fuel that runs Partners at Nemours
the hospital.
When and how
to design parallel
To create the power system at
generators
Nemours, the hospital called
on a team consisting of Schnei-  Creating a
life-saving
der Electric EcoXpert™ and environment
other technology partners. through the
Through collaboration, innovation, and expertise, these four firms combined resources IoT-enabled
EcoStruxure™
to make the Nemours vision for reliable power a reality. architecture and
platform
Located in Sanford, Florida, MC2 is a system integrator and Schnei- Data center design
der Electric Master EcoXpert™ partner, certified in building manage- considerations
ment systems and critical power solutions. Schneider Electric’s EcoX-
Content archive
pert Partner Program unites the world’s leading technology providers,
who exhibit best-in-class system integration competencies in facility
optimization, reliable infrastructures and energy management, with its customers around
the globe. Stronger together, this partnership cultivates collaboration, connects exper-
tise, and delivers best-in-class services and solutions.
27
TLC Engineering for Architecture is a national company with expertise

in mechanical, electrical, plumbing, structural, life safety, fire protec-
tion, and energy services. Recognized in 2017 as a JUST organization, and its interaction with
TLC’s priority is to excel in their design policies and operations. JUST other codes
is a transparency platform that informs about an organization’s opera-
TLC Engineering &
tions, financial and community investments. Schneider Electric –
Partners at Nemours
Borrell Electric is a well-respected electrical service and con-
tracting company headquartered in Tampa, Florida. Serving When and how
to design parallel
its community for more than 70 years, Borrell’s electrical
generators
technicians are trained and certified on the latest tools and technologies to drive innova-
tion. Borrell proudly boasts a 100% safety record with zero unscheduled outages.  Creating a
life-saving
environment
Graybar is a premier supplier of Schneider Electric products through- through the
out Florida and nationwide. An employee-owned Fortune 500 com- IoT-enabled
EcoStruxure™
pany, Graybar brings deep electrical expertise to everyday power distribution problems as architecture and
well as more complex design and build projects like Nemours Children’s Hospital. Graybar platform
also specializes in advanced supply chain management services for contractors. Data center design
considerations
By working with the hospital from the very beginning, and then executing as a closeknit
Content archive
team, the partner companies were able to deliver a world-class power system for Ne-
mours that has met all expectations.
Delivering the solution

“One of the things the owner chartered us to do was to provide redundancy, resiliency,
and safety in our designs,” said Vince Rea, Senior Electrical Engineer for TLC Engineering. 28
To meet this goal, the team developed a 3-tiered

power system: critical power for operating the- up to The role of NFPA 110
aters and essential medical devices; equipment
30%
power for AC, heat, and less critical medical other codes
equipment; and convenience power for lighting
operational efficiency TLC Engineering &
and amenities in non-critical locations. Incoming Schneider Electric –
24 x 7
electrical power is treated, cleaned, and diverted Partners at Nemours
to the relevant areas and equipment across the
hospital, ensuring that sensitive medical equip- Assured, continuous When and how
to design parallel
ment receives the precisely the power needed. power operation
generators
The system is monitored in real-time by the hos-
pital facilities team. Using EcoStruxure Power  Creating a
life-saving
Monitoring Expert and Asset Advisor software, the team can use graphical interfaces to environment
view status, generate reports, and identify potential problems before they affect safety through the
or operational efficiency. IoT-enabled
EcoStruxure™
architecture and
Resilience is built into the system. If Nemours ever experiences a power loss, genera- platform
tors automatically fire up and restore full power within ten seconds across the hospital. Data center design
Schneider Electric switchgear equipment transfers power to the hospital circuits, mon- considerations
itoring and controlling power levels to ensure proper distribution of energy to medical
Content archive
equipment, lighting, and utilities.
While ten seconds to power is rapid response, it is not fast enough in some cases. To fill
this gap, Schneider Electric Uninterruptable Power Supplies (UPS) provide immediate
battery backup for sophisticated critical assets such as incubators, life-support machines,
operating theatres, radiology equipment, and data center. 29

other codes
TLC Engineering &

Partners at Nemours
When and how

to design parallel
generators
 Creating a
life-saving
environment
through the
IoT-enabled
EcoStruxure™
The result is a smooth-running, seamless power system that keeps the lights on and the architecture and
equipment operating at all times. Caregivers and patients are never left in the dark, and platform
care is never interrupted by power failures. Data center design
considerations
The power of partnership Content archive
Technology played a big part in the project success. But everyone involved agrees that
it’s also a story about partnership and teamwork. “I think the teamwork in innovative
product delivery from the very beginning, with all the team players involved, made it a
very successful project,” said Vince Rea of TLC Engineering.
Justin Powell of Graybar agreed: “There’s a personal touch, a personal relationship.” 30


other codes
TLC Engineering &

Partners at Nemours
When and how

to design parallel
generators
 Creating a
life-saving
And Roy Hoffman, president of MC2, stated “We were able to work as an extension of environment
Schneider Electric.” through the
IoT-enabled
EcoStruxure™
Brian Keemer, Manager Healthcare Solutions, Schneider Electric, summed it up well: architecture and
“There is nothing more important than our value chain. All the players are doing what platform
they’re supposed to do. They have expertise, skill, and capability in the industry to make Data center design
it all happen for the end user so that kids are able to receive their treatment, and doctors considerations
are able to deliver their skill.”
Content archive
Most importantly is the satisfaction that the people at Nemours Children’s Hospital feel
about the performance of the partners who delivered this project. As Mr. Roque states,
“We have a number of partners that we have had for many years. I can’t brag enough
about the quality of work that they’ve done for us. To have partners that continue to per-
31

other codes
TLC Engineering &

Partners at Nemours
When and how

to design parallel
generators
 Creating a
life-saving
environment
through the
IoT-enabled
EcoStruxure™
form for you every time, every job, is wonderful.” architecture and
platform
IoT-enabled system architecture and platform that drives op- Data center design
erational and energy efficiency considerations
EcoStruxureTM by Schneider Electric is our open, interoperable, IoT-enabled system Content archive
architecture and platform. EcoStruxure delivers enhanced value around safety, reliability,
efficiency, sustainability, and connectivity for our customers. We leverage technologies
in IoT, mobility, sensing, cloud, analytics, and cybersecurity to deliver Innovation at Every
Level. This includes Connected Products, Edge Control, and Apps, Analytics & Services.
EcoStruxure has been deployed in 450,000+ installations, with the support of 9,000 sys-
32
tem integrators, connecting over 1 billion devices.

Connected Products and its interaction with
The Internet of Things starts with the best things. Our IoT-enabled best-in-class con- other codes
nected products include breakers, drives, UPSs, relays, sensors, and more. Devices with
TLC Engineering &
embedded intelligence drive better decision-making throughout operations. Schneider Electric –
Partners at Nemours
Edge Control
Mission-critical scenarios can be unpredictable, so control of devices at the edge of the When and how
to design parallel
IoT network is a must. This essential capability provides real-time solutions that enable
generators
local control at the edge, protecting safety and uptime.
 Creating a
life-saving
Apps, Analytics & Services environment
Interoperability is imperative to supporting the diverse hardware and systems in build- through the
ing, data center, industry, and grid environments. EcoStruxure enables a breadth of ag- IoT-enabled
EcoStruxure™
nostic Applications, Analytics & Services for seamless enterprise integration. architecture and
platform
Data center design

considerations
Content archive
33
Data center design considerations
This article provides guidelines on distribution systems’ levels of
redundancy, the correct generator rating to use, and whether solar power
can be used in a data center.
other codes
O ver the past several years, mission critical clients seem to be asking the same
series of questions regarding data center designs. These questions relate to the
best distribution system and best level of redundancy, the correct generator rating to
TLC Engineering &
Partners at Nemours
use, whether solar power can be used in a data center, and more. The answer to these Children’s Hospital
questions is “It depends,” which really doesn’t help address the root of their questions. When and how
For every one of these topics, an entire white paper can be written to highlight the attri- to design parallel
butes and deficiencies, and in many cases, white papers are currently available. However, generators
sometimes a simple and concise overview is what is required rather than an in-depth Creating a life-saving
analysis. The following are the most common questions that this CH2M office has re- environment through
the IoT-enabled
ceived along with a concise overview.
EcoStruxure™
architecture and
What is the best system topology? platform
There isn’t a single “best” system topology. There is only the best topology for an indi-  Data center design
vidual data center end user. The electrical distribution system for a data center can be considerations
configured in multiple topologies. While the options and suboptions can be myriad, the Content archive
following topologies are commonly deployed (see Figure 1).
• 2N: Simply designing twice as much equipment as needed for the base (i.e., N) load
and using static transfer switches (STS), automatic transfer switches (ATS), and the
information technology (IT) and HVAC equipment’s dual cording to transfer the load
between systems. The systems are aligned in an “A/B” configuration and the load
is divided evenly over the two systems. In the event of failure or maintenance of one
34

other codes
TLC Engineering &

Partners at Nemours
When and how

to design parallel
generators
environment through
the IoT-enabled
EcoStruxure™
architecture and
platform
system, the overall topology goes to an Figure 1: Conceptual one-line configurations of
N level of redundancy.
electrical topologies highlighting redundancy.  Data center design
Image courtesy: CH2M considerations
Content archive
• 3M2: This topology aligns the load over more than two independent systems. The
distributed redundant topology is commonly deployed in a “three-to-make-two”
(3M2) configuration, which allows more of the capacity of the equipment to be used
while maintaining sufficient redundancy for the load in the event of a failure (see Fig-
ure 2). The systems are aligned in an “A/B/C” configuration, where if one system fails
(e.g., A), the other two (B and C) will accept and support the critical load. The load is
evenly divided with each system supporting 33.4% of the load or up to 66.7% of the
35
equipment rating. In the event of a component failure or maintenance in one system,

the overall topology goes to an N level of redundancy. In theory, additional systems
could be supplied, such as 4M3 or 5M4, but deployment can significantly complicate and its interaction with
the load management and increases the probability of operator error. other codes
TLC Engineering &

• N+1 (SR): The shared-redundant (SR) topology concept defines critical-load blocks. Schneider Electric –
Each block is supported 100% by its associated electrical system. In the event of Partners at Nemours
maintenance or a failure, the unsupported equipment would be transferred to a
backup system that can support one or two blocks depending on the design. This When and how
to design parallel
backup system is shared across multiple blocks, with the number of blocks support-
generators
ed being left to the design team but typically in the range of 4:1 up to 6:1.
environment through
• N+1 (CB): The common-bus (CB) redundant system is like the shared redundant the IoT-enabled
system in that the IT equipment’s A and B sources are connected to an N+1 unin- EcoStruxure™
terruptible power supply (UPS) source, but in the event of a failure or maintenance architecture and
platform
activities, the load is transferred to a raw power source via STS. The raw power
source has the capability of being backed up by generators that are required to be  Data center design
considerations
run during maintenance activities to maintain the critical load.
Content archive
The above topologies assume a low-voltage UPS installation. However, similar systems
can be developed using a medium-voltage UPS. Beyond the redundancy configuration,
these low-voltage UPS topologies also can be evaluated on ease-of-load management,
backup power generation, their ability to deploy and commission initially and when
expanding, first costs and total cost of ownership, physical footprint of the equipment
comprising the topology, and time to construct the initial installation as well as expan-
sion of the system. 36
A commonality between the

different topologies presented
is the need to transfer load be- and its interaction with
tween systems. No matter the other codes
system topology, the requirement
TLC Engineering &
to transfer load between electri- Schneider Electric –
cal systems—either for planned Partners at Nemours
maintenance activities, expan-
sions, or failure modes—must be When and how
to design parallel
done. Load management refers
generators
to how the load is managed
across multiple systems. Creating a life-saving
environment through
the IoT-enabled
2N topology. The premise be- EcoStruxure™
hind a 2N system is that there are architecture and
platform
two occurrences of each piece of
critical electrical equipment to al-  Data center design
considerations
low the failure or maintenance of
any one piece without impacting Content archive
the overall operation of the data
center IT equipment. This config-
uration has a number of impacts:
Figure 2: Mission critical-electrical room
showing overhead conduit routing andd
• Load management: Among the topologies pre- complexity for a 3.6-MW, three-to-make-
sented here, 2N has a relatively simple load-man- two (3M2) distribution system. Image
agement scheme. The system will run inde- courtesy: CH2M
37
pendently of other distribution systems and can be sized to accommodate the total
demand load of the IT block and associated HVAC equipment, minimizing the failure
zone. The primary consideration for load management is to ensure the total load and its interaction with
doesn’t overload a single substation/UPS system. other codes
TLC Engineering &

• Backup power generation: This topology uses a 2N backup generation with the Schneider Electric –
simplest of schemes: having the generator paired to the distribution block. Each Partners at Nemours
generator is sized for the entire block load and will carry 50% of the load under nor-
mal conditions. For large data centers, the option exists to parallel together multiple When and how
to design parallel
generator sets to create an “A” backup source and parallel together an equal num-
generators
ber of generators to create a “B” backup source, distributing power via two different
sets of paralleling switchgear. Typically, this is more expensive due to the addition of Creating a life-saving
environment through
paralleling switchgear and controls. Selection of the voltage class usually depends the IoT-enabled
on the size of load as well as physical space and cost to route cable from the gener- EcoStruxure™
ator to the switchgear. The ability to parallel generators tends to be limited by the architecture and
platform
paralleling switchgear bus ampacity ratings as well as short-circuit ratings. Beyond
6,000 amps at 480 V, consider using 15-kV-class generators.  Data center design
considerations
• Deployment: Each 2N system can be designed to accommodate a discrete IT block. Content archive
This allows multiple systems to be deployed independently, facilitating procurement,
construction, commissioning, and operations with no impact to existing or future
systems.
• First cost/TCO: The 2N system requires twice the quantity and capacity of electri-
cal equipment than the load requires, causing the system to run at nominally 50%
of nameplate capacity. Due to the nature of how electrical equipment operates, this 38
tends to cause the equipment to run at a lower efficiency than can be realized in oth-
er topologies. An additional impact of the 2N system topology is that the first cost
tends to be greater because of the quantity of equipment. Also, because there are and its interaction with
additional systems in place, the ongoing operational and maintenance costs tend to other codes
be greater.
TLC Engineering &
• Spatial considerations: Because it generally has the most equipment, the 2N con- Partners at Nemours
figuration typically has the largest physical footprint. However, this system is the
simplest to construct as a facility is expanded, thereby minimizing extra work and When and how
to design parallel
allowing the facility to grow with the IT demands.
generators
• Time to market: As has been discussed, this system will have more equipment to Creating a life-saving
environment through
support the topology, therefore there may be additional time to construct and com- the IoT-enabled
mission the equipment. The systems are duplicates of each other, which allows for EcoStruxure™
construction and commissioning efficiencies when multiple systems are installed, architecture and
platform
assuming the installation teams are maintained.
 Data center design
considerations
Distributed redundant (3M2) topology. The premise behind a 3M2 system is that there
are three independent paths for power to flow, each path designed to run at approxi- Content archive
mately 66.7% of its rated capacity and at 100% during a failure or maintenance event.
This configuration is realized by carefully assigning load such that the failover is properly
distributed among the remaining systems.
This configuration has a number of impacts to the distribution:
• Load management: The load management for the 3M2 system should be carefully 39
considered. The
load will need to
be balanced be- and its interaction with
tween the A, B, other codes
and C systems to
TLC Engineering &
ensure the critical Schneider Electric –
load is properly Partners at Nemours
supported with-
out overloading When and how
to design parallel
any single system.
generators
Load management
of a system like Creating a life-saving
environment through
this can be aided the IoT-enabled
by a power-moni- EcoStruxure™
toring system. architecture and
platform
Figure 3: Installation of feeders
• Backup power generation: This topology follows the

and control cables for a .6-MW
3M2 distribution system. Image
considerations
normal power flow and uses a 3M2 backup generation courtesy: CH2M
where the generator is paired to the distribution block. Content archive
Each generator is sized for the entire block load and will carry 66.7% of their capacity
under normal conditions. Parallel generator configurations are rarely used for 3M2
systems. Like 2N systems, the selection of the voltage class depends on the size of
load as well as physical space and cost to route cable from the generator to switch-
gear (see Figure 3).
• Deployment: Each 3M2 system can be designed to accommodate a discrete IT 40

block. Expansion within a deployed 3M2 system is exceptionally challenging and

difficult, if not impossible to commission. Deployment of multiple 3M2 systems is the
best option for addressing expansion and commissioning. and its interaction with
other codes
• First cost/TCO: The 3M2 system requires about 1.5 times the capacity of electrical
TLC Engineering &
equipment than the load requires and runs at 66.7% of its rated capacity. Because Schneider Electric –
the equipment is running at a higher percentage, the 3M2 system tends to be more Partners at Nemours
energy-efficient than the 2N, but less efficient than either of the shared redundant
systems. An additional impact of the 3M2 system topology is that lower-capacity When and how
to design parallel
equipment can be used to support a similar size IT block, thereby causing the sys-
generators
tem to have a higher cost per kilowatt to install. However, if the greater capacity is
realized by either sizing the IT blocks large enough to realize the benefits of this Creating a life-saving
environment through
topology or by installing two IT blocks on each distribution system, then there will the IoT-enabled
be a lower first cost. Essentially, the 2N system needs two substations and associated EcoStruxure™
equipment for each IT block while the 3M2 system would need only three substa- architecture and
platform
tion systems to support the IT block. First-cost savings is in addition to operational
savings because there are fewer pieces of equipment to maintain. And the energy  Data center design
considerations
savings is because the equipment is running at a higher efficiency.
Content archive
• Spatial considerations: Similar to the first-cost discussion above, the spatial layout
can either be smaller or larger than a 2N system depending on how the topology is
deployed and how many IT blocks each system supports.
• Time to market: The balance between the IT blocks supported by each system and
the quantity of equipment will have an impact on the time to market, though the bal-
ance for this system is unlikely to be significant. The additional equipment should be 41
balanced against smaller

pieces of equipment,
allowing faster installa- and its interaction with
tion time per unit. The other codes
systems are duplicates of
TLC Engineering &
each other, which allow Schneider Electric –
for construction and com- Partners at Nemours
missioning efficiencies
when multiple systems When and how
to design parallel
are installed, assuming
generators
the installation teams are
maintained. Creating a life-saving
environment through
the IoT-enabled
N+1 shared redundant (N+1 SR). The premise Figure 4: Rendering of a CH2M design EcoStruxure™
behind the N+1 SR system is that each IT block is of a data center, conference center, and architecture and
offce buildings for Saudi Airlines. Image platform
supported by one primary path. In the event of main- courtesy: CH2M
tenance or a failure, there is a redundant but shared  Data center design
considerations
module that provides backup support. The shared module in this topology has the same
equipment capacities and configuration as the primary power system, minimizing the Content archive
types of equipment to maintain.
For example, if six IT blocks are to be installed, then seven distribution systems (substa-
tions, generators, and UPS) will need to be installed for an N+1 system. This N+1 system
can easily be reconfigured to an N+2 system with minimal impact (procuring eight sys-
tems in lieu of seven). This reconfiguration would allow the system to provide full reserve
capacity even while a system is being maintained. 42
This configuration has several impacts to consider:

• Load management: The N+1 SR system has the simplest load management of to- other codes
pologies presented. As long as the local UPS and generator are not overloaded, the
TLC Engineering &
system will not be overloaded. Schneider Electric –
Partners at Nemours
• Backup power generation: This topology follows the normal power flow and uses
an N+1 SR backup generation where the generator is paired to the distribution When and how
to design parallel
block. Each generator is sized for the entire block load, with the SR generator also
generators
sized to carry one block. Parallel generation can be used for block-redundant sys-
tems. However, carefully consider the need for redundancy in the paralleling switch- Creating a life-saving
environment through
gear. True N+1 redundancy would require redundant paralleling switchgear. Howev- the IoT-enabled
er, this level of redundancy while on generator power may not be required. EcoStruxure™
architecture and
platform
• Deployment/commissioning: The deployment of the N+1 SR system is modular be-
cause each system functions independently. However, commissioning a new system  Data center design
considerations
with an existing redundant system may be challenging if the redundancy needs to
be always available for the critical load. In the event of a multiple-fault scenario (mul- Content archive
tiple generators failing to operate or multiple UPS failing to support the load while
generators start), the faults will cascade and overload the redundant system. There
are multiple ways to mitigate this risk (load-management tripping breakers or inhibit-
ing the STS), but the concern is valid. Any of the methods implemented to prevent a
cascading failure will cause some IT loads to go offline.
• First cost/TCO: For a large-scale deployment (i.e., exceeding two modules), the 43
N+1 SR system has the lowest installed cost per kilowatt of the systems explored
here that have full UPS protection for both the normal and redundant power distri-
bution systems, due to the lower quantity of equipment. In addition, less equipment and its interaction with
should also result in lower ongoing operation and maintenance costs. other codes
TLC Engineering &

• Spatial considerations: The N+1 SR layout will have the smallest spatial impact. Schneider Electric –
Additional distribution is required between modules as well as a central location to Partners at Nemours
house the redundant system.
When and how
to design parallel
• Time to market: The balance between the IT-blocks distribution system and the
generators
quantity of equipment will have an impact on the time to market. However, due to
the fact that the N+1 SR has the smallest quantity of equipment, this configuration Creating a life-saving
environment through
potentially has the shortest time to market of any system explored so far. This timing the IoT-enabled
is further supported due to system duplicates, which should allow for construction EcoStruxure™
and commissioning efficiencies on the subsequent installations, assuming the teams architecture and
platform
are maintained.
considerations
N+1 common bus (N+1 CB). The premise behind the N+1 CB system is there is one pri-
mary path that supports each IT block. This path also has an N+1 capacity UPS to facili- Content archive
tate maintenance and function in the event of a UPS failure. The system is backed up by
a simple transfer switch system with a backup generator.
This configuration has a number of impacts on the distribution:
• Load management: Similar to the N+1 SR system, the load management for the
N+1 CB is simple. As long as the local UPS/generator combination is not overload- 44
ed, the system will not be overloaded.

• Backup power generation: Like the previous topology, there is a generator paired and its interaction with
to each distribution block including the redundant block. other codes
TLC Engineering &

• Deployment/commissioning: The deployment of the N+1 CB system is a modular Schneider Electric –
deployment because each system functions independently. The only location where Partners at Nemours
existing work has to be tested with the new equipment is on the common bus sys-
tem. When and how
to design parallel
generators
• First cost/TCO: The N+1 CB system potentially has the lowest installed cost per
kilowatt of any of the systems. This lower cost is due to a combination of lower quan- Creating a life-saving
environment through
tities of UPS and generators coupled with simpler distribution. Additionally, less the IoT-enabled
equipment means ongoing operation and maintenance costs should be lower as EcoStruxure™
well. architecture and
platform
• Spatial considerations: The N+1 CB layout will have a small spatial impact. Addi-  Data center design
considerations
tional distribution is required between modules as well as a central location to locate
the central bus system (transfer switches and generator). Content archive
• Time to market: Similar to the N+1 SR system, the N+1 CB has significantly fewer
pieces of equipment than the 2N or 3M2 systems. This equipment count should sup-
port a faster time to market. However, it is difficult to determine which of the N+1
systems would have a quicker time to market.
The above topology descriptions only highlight a few systems. There are other topol- 45
ogies and multiple

variations on these
topologies. There isn’t and its interaction with
a ranking system for other codes
topologies; one isn’t
TLC Engineering &
better than another. Schneider Electric –
Each topology has pros Partners at Nemours
and cons that must be
weighed against the When and how
to design parallel
performance, budget,
generators
schedule, and the ulti-
mate function of each Creating a life-saving
environment through
data center. the IoT-enabled
EcoStruxure™
What generator architecture and
Figure 5: The photo shows the outdoor platform
rating should be used for a data installation of the power electronic switch of
center? the medium-voltage UPS at Michigan State  Data center design
University. Image courtesy: CH2M
considerations
Generators need to be able to deliver backup
power for an unknown number of hours when util- Content archive
ity power is unavailable. To help select the appropriate generator, manufacturers have
developed ratings for engine-generators to meet load and run time requirements under
different conditions. The International Standards Organization (ISO) Standard 8528-2005,
Reciprocating Internal Combustion Engine Driven Alternating Current Generating Sets,
tries to provide consistency across manufacturers. However, the ISO standard only de-
fines the minimum requirements. If the generator is capable of a higher performance,
then the manufacturer can determine the listed rating. To complicate generator ratings 46
even more, some industries have their own ratings specific to that industry and applica-
tion. These various ratings can make selecting the correct generator type complicated.
There are four ratings defined by ISO-8528: other codes
TLC Engineering &

1. Continuous power: designed for a constant load and unlimited operating hours; Schneider Electric –
provides 100% of the nameplate rating for 100% of the operating hours. Partners at Nemours
2. Prime power: designed for a variable load and unlimited running hours; provides When and how
to design parallel
100% of nameplate rating for a short period but with a load factor of 70%; 10%
generators
overload is allowed for a maximum of 1 hour in 12 hours and no more than 25
hours/year. Creating a life-saving
environment through
the IoT-enabled
3. Limited running: designed for a constant load with a maximum run time of 500 EcoStruxure™
hours annually; same nameplate rating as a prime-rated unit but allows for a load architecture and
platform
factor of up to 100%; there is no allowance for a 10% overload.
considerations
4. Emergency standby power: designed for a variable load with a maximum run time
of 200 hours/year; rated to run at 70% of the nameplate. Content archive
The generator industry also has two additional ratings that are not defined by ISO-8528:
mission critical standby and standby. Mission critical standby allows for an 85% load
factor with only 5% of the run time at the nameplate rating. A standby-rated generator
can provide the nameplate rating for the duration of an outage assuming a load factor of
70% and a maximum run time of 500 hours/year.
47
Data center designs

assume a constant
load and worst-case and its interaction with
ambient tempera- other codes
tures. This does not
TLC Engineering &
reflect real-world op- Schneider Electric –
eration and results in Partners at Nemours
overbuilt and excess
equipment. Further- When and how
to design parallel
more, it is unrealistic
generators
to expect 100% load
for 100% of the op- Creating a life-saving
environment through
erating hours, as the the IoT-enabled
generator typically EcoStruxure™
requires maintenance architecture and
platform
and oil changes after every 500 hours of run time. Figure 6: The photo shows the outdoor
Realistically during a long outage, the ambient tem- installation of the switchgear in an enclosure  Data center design
used in conjunction with the power considerations
perature will fluctuate below the maximum design
electronic switch. Image courtesy: CH2M
temperature. Similarly, the load in a data center is Content archive
not constant. Based on research performed by Caterpillar, real-world data center appli-
cations show an inherent variability in loads. This variability in both loads and ambient
temperatures allows manufacturers to state that a standby-rated generator will provide
nameplate power for the duration of the outage and it’s appropriate for a data center
application. However, if an end user truly desires an unlimited number of run hours, then
a standby-rated generator is not the appropriate choice.
48
What type of transformer is best?

The type of transformer to be used for a data center is constantly questioned and chal-
lenged by end users trying to understand if they should invest in a high-performance and its interaction with
transformer or not. There are two categories of distribution transformers: dry-type and other codes
liquid-filled. Within each category, there are several different types. The dry-type catego- TLC Engineering &
ry can be subdivided into five categories with the following features: Schneider Electric –
Partners at Nemours
1. Open-wound transformers apply a layer of varnish on heated conductor coils and
bake the coils until the varnish cures. When and how
to design parallel
generators
2. Vacuum-pressure impregnated (VPI) transformers are impregnated with a high-tem-
perature polyester varnish, allowing for better penetration of the varnish into the
environment through
coils and offering increased mechanical and short-circuit strength. the IoT-enabled
EcoStruxure™
architecture and
3. Vacuum-pressure encapsulated (VPE) transformer windings are encapsulated with platform
silicon resin typically applied in accordance with a military spec and used in loca-
tions exposed to salt spray, such as shipboard applications with the U.S. Navy. VPE  Data center design
considerations
transformers are superior to VPI transformers, with better dielectric, mechanical,
and short-circuit strength. Content archive
4. Encapsulated transformers have open wound windings that are insulated with ep-
oxy, which makes them highly resistant to short-circuit forces, severe climate condi-
tions, and cycling loads.
5. Cast-coil-type transformers have windings that are hermetically sealed in epoxy to

49
provide both electri-

cal and mechanical
strength for higher and its interaction with
levels of perfor- other codes
mance and environ-
TLC Engineering &
mental protection Schneider Electric –
in high-moisture, Partners at Nemours
dust-laden, and
chemical environ- When and how
to design parallel
ments.
generators
For liquid-filled trans- Creating a life-saving

environment through
formers, various types the IoT-enabled
of fluids can be used to EcoStruxure™
insulate and cool the architecture and
Figure 7: The photo shows the outdoor
platform
transformers. These include less-flammable fluids,
nonflammable fluids, mineral oil, and Askarel.
installation of the medium-voltage UPS
system with batteries in an enclosure  Data center design
considerations
equipped with a heating and ventilating
system and integrated air conditioning.
When put into the context of a mission critical envi- Content archive
Image courtesy: CH2M
ronment, two transformers stand out: the cast-coil
transformer due to its exceptional performance and the less-flammable liquid-immersed
transformer due to its dependability and longevity in commercial and industrial environ-
ments. While both transformer types are appropriate for a data center, each comes with
pros and cons that require evaluation for the specific environment.
Liquid-filled transformers are more efficient than cast coil. Because air is the basic cool- 50
ing and insulating system for cast coil transformers, they will be larger than liquid-filled
units of the same voltage and capacity. When operating at the same current, more ma-
terial and more core and coil imply higher losses for cast coil. Liquid-filled transformers and its interaction with
have the additional cooling and insulating properties associated with the oil-and-paper other codes
systems and tend to have lower losses than corresponding cast coil units.
TLC Engineering &
Liquid-filled transformers have an average lifespan of 25 to 35 years. The average lifes- Partners at Nemours
pan of a cast coil transformer is 15 to 25 years. Because liquid-filled transformers last
longer than cast coil, they save on material, labor to replace, and operational impact due When and how
to design parallel
to replacement.
generators
Recommended annual maintenance for a cast coil transformer consists of inspection, in- Creating a life-saving
environment through
frared examination of bolted connections, and vacuuming of grills and coils to maintain the IoT-enabled
adequate cooling. Most times, cleaning of the grills and coils requires the transformer to EcoStruxure™
be de-energized, which often leads to this maintenance procedure being skipped. The architecture and
platform
buildup of material on the transformer grills and coils can lead to decreased transformer
efficiency due to decreased airflow.  Data center design
considerations
Maintenance for a liquid-filled transformer consists of drawing and analyzing an oil sam- Content archive
ple. The oil analysis provides an accurate assessment of the transformer condition and
allows for a scheduled repair or replacement rather than an unforeseen failure. This kind
of assessment is not possible on a cast coil transformer. Additionally, omitting the oil
sampling does not decrease the transformer efficiency.
Cast-coil-type transformers have a history of catastrophic failures within data centers due
to switching induced transient voltages when switched by upstream vacuum breakers. 51
There has been significant research by IEEE committees, which resulted in guidelines for
mitigating techniques (i.e., resistive-capacitive [RC] snubbers) published in IEEE C57.142-
2010: IEEE Guide to Describe the Occurrence and Mitigation of Switching Transients and its interaction with
Induced by Transformers, Switching Device, and System Interaction. Liquid-filled trans- other codes
formers seem less susceptible to this problem, as there is no published data on their
TLC Engineering &
failure. Regardless of the transformer type installed, best industry practice is to perform a Schneider Electric –
switching transient study and install RC snubbers on the systems if warranted. Partners at Nemours
When a transformer fails, a decision must be made on whether to repair or replace it. When and how
to design parallel
Cast coil transformers typically are not repairable; they must be replaced. However, there
generators
are a few companies who are building recyclable cast coil transformers. On the other
hand, in most cases, liquid-filled transformers can be repaired or rewound. Creating a life-saving
environment through
the IoT-enabled
When a cast-coil transformer fails, the entire winding is rendered useless because it is EcoStruxure™
encapsulated in epoxy resin. Because of the construction, the materials are difficult and architecture and
platform
expensive to recycle. Liquid-filled transformers are easily recycled after they’ve reached
the end of their useful life. The steel, copper, and aluminum can be recycled.  Data center design
considerations
Cast-coil transformers have a higher operating sound level than liquid-filled transform- Content archive
ers. Typical cast coil transformers operate in the 64 to 70 dB range while liquid-filled
transformers operate in the 58 to 63 dB range. A decibel is a logarithmic function and
sound pressure doubles for every 3-dB increase.
Liquid-filled transformers have less material for the core and coil and use highly effective
oil and paper cooling systems, which allow them to be small in physical footprint and
weigh less than the corresponding cast coil unit. Because cast coil transformers are air- 52
cooled, they are often larger than their liquid counterparts assuming the same voltage
and capacity (kVA rating). Cast coil transformers have more core material, which implies
higher costs and losses. and its interaction with
other codes
Dry-type transformers have the advantage of being easy to install with fire-resistant
TLC Engineering &
and environmental benefits. Liquid-filled transformers have the distinct disadvantage of Schneider Electric –
requiring fluid containment. However, advances in insulating fluids, such as Envirotemp Partners at Nemours
FR3 by Cargill, a natural ester derived from renewable vegetable oil, is reducing the ad-
vantages of dry-type transformers. When and how
to design parallel
generators
For indoor installations of transformers, cast coil must be located in a transformer room
with minimum 1-hour fire-resistant construction in accordance with NFPA 70-2017: Na- Creating a life-saving
environment through
tional Electrical Code (NEC) Article 450.21(B). However, if less-flammable liquid-insulated the IoT-enabled
transformers are installed indoors, they are permitted in an area that is protected by an EcoStruxure™
automatic fire-extinguishing system and has a liquid-confinement area in accordance architecture and
platform
with NEC Article 450.23.
considerations
Traditionally less-flammable liquid-filled transformers are installed outdoors. However,
both types can be installed outdoors. This option of outdoor installation has the addi- Content archive
tional advantage of reducing data center cooling requirements. In this case, cast coil
transformers need to have a weatherproof enclosure and cannot be located within 12 in.
of combustible building materials per NEC Article 450.22. The liquid-filled transformer
must be physically separated from doors, windows, and similar building openings in ac-
cordance with NEC Article 450.27.
The choice between a cast coil and a less-flammable liquid-filled transformer can be a 53
challenging one to make. A liquid-filled transformer is a solid choice for a data center
application because it is more efficient, physically smaller and lighter, quieter, recyclable,
and has a longer lifespan. However, if the demand for high electrical and mechanical and its interaction with
performance is of the utmost concern, then cast coil would be the appropriate choice. other codes
TLC Engineering &

What IT distribution voltage should be used? Schneider Electric –
By now it’s well understood in the data center industry that 3-phase circuits can provide Partners at Nemours
more power to the IT cabinet than a single-phase circuit. However, the choice of distri-
bution voltage between 208 Y/120 V or 415 Y/240 V depends on the answers to several When and how
to design parallel
questions, such as:
generators
• How much power needs to be delivered to each IT cabinet initially, and what does Creating a life-saving
environment through
the power-growth curve look like for the future? the IoT-enabled
EcoStruxure™
• What are the requirements of the IT equipment power supplies? architecture and
platform
• Will legacy equipment be installed in the data center?  Data center design
considerations
• Can the facilities team decide on the power supplies to be ordered when new IT Content archive
equipment is purchased?
Let’s start with the power of a 3-phase circuit. A 208 Y/120 V, 3-phase, 20-amp circuit can
power up to a 5.7-kVA cabinet. Per NEC Article 210.20, branch-circuit breakers can be
used up to 80% of their rating, assuming it’s not a 100%-rated device. Therefore, a 208 V,
3-phase, 20-amp circuit can power a cabinet up to 5.7 kVA (20 amps x 0.8 x √3 x 208 V).
Now, if that same 20-amp circuit was operating at 415 Y/240 V, 3-phase, then that circuit 54
could power a cabinet up to 11.5 kVA (20 amps x 0.8 x √3 x 415 V). That’s more than twice
the power from the same circuit for no extra distribution cost.
If the specifications for the IT equipment can be tightly controlled, the decision to stan- other codes
dardize on 415 Y/240 V distribution is a pretty simple one. However, if the IT environment
TLC Engineering &
cannot be tightly controlled, the decision is more challenging. Currently, most IT power Schneider Electric –
supplies have a wide range of operating voltage, from 110 V to 240 V. This allows the Partners at Nemours
equipment to be powered from numerous voltage options while only having to change
the plug configuration to the power supply. However, legacy equipment or specialized IT When and how
to design parallel
equipment may have very precise voltage requirements, thereby not allowing for opera-
generators
tion at the higher 240 V level. To address this problem, both 208 Y/120 V and 415 Y/120
V can be deployed within a data center, but this is rarely done as it creates confusion for Creating a life-saving
environment through
deployment of IT equipment. the IoT-enabled
EcoStruxure™
The follow-on question typically asked is if the entire data center can run at 415 V, rather architecture and
platform
than bringing in 480 V and having the energy loss associated with the transformation to
415 V. While technically feasible, the equipment costs are high because standard HVAC  Data center design
considerations
motors operate at 480 V. Use of 415 V for HVAC would require specially wound motors,
thus increasing the cost of the HVAC equipment. Content archive
Must we install an emergency power-off system?

Emergency power-off (EPO) buttons are the fear of every data center operator. With the
push of a button, the entire data center power and cooling can be shut down. Because
of the devastation that activation of an EPO can cause, EPOs typically are designed with
a two- or three-step activation process, such as lifting a cover and pressing the button or
having two EPO buttons that must be activated simultaneously. These multistep options 55
assume that the authority having jurisdiction has provided approval for such a design.
However, EPOs are not necessarily required. The need for an EPO is typically triggered
by NEC Article 645.10, which allows alternative and significantly relaxed wiring methods and its interaction with
in comparison with the requirements of Chapter 3 and Articles 708, 725, and 770. These other codes
relaxed wiring methods are allowed in exchange for adding an EPO system and ensuring
TLC Engineering &
separation of the IT equipment’s HVAC occupancies from other occupancies. The princi- Schneider Electric –
ple benefit of using Article 645.10 is to allow more flexible wiring methods in the plenum Partners at Nemours
spaces and raised floors. However, if the wiring is compliant with Chapter 3 and Articles
708, 725, and 770, the EPO is not required. When and how
to design parallel
generators
Can we use photovoltaic systems to power our data center?
Corporations and data center investors are demanding sustainability be built into the Creating a life-saving
environment through
data center. The positive impact on public relations by showcasing a sustainable data the IoT-enabled
center can’t be underestimated, especially considering how much of a power hog data EcoStruxure™
centers can be. Additionally, many utility companies will offer incentives for the use of architecture and
platform
energy-efficient and sustainable technologies. An often-questioned item is whether
photovoltaic (PV) systems can be used to meet some of the sustainability requirements in  Data center design
considerations
a data center environment. The answer is yes, but a good understanding of PV systems
and the limitations and impacts on a data center are required prior to making the invest- Content archive
ment (see Figure 4).
The power production of PV equipment varies considerably depending on type and

location of the system installed. There are three main types of solar panel technolo-
gies. Crystalline silicon (c-Si) is the most common PV array type, along with thin-film and
concentrating PV. Thin-film is generally less efficient than c-Si, but also less expensive.
Concentrating PV arrays use lenses and mirrors to reflect concentrated solar energy onto 56
high-efficiency cells. Concentrating PV arrays require direct sunlight and tracking systems
to be most effective and are typically used by utility companies.
Solar cells are not 100% efficient. In the infrared region of light, solar cells are too weak other codes
to create electricity; and in the ultraviolet region of light, solar cells create heat instead
TLC Engineering &
of electricity. The amount of power that can be generated with a PV array also varies due Schneider Electric –
to the average sunshine (insolation, or the delivery of solar radiation to the earth’s sur- Partners at Nemours
face) along with the temperature and wind. Typically, PV arrays are rated at 77˚F, allowing
them to perform better in cold rather than in hot climates. As temperatures rise above When and how
to design parallel
77˚F, the array output decays (the amount of decay varies by type of system). Ultimate-
generators
ly, what this means is that the power generation of an array can vary over the course of
a day and year. Added to this are the inefficiencies of the inverter, and if used, storage Creating a life-saving
environment through
batteries. the IoT-enabled
EcoStruxure™
The physical space required to install the PV array can be significant. A simple rule is to architecture and
platform
assume 100,000 sq ft (about 2.5 acres) for a 1-MW PV-generating plant. However, this
does not include the space required for access or other ground-mounted appurtenanc-  Data center design
considerations
es. The total land required is better estimated at about 4 acres per MW. This estimate
assumes a traditional c-Si PV array (without trackers). Increase this area by 30%, for a total Content archive
of about 6 acres, if thin-film technology (without trackers) is used due to the inefficiencies
of the technology.
A PV system may or may not provide power during a utility power failure, depending on
the type of inverter installed. A standard grid-tied inverter will disconnect the PV system
from the distribution system to prevent islanding. The inverter will reconnect when util-
ity power is available. An interactive inverter will remain connected to the distribution 57
system, but it is designed to only produce power when connected to an external power
source of the correct frequency and voltage (i.e., it will come online under generator
power). Typically, interactive inverters include batteries to carry the system through pow- and its interaction with
er outages, therefore the system should be designed such that there is enough PV-array other codes
capacity to supply the load and charge the batteries.
TLC Engineering &
Most data centers do not have the necessary land to install a PV system, which substan- Partners at Nemours
tially offsets the power demand. Then there is the question of what happens when the
PV system is generating low or no power. Interactive inverters and deep-cycle storage When and how
to design parallel
batteries can be installed to cover these low-PV production periods, but this introduces
generators
new equipment, maintenance, and space requirements into the data center, thus creat-
ing more costs and more maintenance than may have been originally envisioned. Gen- Creating a life-saving
environment through
erally, data center sustainability is addressed more directly through efficient cooling and the IoT-enabled
electrical distribution systems. Sustainability achieved through solar power, while nice to EcoStruxure™
have, is generally not the focus of data center investments. architecture and
platform
The trend is to provide a PV system that offsets some of the noncritical-administration  Data center design
considerations
power usage. These systems are typically small (less than 500 kW) and can be located on
building rooftops, carports, and on the ground. They use a standard grid-tied inverter Content archive
connected through the administration electrical distribution system, which ultimately ties
into the site-distribution switchgear where the utility meter resides. A grid-tied inverter
system will disconnect from the utility if there is a failure or when on generator power.
Because the grid-tied inverter connection is downstream from the utility revenue meter,
a billing mechanism known as net metering generally is used. With net metering, owners
are credited for any electricity they may add to the grid when the PV production is great- 58
er than the site usage. Although in most data centers, the critical load dwarfs the non-
critical load; therefore it’s rare that a PV system would generate power on the grid. There
are differences between states and utility companies regarding the implementation, reg- and its interaction with
ulations, and incentives for net metering. Furthermore, there are some utility companies other codes
that perceive net metering as lost revenue and will not allow connection to their system.
TLC Engineering &
A great resource for PV and renewable energy in general is the National Renewable Partners at Nemours
Energy Laboratory. The NREL website provides information on PV research, applications,
publications as well as a free online tool, PVWatts, which estimates the energy produc- When and how
to design parallel
tion and cost of energy of grid-tied PV systems throughout the world. The PVWatts tool
generators
easily develops estimates of potential PV-installation performance.
environment through
A medium-voltage alternative to low-voltage UPS the IoT-enabled
Design topology evaluation also should consider the medium-voltage uninterruptible EcoStruxure™
power supply (UPS). Like the topologies using the low-voltage UPS, the medium-voltage architecture and
platform
UPS can be deployed in 2N, N+1, and 3N/2 configurations. Regardless of the topology
used, medium-voltage UPS systems offer advantages over low-voltage UPS systems.  Data center design
considerations
They generally are installed outdoors in containers, thereby minimizing the conditioned
building footprint. While not required, medium-voltage UPS topologies are typically Content archive
used for full-facility protection rather than using independent information technology
and mechanical-cooling UPS systems, further reducing the building footprint. Medi-
um-voltage UPS systems are large systems, starting at 2.5 MVA and scalable up to 20
MVA per UPS. Different manufacturers have different voltage offerings, but medium-volt-
age UPS systems can range from 5 kV up to 25 kV, with medium-voltage diesel rotary
UPS systems going as high as 34.5 kV.
59
In early 2018, Michigan State University is expected to complete construction on a new

25,000-sq-ft data center with 10,600 sq ft of server space and initially hosting about 300
server racks. This $46 million facility will use a medium-voltage UPS system, starting with and its interaction with
2.5 MW of critical power and the ability to increase in critical power as needed. The util- other codes
ity infrastructure is built to support an increase of load up to 10 MW. Figures 5, 6, and 7
TLC Engineering &
highlight the outdoor power electronic switch, switchgear, and the medium-voltage UPS Schneider Electric –
installed by the university. Partners at Nemours
Debra Vieira is a senior electrical engineer at CH2M, Portland, Ore., with more than 20 When and how
to design parallel
years of experience for industrial, municipal, commercial, educational, and military cli-
generators
ents globally.
environment through
the IoT-enabled
EcoStruxure™
architecture and
platform

considerations
Content archive
60
How to design a grounded
power supply system
Transformerless uninterruptible power supply (UPS) systems operate
ungrounded during power transfer to a backup source, but a robust
grounding design can accommodate the requirement of both grounded and
ungrounded systems.
I n any facility containing critical loads, whether related to life safety or sensitive com-
puter loads vital to facility operation, one of the most important pieces of equipment
specified in the design is the uninterruptible power supply (UPS), which uses stored ener-
gy to supply power to these critical loads when normal power is lost and a backup power
source is starting up to supply the building loads.
When selecting UPS modules to power critical loads in a facility, one key decision to
make is whether to use a UPS with or without input and/or output transformers. See Fig-
ures 2 and 3 for notional diagrams depicting a transformer-based and a transformerless
UPS module, respectively.
Over the last decade, transformerless UPS systems have grown rapidly in popularity,
eclipsing transformer-based designs. This transition is not surprising, as transformerless
modules offer many advantages over a UPS with transformers. The largest advantage is
efficiency. A UPS without transformers can see efficiency advantages of 5% or greater, as
compared with those with transformers. Not only does this mean lower electricity bills,
but it also represents lower heat loads in the room housing the UPS, resulting in reduced
HVAC requirements.
61
How to design a grounded power supply system
In facilities with a large amount of crit-

ical load, the savings can be dramatic.
Additionally, transformerless UPS sys-
tems reduce the weight and footprint
of each UPS module when compared
with transformer-based systems, re-
ducing the size and structural require-
ments of electrical rooms and leaving
more room for white space or other
portions of the building.
However, the output transformer of

a transformer-based UPS does pro-
vide an option that is not available
for transformerless UPS systems: The
electrical isolation provided by a trans-
former gives the opportunity to create
a separately derived neutral-to-ground
connection at the output of the UPS.
In certain situations—such as a sys-
tem served by an ungrounded delta
service, a service grounded through
a high-resistance ground, or systems in which Figure 1: The Bexar Mtreo 911 Network District
there is the potential that the two sources of a Regional Operations Center in San Antonio,
Texas, is a 2-story building that houses an
dual-input UPS may come from two indepen-
dent sources—it may be desirable to derive a
emergency call center. Image courtesy: Dror
Baldinger FAIA, Architectural Photography LLC 62
neutral at the UPS without a transformer, to provide the UPS with a stable ground refer-
ence that it can use for voltage regulation at its output and on its dc bus.
If such a neutral is not derived in a transformerless UPS system, then while the UPS bat-
tery is discharging during an input power failure and the UPS input circuit breaker is
open, the downstream system is operating ungrounded. In most installations, there will
be one or more downstream transformers, external to the UPS, served by the critical
power system. These downstream transformers are usually housed in a power distribu-
tion unit, and on their secondary side a grounded system can be derived, but that por-
tion of the system on the primary side will nonetheless be ungrounded during this peri-
od.
Most design engineers are used to working with grounded systems, and the prospect
of leaving a portion of the building ungrounded, even during a generally brief transition
period between input power failure and the facility backup power system starting up,
may seem worrisome. However, creating a safe, robust, and code-compliant ungrounded
power system is relatively simple, requiring only minor modifications from the grounding
and bonding systems required in any grounded power system.
Grounded versus ungrounded

To understand the special requirements of an ungrounded system, it is important to first
define what is meant by “grounded” and “ungrounded.” System grounding is achieved
by intentionally connecting a current-carrying conductor to the ground (i.e., the earth) or
to something that serves in place of the ground. Usually, this is achieved by connecting
the system’s neutral wire to the ground at the supply source, often the wye-connect-
ed secondary side of a transformer or the stator of a generator, as well as at the main 63
service-disconnecting means of the facility. Therefore, an ungrounded system is one in

which none of the current-carrying conductors are intentionally connected to the ground.
Grounded systems are typically preferred to ungrounded systems for several reasons.
Grounded systems stabilize voltage levels throughout the system, ensuring that all
equipment in the system operates under the same potential difference. This is particular-
ly important for a UPS, as it is tasked with precisely regulating voltage levels both at its
output and on its dc bus, and accurate voltage regulation requires a solid, stable refer-
ence to ground to maintain. Grounded systems also mitigate voltage spikes due to light-
ning strikes, help prevent potential differences between different pieces of equipment in
the system, and provide a circuit for ground-fault current to flow through the grounded
circuit conductors back to the supply source, allowing overcurrent protective devices to
quickly operate and isolate the fault.
NFPA 70: National Electrical Code (NEC), Article 250.4, provides general, perfor-
mance-based requirements for both grounded systems in 250.4(A) and ungrounded sys-
tems in 250.4(B). Grounded systems have five requirements: electrical system grounding,
grounding of electrical equipment, bonding of electrical equipment, bonding of electri-
cally conductive materials, and effective ground-fault current paths.
Notably, the four requirements of ungrounded systems listed in 250.4(B) are similar or
identical to the latter four requirements of grounded systems. Just as in grounded sys-
tems, ungrounded systems require that noncurrent-carrying conductive materials enclos-
ing electrical conductors or equipment, and those that are likely to become energized,
be connected to the ground through a low-impedance path. The overlap between these
two sets of requirements illustrates the notion that designing an ungrounded system is 64
not too terribly

different than de-
signing a ground-
ed one.
To understand
what the NEC
requires of a
transformerless
UPS system when
it is operating
ungrounded
during battery discharge, we must first identify Figure 2: This shows a conceptual single-line
drawing of a transformer-based uninterruptible
how this system is defined using the language power supply (UPS). A neutral is derived at the
of the NEC. When the UPS input circuit breaker secondary connection of the output transformer.
is open, the UPS is not connected to the up- Image courtesy: Page Inc.
stream power system, and therefore to the supply service, through any circuit conductors
other than those used for grounding and bonding.
It is important to note that even though the UPS enclosure and downstream equipment
may still be effectively bonded to the supply-source enclosure in this condition, the
system is not considered grounded unless a current-carrying conductor is connected to
ground. Therefore, the NEC defines the system in this state as a separately derived sys-
tem and the UPS batteries as a separately derived source. The grounding requirements
for ungrounded separately derived systems are defined in Article 250.30(B).
65
This section of the NEC

requires three components:
grounding electrode con-
ductor, grounding-elec-
trode system, and sup-
ply-side bonding jumper.
The latter of these compo-
nents is only required when
the source of a separately
derived system is located in
a separate enclosure than
the first disconnecting means. This will usually Figure 3: In this conceptual single-line
drawing of a tranformerless uninterruptible
not be the case for a UPS, as the output circuit power supply (UPS), a neutral cannot be
breaker of the UPS is normally housed in the UPS derived at the output of the inverter without a
transformer. Image courtesy: Page Inc.
enclosure.
All three of these grounding components are required in grounded separately derived
systems as well. Essentially, the UPS enclosure must be connected to ground through the
building’s grounding-electrode system by way of a grounding electrode conductor. This
connection in an ungrounded system serves as the grounding reference point for all con-
ductive equipment in the ungrounded system that does not carry current under normal
conditions.
Grounding the system

UPS manufacturers have a variety of solutions for the issue of how to ensure the UPS
maintains a reference to the ground during ungrounded conditions, to ensure that the 66
UPS voltage regulation remains stable. Some manufacturers derive a so-called “virtual
ground” at the common point of the input and output filters of the UPS to achieve this
purpose. This is often a standard feature, especially on newer UPS models, but an op-
tional accessory is required in some cases. When specifying a transformerless UPS, espe-
cially in a 3-phase, 3-wire system, take care when considering how it will operate under
ungrounded conditions.
The rules governing the grounding-electrode system and grounding electrode conduc-
tors are found in Part III of Article 250 of the NEC. The same grounding-electrode system
used for the building as a whole must also be used for any separately derived systems,
as per NEC 250.58, so all that is required is a connection between the building’s ground-
ing electrode and the UPS enclosure through a grounding electrode conductor. Accord-
ingly, all of the normal requirements for grounding-electrode system materials listed in
NEC 250.52 and installation listed in 250.53 apply in this situation.
Similarly, the rules governing grounding electrode conductors are not different between
grounded and ungrounded systems. NEC Articles 250.62 and 250.64 govern the ma-
terials and installation methods of grounding electrode conductors, respectively. The
required size of the grounding electrode conductors used must be determined through
the requirements of NEC 250.66, which vary based on the type of grounding electrode
used, the size of the largest ungrounded conductor or set of conductors in the system,
and the material of the grounding electrode conductor.
No matter the size of the system, the grounding electrode conductor must always be at
least as large as #8 AWG for copper or #6 AWG for aluminum, and unless superseded
by local amendments or authority having jurisdiction (AHJ) requirements, the grounding 67
electrode conductor is not required to be larger than #3/0 AWG for copper or 250 kcmil
for aluminum. Finally, requirements for bonding grounding electrode conductors to the
grounding-electrode system are covered in NEC 250.68.
Ungrounded systems
Thus far, the grounding rules discussed covering ungrounded systems are very similar to
those covering grounded systems. Indeed, if one employs a robust grounding design for
a normally grounded system and ensures that the UPS and battery-cabinet enclosures
are connected to the building’s grounding-electrode system through appropriately sized
grounding electrode conductors, almost all requirements for an ungrounded system
will be met when the UPS discharges its batteries and becomes an ungrounded system
during power transfer.
However, there is a key difference between the behavior of grounded and ungrounded
systems that imposes an additional requirement on ungrounded systems. This difference
appears when a single line-to-ground fault occurs in the system.
In a solidly grounded system, the connection of (usually) the neutral wire to ground at
the supply source means that a complete circuit will be formed when a line-to-ground
fault occurs. This allows a large amount of fault current to flow through the low-im-
pedance path created by the fault, causing an overcurrent protective device (OCPD)
equipped with ground-fault detection to operate and quickly isolate the fault.
In an ungrounded system, though, there is no circuit created when a single line-to-

ground fault occurs through which fault current can flow. Instead, the faulted conductor
simply becomes grounded and the line-to-line potentials between the faulted phase and 68
the other unfaulted phases become line-to-ground potentials. The value of the potential
difference between the phases, however, does not change. This will not have a notice-
able effect on the system’s performance when it occurs, but if the fault is left unrepaired
and a second line-to-ground fault occurs, this will result in a double line-to-ground fault,
drawing larger fault currents and creating the potential for greater damage to electrical
equipment and greater risk to personnel safety. As in grounded system, a phase-to-
phase fault in an ungrounded system will generate fault current and will typically cause
an overcurrent protective device to operate and isolate the fault.
To ensure that single line-to-ground faults do not go undetected, NEC 250.21(B) requires
that ungrounded systems be outfitted with ground detectors at a point as close as prac-
ticable to the system supply source. A ground detector monitors the potential difference
between the phase conductors of the system and the ground in the ungrounded portion
of the system to which it is connected. If a ground fault is present in the system, it emits
a visual and/or audible signal to alert operators or maintenance personnel. The opera-
tors can then initiate an orderly shutdown of the system so that the fault can be located
and repaired. This is especially important in a system served by a UPS, as orderly shut-
downs to critical loads are generally necessary to minimize risk to life safety or disruption
to business functions.
For example, it may be costly to initiate a shutdown of a critical computer system due
to the presence of a ground fault on the system, but it will certainly be less so than an
abrupt disconnection of power to those same computers. Most UPS systems will contain
a ground-detection mechanism, but it is important to verify this component is included
to ensure compliance with this requirement.
69
Detection of ground faults is especially important when a system becomes temporarily

ungrounded, such as while a transformerless UPS is discharging its battery due to an
input source failure, because it is likely to become grounded again when the input power
returns. When power is restored, either through a return of the utility source or due to a
generator source coming online, the UPS input circuit breaker will close and the system
will once again be grounded. If a ground fault is still present in the system when this
occurs, ground-fault current will flow through the fault. A ground detector in the UPS can
prevent this situation through a pre-emptive shutdown before fault current has a chance
to flow.
Detection
To ensure that a ground-fault detector can operate correctly throughout the unground-
ed system, it is important to verify compliance with the requirements for connection of
the UPS enclosure to the grounding-electrode system (discussed above), as well as the
requirements for bonding of metallic objects that do not carry current, covered in Part
V of NEC Article 250. This ensures that any point on the ungrounded system at which a
ground fault is likely to occur has a solid reference to the ground by way of the UPS en-
closure’s connection to the grounding-electrode system, and that the detector can accu-
rately detect a ground-fault condition.
In addition to the requirements for ungrounded ac systems, the NEC contains additional
requirements for ungrounded dc systems. This applies to the more common 2-wire dc
system, as Section 250.162(B) requires that all 3-wire dc systems be grounded. Of course,
a UPS will contain a dc system, namely the connection between the energy-storage
system and the output inverter. NEC Article 250.169 lists the requirements for an un-
grounded dc separately derived system. These requirements are similar to those of the 70
ac system, namely that the source enclosure be grounded through the building’s ground-
ing-electrode system by way of a grounding electrode conductor.
Note, though, that the sizing of the grounding electrode conductor for a dc system is
governed by a different section than that of an ac system, namely Section 250.166, which
requires that the grounding electrode conductor not be smaller than the largest con-
ductor supplied by the dc system. However, the same minimum- and maximum-required
grounding electrode conductor sizes for various installations apply to both dc and ac
systems. With most UPS installations, no special considerations are required for the dc
system, as typically the dc system is grounded at the battery enclosure, though it is im-
portant to verify this will be the case in a given installation.
In the event that special design considerations do need to be made, to consider what
may constitute the best practice for designing the grounding system for the dc portion
of a UPS during ungrounded system conditions, it may be useful to consider a somewhat
analogous situation: an ungrounded solar photovoltaic (PV) system.
Ungrounded PV systems are not uncommon for many of the same reasons that transfor-
merless UPS systems have become popular. Similar to the dc portion of a UPS system
during power transfer, an ungrounded PV system is an ungrounded dc system supplying
an ac system through an inverter. In fact, most PV systems have energy-storage systems
as part of the dc system, just as in a UPS. So while they are not requirements, the rules
and commentary of the NEC regarding PV systems contained in Article 690, particularly
Part III and Part V, can be instructive in thinking about grounding of the UPS energy-stor-
age system.
71
Of particular note is Section 690.15(D), which describes the requirements of equip-

ment-disconnecting means for PV systems. This section draws attention to a more gener-
al requirement found in Section 210.4(B), which requires that disconnecting means must
simultaneously disconnect all ungrounded conductors of the circuit to which it is con-
nected. Unlike a grounded system, in which the disconnection of the neutral wire does
not need to be simultaneous with the disconnection of the phase wires, in an unground-
ed system all wires must be simultaneously disconnected, as none are grounded. This is
not often a concern, but it is something to keep in mind when specifying the dc circuit
breaker protecting the energy-storage system of the UPS.
Ben Stevens is an associate electrical engineer at Page. He has worked for Page for 3
years and specializes in science and technology projects.
72
Critical power: hospitals and
data centers
Electrical engineers must consider many factors when designing backup,
standby, and emergency power systems. Specific requirements for
emergency power vary based on the mission critical facility, such as a
hospital or data center.
S afety, maintainability, code compliance, and economics play crucial roles in determin-
ing the topology of an emergency system for a critical facility. Specific requirements
for emergency power vary based on the mission critical facility. Hospitals and data cen-
ters have the one of the highest power availability requirements in the industry to ensure
minimal to no interruption of service in critical areas.
Availably is the ability of an item—under combined aspects of its reliability, maintain-

ability, and maintenance support—to perform its required function at a stated instant of
time or over a stated period of time. Reliability is the ability of a component or system to
perform required functions under stated conditions for a stated period of time.
As building typologies, hospitals and data centers are inherently distinct, one being
largely unoccupied and the other providing life-safety services to inpatients. The electri-
cal systems of hospitals appear to be trending more toward the robust designs found in
data centers.
Data centers house computer equipment used to process, analyze, and store electronic
data. Many businesses today rely on services performed within data centers to maintain
their own functionality, including day-to-day IT/communication infrastructure. The need
73
Critical power: hospitals and data centers
for electronic data is growing while the need Figure 1: A hospital served by an N+2 generator
to process or access this data at all times plant has two critical busway feeders to the
hospital wing where the operating rooms ae
also is increasing. Keeping these buildings located. The design allows for maintenance while
available 7×24 is a necessity, not a luxury. continuing to provide power. Image courtesy:
Casey Dunn Photography
Hospitals, on the other hand, are occupied, public buildings that provide critical care,
and depending on the type provided, these facilities must be available 7×24. While code
requirements compel hospitals to have emergency, standby generators primarily for 74
life-safety and essential reasons, they are not as stringent as the standard practices for
redundancy in data centers.
Maintaining demand for power

Demand for uninterruptable reliable power in data centers is now the norm. As most of
these facilities are not regulated by the government, their electrical system topology is
flexible and innovative. Guidelines, such as Uptime Institute and Telecommunications
Industry Association (TIA), are used industrywide.
Hospitals are increasingly requiring redundancy in their electrical systems, especially as

the use of computers and electronic equipment to save lives and improve the patient
experience has become integrated into critical care. As technology evolves, the number
and complexities of hospital systems requiring uninterruptable power steadily increases.
Many hospitals are growing their budgets to include support of electrical reliability and
availability of their systems. Examples of hospital equipment requiring continuous pow-
er include operating room machinery, life support, nurse calls, blood storage, fire alarm,
mass notification, etc.
Comparing hospitals, data centers

Hospitals and data centers have different type of requirements because they are serv-
ing distinct types of loads. The difference in their requirements influence their electrical
topologies as described below:
• Data centers predominately use dual-corded equipment for critical loads, while hos-
pitals rely on single-corded systems. 75
• Data centers use central-

ized uninterruptable power
supply (UPS) systems for
critical loads, while hospi-
tals have preferred to use
internal equipment UPS
to allow equipment to be
transferred from room to
room. We are starting to
see the uses of centralized
UPS systems, especially
for operating room light-
ing applications. Mini-data
centers also are part of
Figure 2: This diagram illustrates
hospital medical record systems that employ central-
electrical system configuration at
ized UPS systems. a data center for a 911-call center.
Critical loads are connected to an
uninterruptible power supply (UPS)
• Data centers are often required to operate 7×24 and have flexible environmental
without critical load scheduled outages. Critical load requirements where generators
is all electrical loads required to perform the intend- didn’t have to start within 10
seconds. Emergency life safety
ed task. Most hospitals occasionally can be shut loads are provided with individual
down for repairs or maintenance, but this trend is battery backups. The 2N electrical
distribution system provides power
changing because electrical system uptime is in-
to the critical load; increased
creasingly considered an expected level of service in
76
reliability is achieved with the dual
hospitals. -chord design. Image courtesy: Page
• Hospital start-up
sequences for larg-
er generator plants
are usually faster
due to critical and
life-safety time
requirements.
• Hospital breaker
coordination is
required by code,
while the data cen-
ter industry consid-
ers it good prac-
tice. Data centers
frequently have had better and more up-to- Figure 3: In hospitals, power supply devices
serving the same purpose are strategically
date documentation on breaker coordination,
connected to the electrical system to allow
but more hospitals are now keeping track of one system or device to be turned off for
this information, too. maintenance while the other system is in use.
Image courtesy: Page
• Hospitals typically use automatic transfer

switches to separate different branches; data centers tend to use electrically operat-
ed breakers in a “main-tie-tie-main” configuration.
77
Next-generation electrical infrastructures

The 7×24 electrical distribution systems are becoming essential to the day-to-day op-
erations of all types of buildings. The more solutions engineers design, the greater the
demand—and need—grows for electrical systems to power these solutions, especially in
data centers and hospitals. The current generation of electrical infrastructures has creat-
ed a high level of expectations among owners and users, who have an increasingly hard-
er time tolerating gaps in high-quality service for these types of buildings.
Reliability and availability for electrical distribution systems will remain on the front line
for electrical designers, and these requirements are here to stay. These qualities not
only determine the service that a building provides, but also have a great impact on the
safety of those who operate these facilities. Additionally, increasing focus on environ-
mental impact and energy consumption will continue to challenge designers during their
innovation of new ways to provide a high level of reliability and availability. In the future,
next-generation hospital electrical distribution systems will more closely resemble those
of data centers.
Freddy Padilla is a principal and the MEP engineering director at Page. He is a member
of the Consulting-Specifying Engineer editorial advisory board.
78
Power-generation systems in high-
Electrical engineers must consider many factors when designing power-
generation systems. Safety, maintainability, efficiency, code compliance,
and economics play crucial roles in determining the topology of a power-
generation system. Specific requirements for power vary based on building
occupancy type, facility use, and critical function.
T he term “high-performance buildings” has generated a great deal of interest over

roughly the past decade. That interest is primarily focused on conservation mea-
sures, specifically with regard to energy and water, and their impact on the environment.
Standby generating systems have received little attention as components of high-perfor-

mance buildings. This general dearth of attention isn’t particularly unexpected, as gener-
ation systems often support the welfare of human beings under adverse conditions and
are, by their nature, high-performance systems. Their unyielding operational and reliabil-
ity requirements often preclude design decisions that might favor energy conservation
and environmental impacts, and the limited run time of standby generators limits oppor-
tunities for generator characteristics to have a substantial impact on energy conservation
or environmental concerns.
History and definition

The term high-performance buildings entered the legal lexicon in the Energy Policy Act
of 2005, commonly called the EPAct. The concept was expanded in the Energy Indepen-
dence and Security Act of 2007 (EISA), which provides this definition for a high-perfor-
mance building: 79
Power-generation systems in high-performance buildings
“… a building that inte-

grates and optimizes on
a lifecycle basis all major
high-performance attri-
butes, including energy
conservation, environ-
ment, safety, security,
durability, accessibility,
cost-benefit, productivity,
sustainability, functionality,
and operational consider-
ations.”
This can be called a “soft” definition: It describes the Figure 1: Paralleling switchgear for
an 8,000-kW system with four 2,000-
focus in general terms, but it doesn’t provide enough
kW generators is shown. All graphics
information to determine whether a particular build- courtesy: Smith Seckman Reid Inc.
ing can be classified as high-performance.
EISA also provided for the creation of an Office of Federal High-Performance Buildings,
under the General Services Administration, to establish and promulgate more detailed
standards for federal buildings. A number of states have followed with high-performance
building programs of their own. The aggregate market for facilities that can qualify as
high-performance buildings, therefore, is quite large, leading to a great deal of interest
and discussion in the building design and construction industries.
80
Of the 10 characteristics of high-performance buildings listed in EISA, the greatest indus-

try interest is focused on energy conservation, environment, and sustainability.
Standby generation systems

The U.S. Environmental Protection Agency (EPA) rules classify standby generation sys-
tems as either emergency systems or as nonemergency systems. The regulations are
complex, but they are presented in a simplified form: Emergency systems, as defined by
EPA rules, are those that operate only when the electric utility service is either unavail-
able or unacceptable, and otherwise for certain specific purposes for limited periods of
time. Nonemergency systems are those that run under any other conditions. Peak-shav-
ing is an example of an application that would be impermissible for an emergency sys-
tem, but it is allowed for a nonemergency system.
The EPA promulgates different emissions regulations for emergency systems and none-
mergency systems. Because they may run at any time, the rules for nonemergency sys-
tems are very restrictive. Rules for emergency systems are, by comparison, relaxed due
to the limited conditions under which they are permitted to operate. Most generating
systems installed at facilities primarily intended for occupancy by human beings are
classed as emergency systems. This article will focus on systems classified as emergency
systems under EPA regulations.
Environment
The design decision that might be expected to have the greatest environmental impact
is the selection of the fuel source for the generating system. NFPA 110-2016: Standard
for Emergency and Standby Power Systems declares that three fuel sources shall be
permitted for standby power systems: liquid petroleum products, liquefied petroleum 81
gases, and natural gas. In practice, these fuels are Figure 2: A simplified schematic of
paralleling switchgear is shown, with a split
diesel fuel, propane, and methane. Propane units
bus. In operation, the tie-breaker is open;
are available only in limited sizes, typically 150 kW generators come to speed and energize
and below, and have limited application as standby high-priority loads. After the system
stabilizes, the two buses are synchronized
units for all but the smallest building loads.
and connected through the tie breaker.
Natural gas has a reputation as a clean-burning fuel, and in fact, it does have lower emis-
sions of almost every type at the point of use, with the exception of water vapor. In terms
of carbon dioxide, the greenhouse gas that currently gets most of the press, natural gas
generates about 30% less than diesel fuel to produce equal amounts of heat. It would
seem, then, that natural gas would be the preferred fuel for generator applications from
an environmental standpoint.
82
The overall emissions picture, though, is less clear. A small portion of natural gas pro-
duced and transported will escape, appearing as atmospheric methane. Methane is a
very effective greenhouse gas, capturing the Earth’s radiated heat about 25 times as
effectively as carbon dioxide over a 100-year period, as reported by the EPA. So, a small
amount of methane released during production, transportation, and delivery can entirely
negate the reduced greenhouse effect of reduced carbon dioxide emissions.
On the other hand, atmospheric methane persists for a few decades at most, with the
bulk converted to other, more benign substances in the first 10 or so years, while carbon
dioxide appears to persist for centuries or longer.
Natural gas engines are somewhat less efficient than diesel engines, though that gap
appears to be closing. In terms of carbon dioxide emissions, the advantage of natural
gas over diesel is therefore less pronounced when comparing equal amounts of energy
delivered at the generator terminals, as opposed to equal heat content.
Trade-offs between the estimated climate effects of these two gases are difficult to esti-
mate, and it appears that general agreement on the equivalence has not been reached
among climate scientists. It’s not entirely clear which of the two options has a lower
impact on climate change, but the balance currently appears to tip slightly in favor of
natural gas. Decisions regarding fuel source will, therefore, be based on other consider-
ations.
Diesel generators command roughly 80% of the standby generator market, due primar-
ily to operational advantages and industry familiarity. Diesel generators have a better 83
ability to track sudden

large changes in load than
similarly sized natural gas
units, making them better
able to meet the 10-second
starting requirements of
NFPA 110 for Level 1 instal-
lations-generators whose
failure could have a serious
impact on the safety of hu-
man beings.
One of the primary advantages of natural gas as a gener- Figure 3: The fuel-rate range
plotted against the percentage of
ator fuel is the fact that it’s provided by an offsite supplier
rated output, expressed in gallons
and doesn’t require onsite storage. For Level 1 installa- per megawatt-hour, for typical
tions where the probability of interruption of the offsite generators rate from 150 to 2,259
kW. Larger units typically are more
fuel supply is high, however, NFPA 110 requires onsite efficient than smaller units.
storage of sufficient fuel for the entire required run time
of the standby system. This requirement will often negate a significant advantage of
natural gas as a generator fuel. The code doesn’t provide guidance on the level of likeli-
hood of failure that triggers the onsite storage requirement. For Level 1 installations, the
acceptable level of risk could be expected to be quite low, particularly where the risk of
interruption of utility power and natural gas service are correlated.
Energy conservation
Emergency standby generators run infrequently and usually for short periods of time. 84
They are permitted by EPA regulations to run for as much as 100 hours/year for testing
and maintenance while the utility is available, and for an unlimited period when the utili-
ty has failed. In practice, their testing and maintenance run time will be much lower than
the allowed maximum, and periods when utility power is unavailable will be limited.
The electric utility industry takes service reliability quite seriously, and will take measures
to improve it-sometimes under pressure from regulators and customers-should outage
frequencies or durations begin to rise. The limited run time of standby systems makes
the efficiency of the engines less interesting from the standpoint of energy conservation.
Standby generators generally operate in a relatively narrow band of roughly 70 to 75 gal/

MWh in their most efficient range-usually 75% to 80% of nameplate capacity-and exhibit
the familiar bathtub curve over their operating range. Larger units are typically a bit more
efficient than smaller units. This narrow range of efficiencies is due to the fact that diesel
engine technology is driven largely by the transportation industry, where fuel efficiency is
a primary driver of purchasing decisions. Modern designs have wrung out about as much
efficiency as the medium can deliver. In general, attempting to select diesel generators
for operating efficiency will yield only marginal benefit, if any.
Paralleling
Paralleling generators can yield meaningful increases in overall fuel efficiency, particularly
for systems whose total load shows a high degree of variability. Generating systems must
be sized to serve the largest loads that they will be required to serve, and they are often
sized to accommodate expansion that may be delayed, or may never occur. In practice,
though, they will normally see a load considerably below their projected peak demand,
resulting in them operating well below their optimal efficiency. 85
Most modern paralleling systems

are capable of adjusting the
number of generators online in
response to changing loads. This
feature is sometimes called “load
demand.” In a load-demand
system, all available generators
will start in response to a power
outage. After the system stabi-
lizes, the system compares the
load to the online capacity, and if
adequate headroom exists, it will
Figure 4: A comparison of fuel
de-energize generators until the load and capacity are
rates versus output for a parallel
well-matched, maintaining an adequate online reserve combination of four 500-kW
capacity of typically 20%. The benefit of this feature, in generators, with generators staged
on as required, and a single 2,000-
terms of system fuel efficiency, is that the control system
kW generator. The parellel system is
can keep the generators running as close to their maxi- more efficient at low loads, while the
mum efficiency as the system load and generating-unit single larger unit is marginally more
efficient at higher loads.
sizes will allow.
From the viewpoint of fuel efficiency, the benefit of paralleling is reduced due to the
limited run time of emergency standby systems. As a simple example, a 2-MW generator
running at 40% would burn about 7 gal more of diesel fuel per megawatt-hour than two
paralleled 500-kW units at 80%. For 100 hours of run time, the difference amounts to 700
gal-about what a single good-sized diesel pickup truck might burn in a single year. The 86
environmental impact of improved efficiency by paralleling is limited.
Parallel systems provide a number of operational advantages in addition to fuel efficien-

cy. An N+1 system can tolerate the failure of a single generator, improving reliability and
maintainability. A system can be designed to be expandable, allowing the postpone-
ment of expenditures for additional units until they are actually needed. Full-load testing
can be simplified by testing one unit at a time, requiring a load bank the size of a single
unit rather than the entire system.
Those advantages come at a considerable cost, in terms of the cost of the paralleling
system itself and the additional complexity of the system. There are many good reasons
to parallel, but energy efficiency and environmental concerns normally will not drive that
decision.
Productivity
Productivity is influenced by the quality of the indoor environment. A variety of studies
have concluded that environments that don’t intrude on the perceptions of building
occupants lead to higher productivity. Generating systems affect indoor environmental
quality in terms of acoustical and visual comfort: noise and views. If a generator is visible
at all from the occupied space, it will have a negative impact on view quality. Such aes-
thetic concerns are the province of the project architect. The engineer, though, can have
a substantial impact on the system’s noise level.
Generator noise will be an important consideration for facilities that are intended to
maintain a level of normal operation during a power outage. Even in facilities that don’t
continue operating through a power outage, some level of noise management may still 87
be necessary to ensure that emergency instructions and communications among emer-

gency responders can be understood. Many municipalities have noise ordinances that
limit the sound-pressure level at the property line from all sources.
The impact of generator noise on occupant productivity will have a limited impact on
overall economic performance, due again to the limited run time of emergency standby
generators.
For outdoor installations, noise-management strategies are based primarily on barriers

and distance. An outdoor generator will require some form of enclosure. The manufac-
turer’s standard offering will typically provide minimal sound attenuation. Where there’s
adequate distance from the generator to the occupied space, or to the property line, no
further sound reduction may be required. Otherwise, a sound-attenuating enclosure will
be necessary.
Sound-attenuating enclosures are normally rated for a specific generator, with a spe-
cific sound-pressure level at approximately 23 ft from the enclosure. The resulting
sound-pressure levels are usually specified in decibels, a logarithmic measure of sound
energy per unit area, and are usually frequency-weighted.
Generator sound pressure levels are typically described in “dBa.” The nomenclature
dBa means that measurements are in decibels, and that the frequency components of
the sound have been weighted using an industry standard scale, arbitrarily named “A,”
giving the greatest weight to frequencies between 1 and 6.5 kHz. Sound-attenuating en-
closures are typically rated to limit generator noise at 23 ft to 85 dBa, 75 dBa, or 65 dBa,
and will depend on the municipality or jurisdiction. A 65-dBa enclosure is quieter, larger, 88
and more expensive than a 75-dBa enclosure.
Distance from the source provides sound attenuation. As sound radiates from its source,
its power is spread of the surface sphere of increasing radius, and the sound power per
unit area decreases with the square of distance from the source. For an uncomplicated
arrangement, without large reflective surfaces near the generators, the sound-pressure
level will decrease to a quarter of its initial intensity when the distance to the source is
doubled.
That decrease corresponds to an attenuation of approximately 6 dB. Looking at a 75-dBa

enclosure, with a sound-pressure level of 75 dBa at a distance of 23 ft, the sound-pres-
sure level would be decreased to 69 dB at a distance of about 46 ft. Distance provides
effective sound attenuation on multibuilding campuses, where generators can be placed
far from principal occupied spaces.
Generators installed inside the structure they serve, as may be the case in tight urban
sites, present a much more complex set of conditions for noise management. In these
installations, generators are surrounded by close reflective surfaces, complicating the
analysis, and the building structure itself will participate in transmitting sound through
the building. In these cases, the project team is well-advised to engage an acoustical
consultant to analyze the installation and recommend attenuation measures.
Tom Divine is a senior electrical engineer and project manager at Smith Seckman Reid
Inc. He is a member of the Consulting-Specifying Engineer editorial advisory board.
89
How to design
electrical rooms
Electrical engineers should coordinate with mechanical engineers, architects,
structural engineers, and others involved in the design of electrical rooms.
E lectrical rooms and mechanical, electrical, and plumbing (MEP) spaces are often an
afterthought when it comes to building design and planning, either relegated to lo-
cations that are left over or deemed undesirable for other planning purposes. This short-
sightedness can have unfortunate consequences on the cost, operations, and flexibility
of the systems for the future.
NFPA 70: National Electrical Code (NEC) dictates the minimum amount of space needed
around the equipment for access, operations, safety reasons, and conduit installation.
Together, with the actual equipment sizes, this defines the overall minimum dimensional
requirements of the room.
There are three types of general interior electrical spaces that factor into new building
design: main equipment rooms, distribution pathways, and local/branch equipment
rooms. Code-required working space and dedicated space needs must be met. This ar-
ticle will outline important considerations for these spaces in the early stages of building
design as they relate to building type, intended occupancy, size, and future expectations
of both the building and the electrical systems.
Working vs. dedicated spaces

Let’s first define what differentiates working and dedicated space as stated by the NEC
90
How to design electrical rooms
(see Figure 1). The working space helps

safeguard a clear working zone around
all equipment and ensures protection
for any workers or occupants within the
room. This includes defining minimum
width, depth, and height requirements
for the working space, which varies due
to voltage and the specific equipment.
The higher the voltage of the equipment,
the greater the depth of the working
space. The width should be equal to the
width of the equipment and no less than
30 in., while allowing for opening any
doors or hinged panels to a full 90 deg.
The height should be 6 ft 6 in. from the floor, Figure 1: The dedicated space (depicted in red)
and the working clearances (depicted in blue) are
or the height of the equipment if greater than shown in a new emergency distribution room.
6 ft 6 in. Image courtesy: CannonDesign
The style and construction type of the electrical equipment dictates whether only front
access is required, or if rear and/or side access also is required. For each point of access
to a piece of equipment, the minimum working clearances must be provided.
Dedicated space is a zone above the electrical equipment. It’s reserved to provide future
access to the electrical equipment, protection of the electrical equipment from foreign
systems, and for installing conduit/other raceways supporting incoming and outgoing
circuits. The requirement for dedicated space applies primarily to switchgear, switch- 91
boards, panelboards, and mo-

tor control centers. The space
should be equal in width and
depth to the equipment size
and extend from the floor to a
height of 6 ft above the equip-
ment (or to a structural ceiling,
whichever is lower). No equip-
ment or systems foreign to the
electrical installation are al-
lowed in this zone by the NEC.
The area above the dedicated

Figure 2: The medium-voltage switchgear sections
space may contain foreign systems, pro-
and unit substation transformers in a large data center
vided proper protection prevents damage installation require additional space and clearances.
from drips, leaks, or breaks in these sys- Image courtesy: CannonDesign
tems. However, it’s good practice to avoid having these systems installed in electrical
rooms altogether.
While installations of equipment greater than 1,000 V generally follow the same princi-
ples, some of the specifics vary, requiring additional clearance around the equipment
due to the increased hazard that these voltages impose (see Figure 2). Access to this
equipment is preferably limited to only those deemed qualified to be there. For this rea-
son, electrical equipment should be installed in rooms or spaces that are dedicated for
that purpose and have controlled access.
92
The first room type we’ll explore in this article—a main equipment room—has distinctive
needs that separate it from distribution spaces or local/branch equipment rooms.
Main equipment rooms

The main electrical room, or service entrance space, should coordinate with the local
electrical utility (refer to NEC Article 230, Services, for additional details pertaining to the
installation of service entrance conductors and equipment). For example, main equip-
ment rooms have requirements that dictate access to the space from the exterior for
servicing, maintenance, and service feeder installation. The type of equipment installed
will also further determine the room requirements. The service entrance room is typically
located on an exterior wall for both code and practical reasons; it makes installation easi-
er and minimizes the length of the service entrance conductors. Because the service con-
ductors are usually the largest in the facility, this can have a substantial impact on cost.
Using arc-resistant switchgear will also impact space needs. This equipment will be taller
and may have a larger footprint. Engineers will also need to account for the potential
exhaust gases and arc flash energy by providing a pathway to expel them and relieve the
pressure buildup from inside the switchgear.
If an exterior transformer is used to provide the service to a building, feeders from the
transformer enter the building and transition to the main service entrance disconnect,
typically a switchgear, switchboard, or panelboard. These feeders are often routed un-
derground into the building through the exterior foundation wall via a coordinated
opening. Additional coordination with the structural engineer is needed to avoid foot-
ings.
93
The elevation of the service entrance conduits many times do not naturally align with the
equipment to which it is routed. Additional space in the form of increased height or foot-
print commonly is required to allow for the successful transition and termination of these
conduits and conductors. Service installations that require medium-voltage equipment
and/or transformers installed indoors will require additional elements including more
space, higher fire ratings of the rooms (per NEC Article 450), and increased ventilation.
The location of any exterior equipment also needs to be coordinated with other archi-
tectural and landscaping elements. Minimum separation distances are often dictated
by local codes/ordinances or utility requirements for proximity to screen walls, fencing,
vegetation, paths of egress, or building fenestration.
Generator installations offer additional challenges when it comes to defining space

needs. Noise, odor, and vibration factor into the location of this equipment within a
building. The equipment should be located to minimize disturbances to building occu-
pants and adjacent properties. Many jurisdictions have specific requirements on noise
emissions, which will impact equipment placement and other components needed to
meet requirements. Increasing the distance of this equipment from sensitive areas is one
way of dealing with the concerns, but this comes with added feeder costs and may prove
to be more costly than other options.
Sound attenuation and equipment required to meet specific emissions requirements,

such as diesel oxidation catalyst, particulate filters, urea tanks, and selective catalytic
reduction units, have significant cost implications and require a large amount of space to
install. The U.S. Environmental Protection Agency defines the performance standards for
stationary combustion engines in 40 CFR Part 60, Subpart III. 94
Tier 4 versus Tier 2 compliance

is usually dictated by an owner’s
desire to use a generator for utility
peak shaving or other nonemer-
gency uses. It is crucial to have
a clear understanding of current
and future implications in both of
these areas from the outset of a
project and to discuss them thor-
oughly with the building owner.
The weight of a generator and the

Figure 3: At the Coppin State University Science
vibration experienced during its operation will
and Technology Center, conduit pathways need
have an impact on the building’s structural de- to be considered when designing electrical
sign. Generators require a lot of ventilation for rooms to ensure proper clearances are met and
that the distribution if efficient. Image courtesy:
cooling and combustion needs; getting air into
Mathew Kirschner Photograph
and out of the room is critical and will impact
placement.
With regard to fuel storage, most installations require a volume of fuel that dictates
an external fuel tank with interconnecting fuel lines. NFPA limits the overall capacity of
diesel fuel inside buildings to 660 gal. The relationship of the exterior tank and the gen-
erator is also important to minimize pumping requirements and allow for gravity-drain
return-fuel piping. This requires the fuel tank to be lower in elevation than the generator.
Direct access to the outside is preferable for maintenance and testing. All of this requires 95
close coordination with the architectural, structural, and mechanical disciplines.
NFPA 110: Standard for Emergency and Standby Power Systems requires the emergen-
cy power supply (generators) for Level 1 installations to be installed in a separate room,
separated from the rest of the building by 2-hour fire-rated construction. While NFPA 110
does allow the emergency power supply system equipment (EPSS; equipment consists
of all components from the emergency power supply, or EPS, to the load terminals of the
transfer switches) to be installed in the same room as the EPS, it is good practice to keep
these separated to help enhance system resiliency. EPS rooms are also prone to addi-
tional dust, moisture, temperature fluctuations, and excessive noise during operation
that limits the ability to have a conversation and may have a negative impact on other
equipment if co-located.
For mission critical facilities (e.g., financial institutions, data centers, and airports) and
other highly sensitive installations, the use of a dry-type, preaction, or another type of
fire protection system that does not rely on a normally wet piping installation is highly
recommended. In cold climates, this has an added advantage of preventing pipes from
freezing, rupturing, and potentially flooding the EPS room.
Distribution pathways
Distribution pathways are needed for interconnecting all the electrical equipment and
end-user devices, and the pathways will affect where rooms are located. Conduits can
be routed above the equipment, below ground, or in the ceiling space of the floor be-
low, though overhead conduits need space within the rooms to leave the equipment
and transition to the desired route going to other parts of the building (see Figure 3).
The routing of the feeders and how they enter/exit the distribution equipment must be 96
evaluated during design and re-

confirmed during the shop drawing
review, as this will impact how the
equipment is constructed and affect
its physical size.
Below grade conduit routing needs

to be coordinated with other util-
ities and footing/foundation ele-
ments. The restrictions that these
place on the routing may impact the
layout of the equipment in the room
and the size of the space needed.
Similarly, beams on the floor above
or below the equipment may re-
quire an offset of conduit or shifting
of the equipment to allow for the
conduit installation to effectively
occur.
Horizontal pathways can define the

placement of electrical rooms, as other building Figure 4: The hospital’s electrical closet is
arranged to meet multiple requirements. First,
elements may impede these routes and affect all code clearances have been met. Addition-
installation. Structural beams and large ductwork ally, it provides a vertical pathway for feeders
can become obstacles, especially in tandem with extending up through the building. Image
high ceilings. Large-volume spaces like gymna-

courtesy: CannonDesign
97
siums and atriums require extra care as to how conduit will be routed across or around
these areas, especially when the entering/exiting pathway would be lower in elevation
than the ceiling.
Vertical risers are typically accommodated in either one of two ways—through shafts (pull
boxes may be required depending on the height of the building and conduit layout) or
stacked electrical closets. Stacked closets allow for the busway or conduits that distrib-
ute power throughout the building to be run through these spaces for a more efficient
and less expensive installation. If these closets are constructed with 2-hour-rated parti-
tions, the stacked rooms can provide the code-required circuit protection for EPSS feed-
ers and fire alarm circuits without having to rely on more costly wiring methods.
Local/branch equipment rooms

A third space type, the local/branch equipment room, is often referred to as an elec-
trical closet (see Figure 4). Distribution panels, branch circuit panels, and low-voltage
transformers are typically located in these spaces and directly serve the end-user loads:
lighting, receptacles, and small equipment. Lighting control system panels and devices
(and other electrical system devices) are sometimes also located in these rooms. Given
the amount of change that occurs in buildings over their lifespan, extra wall space should
always be provided in these rooms for future equipment.
In multistory buildings, these spaces should be stacked. The placement of electrical

closets within a building’s footprint is often an item of much debate and discussion with
the rest of the design team. The NEC has set restrictions on piping and ductwork rout-
ed through these rooms (i.e., dedicated spaces). Conduit needs to be routed out of the
room to the floor or area served; minimizing branch circuit lengths help avoid excessive 98
voltage drop and reduce distribution costs. These rooms should be located as close to
the center of the area served, with conduits routed out in all directions.
Avoid specific adjacencies to other building elements. Often, closets are targeted for
location next to mechanical shafts, but the need to get ductwork and/or piping out
of these becomes challenging and conflicts with the electrical equipment’s dedicated
space. Similarly, locations next to stairs or elevator shafts present other challenges and
limit the routing of conduits out of the electrical rooms. Locating electrical rooms next
to these, especially if placed between, should be carefully evaluated to ensure there is
enough space and flexibility for conduits.
Additional space needs

Outside of working- and dedicated-space needs, there are many special considerations
for electrical rooms that depend on building programs as well as exterior spaces that
will directly impact how the electrical systems are designed. The needs and expectations
associated with an office building are very different from that of a data center or hospital
with regard to the electrical distribution systems. Redundancy and resiliency are essen-
tial for mission critical-type facilities. Flooding due to natural disasters is a key element
in determining equipment placement (NFPA 99: Health Care Facilities Code, Chapter
6, and NFPA 110, Chapter 7). Historically, much of the main equipment was located in
basements or (partially) belowgrade levels, but now this equipment is located above the
anticipated flood levels. This ensures ongoing continued operations during and after an
event.
Mission critical and safety-critical (e.g., hospitals) installations require added redundancy
to ensure the continuity of business operations and avoid potential loss of life or serious 99
personal injury. Redundancy of systems requires more space, as the equipment is sepa-
rated into different rooms in different parts of the building. Having panels that are part
of a redundant distribution arrangement (A and B sources) located adjacent or in close
proximity to each other in the same electrical room greatly minimizes the value that the
intended redundancy offers. The redundant equipment should be located in separately
rated spaces, with the A sources and distribution located apart from the B sources and
distribution.
Additional clearance requirements include allowing for future equipment to be moved

into a room or allowing for the eventual replacement of that same equipment. While
code may only require 3 or 4 ft of clearance in front of a piece of equipment, the physical
dimension of the equipment could be larger. Because of this, the only way to effectively
remove and reinstall a replacement is to leave an area that is larger than the footprint of
the equipment.
Getting equipment from the exterior of a building to its final location may not always be
a concern during the initial building construction, but it will certainly be an issue during
later time periods of equipment modifications, additions, or replacement. The entire
pathway from the building exterior, including doorways, may need to be enlarged due
to the height or width of the equipment. If the equipment is located on a floor level that
is below- or abovegrade, then area wells, reinforced floors, and a pathway or removable
sections of the exterior wall assembly may be required.
Buildings are expected to have a life well beyond the initial install, and yet future growth
and conduit installation are rarely considered. This automatically infers change, which will
likely come in the form of added equipment and conduit. Initial planning and system de- 100
sign should account for this by including spare breakers, additional distribution sections,
and oversized-conduit rack supports.
Supporting the whole building

Appropriate lighting and illumination levels are important for occupant safety. Ventila-
tion and cooling needs must be defined and support the heat load generated. And it’s
important to remember the physical protection of equipment located outside and in the
open. This may entail the installation of crash-tested bollards or barriers to protect the
equipment from accidental vehicle damage, for instance.
The NEC also requires that access to some equipment be limited to qualified persons
only in some instances. Per the NEC, this includes only those who have the skills and
knowledge related to the construction and operation of the electrical equipment, the in-
stallation, and have received safety training to recognize and avoid the hazards involved.
There is no singular method to design electrical systems, nor are two buildings ever the
same. Electrical space needs differ by project and building. Electrical engineers and
designers must consider many factors when making the decisions that lead to the final
design. Most importantly, this includes communication and working hand-in-hand with
mechanical engineers, architects, structural engineers, and others that have a hand in the
design of buildings, and coordinating the electrical installation with these other disci-
plines.
Scott Kesler is a principal and engineering integration leader at CannonDesign. He is an

electrical engineer with more than 25 years of experience.
101
Understanding the changes
to NFPA 70-2017
NFPA 70-2017: The National Electrical Code (NEC) has many requirements that
apply to designing electrical systems for health care facilities. It’s vital to understand
the changes in the 2017 edition in order to stay code-compliant.
T he health care construction industry is one of the most regulated industries in the
United States. There are more than a dozen national, local, state, and government
codes and standards that are applicable to the design, installation, maintenance, and
performance of health care facilities. The most popular codes include NFPA 70 National
Electrical Code (NEC), Article 517 Health Care Facilities; NFPA 99: Health Care Facilities
Code; NFPA 101: Life Safety Code; NFPA 110: Standard for Emergency and Standby
Power Systems; Facility Guidelines Institute (FGI), ASHRAE 170: American Society of
Heating Refrigerating and Air-Conditioning Engineers Standard on Ventilation of Health
Care Facilities; Center for Medicare and Medicaid Services (CMS); United States Pharma-
copeia (USP), Chapter 797 Pharmaceutical Compounding—Sterile Separation and Chap-
ter 800 Hazardous Drugs—Handling in Healthcare Settings; International Building Code
(IBC); and The Joint Commission accreditation standards, just to name a few.
NFPA 70 Article 517 applies to the construction and installation of electrical systems in
health care facilities that provide services to human life. NFPA 99 covers the performance
of electrical systems in health care facilities. Health care facilities are covered in Article
517 in NFPA 70. Electrical systems are covered in Chapter 6 of NFPA 99.
New definitions in NFPA 70

There are new definitions in Article 517 of NFPA 70 that have design implications that
102
Understanding the changes to NFPA 70-2017
need to be considered
by the consulting-speci-
fying engineer.
The governing body of

the health care facility is
“the person/s who has
the overall legal respon-
sibility for the operation
of a health care facility.”
The governing body
establishes and identifies
the patient care space
categories in the health care facility based on the Figure 1: A women’s and children’s medical
center with multiple building occupancies.
anticipated level of patient care and designates
Image courtesy: AECOM
wet-procedure locations. The above areas of
responsibility have many requirements that must be considered by the consulting en-
gineer, including the number and type of receptacles and number of branch circuits in
patient spaces, special protection against electric shock in wet-procedure locations, and
loading on the alternate power source.
An invasive procedure is defined as “any procedure that penetrates the protective sur-
faces of a patient’s body and that is performed with an aseptic field (procedural site).”
Not included in this category are special procedures such as the placement of peripheral
intravenous needles or catheters used to administer fluids or medications, insertion of
urethral catheters, and other similar procedures. What is important for the consulting en- 103
gineer to understand about invasive procedures are what make certain spaces in a health
care facility a Category 1 risk space. Category 1 spaces have special requirements in wir-
ing, protection, and the type of the essential electrical system (EES) that supplies these
spaces. NFPA 99 (Item A.6.7.5.1.3) recommends an automatic battery-operated lighting
unit or system to provide minimal task illumination in invasive-procedure spaces.
A medical office is “a building or part thereof where the following occur: 1) examinations
and minor treatments or procedures performed under the continuous supervision of a
medical or dental professional; 2) only sedation or local anesthesia is involved and treat-
ment or procedures do not render patients incapable of self-preservation under emergen-
cy conditions; and 3) overnight stays for patients or 24-hour operation are not provided.”
What is important for the consulting engineer to understand is that there is a difference
between a health care occupancy and a health care facility. All health care occupancies
are health care facilities, but the opposite is not true. A health care occupancy is used to
provide medical or other treatment or care simultaneously to four or more patients on an
inpatient basis, where such patients are mostly incapable of self-preservation due to age,
physical or mental disability, or because of security measures not under the occupant’s
control. A hospital, for example, is a health care facility but can be of several different
occupancies (health care, business, assembly, ambulatory, etc.). A medical office is a health
care facility, but under NFPA 101 it would typically be a business occupancy.
Patient care spaces as defined in NFPA 70 are now aligned with the same definition in
NFPA 99. A patient care space is “any space of a health care facility wherein patients are
intended to be examined or treated.” There are four categories in health care facilities
identified by the governing body based on 1) the anticipated level of patient care in any
given space; and 2) the risk to the patient from procedures that will be performed. 104
Table 1 pro-
vides a distinc-
tion summary
of patient care
spaces, gives
examples of
these spaces,
and discusses
their design
considerations
as stipulated in
NFPA 70 and
NFPA 99. Figure 2: A hybrid operating room in a hospital—a Cate-
gory 1 (critical care) space. Image courtesy: AECOM
NFPA 70-2017 “Type” classification

The 2017 version of NFPA 70 has now aligned the Type classification of the EES with NFPA
99. Type 1 and Type 2 EES are referenced in NFPA 70 Article 517.29 (Essential Electrical
System Requirements for Hospitals and Other Health Care Facilities) and in Article 517.40
(Essential Electrical Systems for Nursing Homes and Limited Care Facilities), respectively.
It is to be noted that the “type” terminology is also mentioned in NFPA 110 and applies
to the emergency power supply system. The following paragraphs clarify what the type
means as referenced in NFPA 70, NFPA 99, and NFPA 110.
NFPA 99, Item 6.7.1.2.4.1, requires that Type 1 and Type 2 EES power sources be classified
as Type 10, Level 1 generator Class X, per NFPA 110. The type in NFPA 110 is the maxi-
mum time in seconds that the emergency power supply system (EPSS) will permit the load 105
terminals of the trans-

fer switches to be
without power. The
class in NFPA 110 is
the maximum time in
hours for which the
EPSS is designed to
operate at its rated
load without being
refueled. This means
that power must be
restored within 10
seconds, but the
Figure 3: A medical/surgical inpatient room in a
amount of time the EPSS is required to run
hospital—a Category 2 (general care) space. Image
without refueling is not defined in NFPA 110 courtesy: AECOM
and needs to be specified as required by oth-
er codes or by the authority having jurisdiction.
Type 1 EES per NFPA 99 and NFPA 70 consists of three branches: life safety, critical, and
equipment. Type 2 EES has two branches only: life safety and equipment. Type 2 EES does
not have a critical branch. The critical branch in a Type 1 EES is intended to serve loads that,
if they were without power, their failure could result in death or serious injuries. See Figure 5.
Type 1 EES applies to critical care (Category 1) and general-care (Category 2) spaces in
hospitals and other health care facilities where patients are sustained by electrical life sup-
port equipment. Type 2 EES applies to nursing homes and limited-care facilities. For free- 106
standing buildings where admitting and discharging policies preclude providing care for
patients who may need electrical life support equipment, or where no surgical treatment
requiring general anesthesia is offered, an automatic battery-operated system would suf-
fice to provide 1½ hours of backup lighting for exit lights, exit corridors, stairways, nursing
stations, medical preparation rooms, boiler rooms, and communication areas, in addition
to backup power to operate all alarm systems. 107
Item 517.45 clar-

ified the type of
EES required for
other health care
facilities (other
than hospitals,
nursing homes,
and limited-care
facilities). If elec-
trical life support
is required or if
there are critical
care (Category
1) spaces, then
a Type 1 EES shall be provided. If there are Figure 4: A sports medicine clinic—a Category 3 (basic
care) space. Image courtesy: AECOM
general-care (Category 2) spaces, then a
Type 2 EES is required. However, if required by the governing body of the facility, an EES
is permitted and is comprised of an alternate power system capable of supplying a limit-
ed amount of lighting and power for the orderly cessation of procedures during the time
normal power is interrupted.
It is to be noted that Type 3 EES has been eliminated from the 2015 version of NFPA 99
because Category 3 and 4 spaces are not required to be fed by an EES. This means that
no spaces in the health care facility were required to be served from a Type 3 EES. Facili-
ties that had previously required a Type 3 EES typically provided the type of patient care in
which, if normal power was interrupted, there was a minimal risk to patients or staff. Such 108
a facility could safely terminate procedures and Figure 5: This diagram shows the difference
between a Type 1 and Type 2 essential electrical
basically close for business. A Type 3 EES was
system. Image courtesy: AECOM
only required to provide 1½ hours of backup
power, which is consistent with NFPA 101. It is not intended that these facilities no longer
need backup power, but rather the other codes and standards or the governing body will 109
specify limited requirements needed for safe egress from such facilities.
NFPA 70 and NFPA 99 prohibit the use of isolated ground receptacles inside of a patient
care vicinity. A patient care vicinity is a space within a location intended for the examina-
tion and treatment of patients, extending 6 ft beyond the normal location of the patient
bed/table and extending vertically 7.5 ft above the floor. Isolated ground refers to devices
that have a ground connection, but that ground connection is derived from the building
entrance rather than being directly grounded to the enclosure. Examples of where an iso-
lated ground application might be required include diagnostic and treatment equipment
that may require a dedicated or isolated ground separate from the facility’s grounding
system to reduce noise when serving sensitive electronic or medical equipment. The con-
sideration for the consulting-specifying engineer is that isolated ground is not permitted
within a patient care vicinity, but it can be provided outside the patient care vicinity area.
There are other changes in the 2017 NFPA 70 outside of Article 517 that impact all types of build-
ing occupancies, but the following should especially be considered for health care facilities.
Additional arc flash energy-reduction measures have been introduced in Item 240.87,
namely using circuit breakers with instantaneous trip settings that are less than the avail-
able arcing current or with an instantaneous override that is less than available arcing cur-
rent. Although on one hand, instantaneous tripping would reduce arc flash energy—as arc
flash energy is a function of protective device fault-clearing time and tripping current—,
instantaneous tripping will not allow proper circuit breaker selective coordination. The
concept of selective coordination is to delay, not to instantaneously trip upstream circuit
breakers to allow downstream breakers localized to the fault to trip first. Consulting-spec-
ifying engineers should conduct an overcurrent protective study and balance the trade-off 110
between lower arc flash energy and circuit breaker Figure 6: A portable generator/load-bank
connection. Image courtesy: AECOM
selective coordination.
There are also notable changes in NFPA 70-2017 made to Article 700. Article 700.3.F now
calls for an emergency system that relies on a single alternate source of power that will be 111
disabled for maintenance or repair, to include a permanent switching means to connect a

portable or temporary source of power, and be available for the duration of the downtime.
Manual switching is allowed to be used in this case to switch from the permanent source
to the portable source. Some health care facilities have multiple alternate power sources,
such as generators connected in an N or N+1 configuration. In a two-generator scheme
connected in parallel in N configuration and supplying 100% of the total essential load,
either generator would comply with the above requirement in this case, if it can support
the life safety system load by itself. The portable or temporary power-source connection
could also be used to connect a portable load bank to comply with NFPA 110 periodic
maintenance testing and loading requirements, which are mandatory for Level 1 and Level
2 standby and emergency power systems. See Figure 6.
Another important requirement has been added in NFPA 70 Article 700.10(D)(3) is fire
protection for emergency (life safety) feeder circuit wiring for health care occupancies
where patients are not capable of self-preservation, such as hospitals and ambulatory care
occupancies. The code lists a number of options to comply with this requirement including
installing life safety feeders in spaces that are protected by an approved fire-suppression
system, using feeder cables that are a listed electrical circuit protective system, or encased
in a minimum of 2 in. of concrete. As required by other codes, a water-based sprinkler
system is typically provided in health care occupancies; however, the code stipulates that
running feeders above a ceiling in an otherwise fully sprinklered building does not meet
the requirement unless there are also sprinklers in the plenum space above the ceiling.
Without proper consideration to the routing of feeders, this new requirement could have
a very large financial impact on the installation. The consulting engineer might consider
installing emergency feeders in areas that are typically fully sprinklered, such as vertically
within electrical rooms or horizontally in areas without a suspended ceiling, and avoiding 112
horizontal feeder runs above suspended ceilings that are not sprinklered, to avoid the ad-
ditional cost of the other fire protection measures.
Another addition to Article 700 of NFPA 70 which is directly related to Article 517, is a new
requirement for generator-control wiring. The article states that generator-control wiring
installed between the transfer equipment and the emergency generators must be kept en-
tirely independent of all other wiring and must be protected against fire in line with the fire
protection measures listed above. Article 700.10(D) (3) now calls for continuous monitoring
of the integrity of the generator-control wiring. Without this monitoring, an open circuit on
the control wiring would go undetected and won’t be noticed until there is a power outage
and the generator fails to start. This new requirement entails three elements: 1) continuous
monitoring of generator signal circuits; 2) visual and audible annunciation; and 3) automat-
ic transmission of an engine start signal and an alarm when a problem is detected. Several
transfer switch manufacturers offer an easy solution for existing and new installations. Exist-
ing wiring does not need to be changed in this case, but additional circuit-communication
modules would have to be added at both ends of the control wiring: at the generator and
at the transfer equipment.
Article 517 of NFPA 70 has many design considerations for the consulting engineer design-
ing health care facilities. These design considerations cover the wiring and protection of
the electrical distribution system in patient care spaces and the EES in various health care
occupancies.
Firas (Ross) Alkhayri is the electrical department manager at AECOM’s Columbus, Ohio,
office. He has more than 25 years of experience in the design of power and emergency
systems for a wide variety of mission critical facilities. 113
Selecting, sizing transformers
in nonresidential buildings
Transformers are a critical piece of any electrical infrastructure. Knowing how to
properly size and select these transformers will allow the building to function with
minimal interruptions.
T ransformers are a ubiquitous component of any nonresidential distribution system.

With many different sizes, types, and listed uses, it can be quite complicated to pick
the correct transformer for a specific application.
Transformers are a passive component of electrical distribution systems and are used
throughout electrical systems to change the voltage, they will either step-down a voltage
or step-up a voltage with almost zero power loss; the average transformer is roughly 98%
to 99.5% efficient. Typically, step-down transformers are seen in commercial buildings to
bring the voltage from a higher utility-distribution voltage (7 to 25 kV) to a lower utiliza-
tion voltage (120/208 V, 277/480 V, etc.). This is done for safety and efficiency; it is much
more efficient for transmission to take place at high voltages with smaller conductor sizes
and it is much safer for a lower utilization voltage at the end-user equipment.
There are myriad transformer types; however, there is only a handful used commonly.
NFPA 70: National Electrical Code (NEC) 2017 outlines the installation requirements for
these transformers and provides minimal guidance on sizing the transformers.
Transformer ratings
Transformers are rated using industry-accepted terms, such as capacity (kilovolt-amperes
rating or kVA rating), voltage, temperature rise, and insulation class.
114
Selecting, sizing transformers in nonresidential buildings
Capacity: Transformers are rated to

carry a load, which is indicated by their
kilovolt-amperes nameplate ratings at
the rated output voltage and frequency.
These are standard within the industry
and include: 15, 30, 45, 75, 112.5, 150,
225, 300, 500, 750, and 1,000 kVA. There
are several transformers larger than this
on the utility level; however, they are not
common, with the exception of large
projects with campus-style distribution
schemes.
Voltage: A transformer will have a

primary and secondary voltage rating
along with the wiring configuration. In
U.S. facilities, the overwhelming ma-
jority of transformer installations are a
step-down 480-V delta primary to 120/208-V wye Figure 1: This is a sample nameplate from a
transformer manufactured in 1997. This 300-
secondary. Other voltage ratings and configura-
kVA transformer is K-rated with a K factor of
tion types exist for special applications. 13. Image courtesy: NV5
Unless there are strong technical reasons otherwise, most transformer installations
should have a wye output. Whether the neutral is required for the downstream distri-
bution system, this configuration provides a point to intentionally create the reference 115
to ground; this is important as transformers are separately derived systems in which the
ground reference is lost on the primary side and must be re-established on the second-
ary. A corner-grounded delta is possible, and often encountered in older installations,
but has several disadvantages. The grounded phase must be labeled throughout the
system, a higher line-to-ground voltage exists on two phases resulting in a reduced
fault-current rating on circuit breakers, and circuit breakers must be marked 1- to
3-phase.
Energy efficiency: The U.S. Department of Energy (DOE) mandates transformer efficien-
cies. The most recent regulation, commonly referred to as DOE-2016, was adopted start-
ing Jan. 1, 2016, and is a requirement for all transformers produced in or imported into
the U.S. Manufacturers are permitted to sell all remaining inventories; most transformers
at this time are compliant with DOE-2016.
Sound level: All transformers will give off both vibrations and audible noise caused by
magnetic expansion and contraction of the cores. These vibrations cannot be entirely
eliminated but may be mitigated with measures such as vibration-isolation pads and
acoustically dense room construction.
Insulation class (also known as temperature class): This rating describes the maximum
temperature, in Celsius, at which the windings may operate without damage to the insu-
lation. Several common insulation classes are 105, 150, 180, and 220.
It should be noted that NEC 450.21(B) requires a fire-resistant room (1-hour fire-rated
construction) for transformers that are larger than 112.5 kVA with an insulation class be-
low 155°C. The typical transformer specified today has an insulation class of 220°C and 116
falls within exception No. 2.
Temperature rise: This is the average change in temperature at the windings from a
no-load to full-load situation. Typically, this is expressed in degrees Celsius. This rating is
usually standard based on the insulation class.
Standard dry-type temperature rise is 80°C, 115°C, or 150°C.
Standard oil-filled temperature rise is 60°C.
Hot spot allowance: The windings of a transformer core are not evenly heated during
operation, instead the interior of the windings is hotter than the surrounding areas be-
cause they are farther from any ventilated openings. This is a set number as defined by
industry standards and is associated with the insulation class. For example, a Class 105°C
transformer will be allotted a 10°C hot spot allowance.
Ambient temperature: This rating specifies the average temperature of the space that
the transformer will occupy over a 24-hour period. Typically, this rating is 40°C.
It should be noted that typically the maximum operating temperature for a trans-
former is:
Maximum temperature (°C) = ambient temperature + temperature rise + hot spot

allowance
Nameplate: Each transformer is required to have a permanently affixed label indicating 117
the requirements outlined in NEC 450.11(A). These are: manufacturer, rated kilovolt-am-
peres, frequency, primary and secondary voltages, impedance (if greater than 25 kVA),
required clearances, quantity and type of insulating liquid (if any), and temperature class
(see Figure 1).
Common transformer types

Liquid-filled transformer: These units are filled with a liquid that acts as the cooling fluid
and dielectric medium between the transformer cores. The most common types of liquids
in use are mineral oil and less-flammable bio-based oils. Mineral oil is commonly used for
exterior pad-mounted utility transformers and is considered combustible, with a flashpoint
of less than 300°C. Typically the flashpoint of mineral oil is around 155°C. Bio-based oils
do not contain petroleum and are made from vegetable oils. These bio-based oils have a
much lower flashpoint—around 330°C—and are much more environmentally friendly; in
the event of a leak, they will biodegrade within a month under normal conditions.
Indoor use is restricted to vault rooms and the installation must comply with NEC 450.26.
These vaults are characterized by having exterior access, 3-hour fire-rated construction,
liquid containment, and exterior ventilation per NEC 450.42. FR3 is a less flammable liquid
with a flashpoint of around 316°C and is preferred for liquid-filled transformers located
within the building. Installation requirements for transformers insulated with less flamma-
ble liquids are addressed under NEC 450.23. Indoor installations that comply with NEC
450.23(A)(1) do not require a vault or sprinkler protection.
Dry-type transformer: These units are air-cooled and are not liquid-filled. As these units
rely on air to cool the core and windings through convection, they are typically larger than
their liquid-filled counterparts. Within the dry-type family there are two specific subtypes: 118
ventilated and nonventilated. Ven-

tilated dry-type transformers have
openings within their enclosures
and allow air movement from the
exterior of the enclosure to the coils
within the enclosure. Nonventilated
(or sealed) dry-type transformers are
completely sealed and allow cool-
ing through the surface area of the
enclosure. These units are well-suit-
ed for wash-down areas as well as
corrosive, combustible, or otherwise
harmful conditions.
It should be noted that NEC-2017

currently has a typographical error
in section 450.23(A) where it ap-
pears to limit indoor installations to
vault rooms. However, this has been
corrected in errata number 70-17-
6, issued July 5, 2018. This corrects
NEC 450.23(A)(1)(e), which requires less Figure 2: Example of a standard dry-type 112.5-kVA trans-
former during the installation process. This transformer
flammable liquid-insulated transformers
has been supplied with vibration-isolation pads at the feet
installed indoors to be supplied with and a flex connection for the conduit to reduce vibrations
automatic fire suppression and liquid transferred to the structure. Each of the three cores is also
containment. NEC 450.23(A)(1)(f) requires

visible in this photo. Image courtesy: NV5
119
that the less flammable oil-insulated transformer is installed in a vault to be subsets of NEC
450.23(A) instead of being supplemental requirements.
This correction now allows three options for indoor installation in lieu of the single way that
the code appears to outline. For transformers less than 35,000 V, all three installations are
acceptable: installation with liquid confinement and no combustibles stored in the room,
installation with automatic fire suppression, or installation in a vault room. For transformers
larger than 35,000 V, there is only one code-compliant installation, which is constructing a
vault room.
Liquid-filled transformers operate with a standard temperature rise of 60°C above ambient.
These transformers can sustain bursts of overload up 50% above a nameplate rating for
short periods of time. The amount of time and overload capacity is directly related to the
previous continuous loading and the winding temperature.
Resin cast coil: A cast coil or epoxy cast coil transformer is a dry-type transformer con-
structed with the primary and secondary windings fully encapsulated in a resin, which pro-
tects the transformer from moisture, corrosion, or other aggressive contaminants. These
transformers have a standard ventilated dry-type enclosure.
The properties of this insulation provide a higher short-circuit strength and can sustain
repeated short-duration overloads.
Harmonic mitigating transformer (HMT): These transformers are specially designed to

cancel out problematic harmonic currents (often these are the 3rd, 9th, 15th, or triplen har-
monics) resulting from nonlinear loads. These transformers accomplish this through phase 120
shifting and zero-sequence flux cancellation within the cores, preventing the harmonics
from being reflected back to the primary winding of the transformer
K-rated transformer: These transformers are commonly used in applications involving

harmonic currents, but they will not cancel out the harmonics in the same process as an
HMT. Instead, this transformer has a derated core and will withstand the heating effects
of harmonic currents. Additionally, this type of transformer will not prevent the harmonics
from propagating up through the distribution system through the transformer primary.
The industry commonly refers to K-rated and HMT transformers as alternatives to each
other because they are commonly used to address the same issue. However, the HMT
will remedy the issue while the K-rated transformer is merely a solution to the transformer
damage and does not address the harmonic currents for the entire distribution system.
Load types and sizing

When selecting a transformer, load calculations are completed as outlined within the NEC.
Sizing of panelboards and branch circuits is typically done through calculations outlined in
NEC Article 210, Branch Circuits, and NEC Article 230, Services. This calculation or summa-
tion of calculations shall be used in sizing the transformer. For example, a building that has
a total NEC demand load of 60 kVA on the 120/208-V, 3 phase, 4-wire system.
To size the transformer, it is recommended to allow 25% future growth to the circuit and
then adopt the next standard transformer size up. In this example, 60 kVA of demand load
shall be multiplied by a factor of 1.25, resulting in a load of 89 kVA. Obviously, 89 kVA is not
a standard transformer size and it is recommended to use the next standard size, which is
112.5 kVA. 121
Once the size of the transformer has been decided, the type of transformer shall be select-
ed. The next logical step is to determine the load types that are in use on the downstream
system. If the system has a high number of resistive or linear loads, the selection is quite
easy and sends us to a standard dry-type or oil-filled transformer (depending on location,
size, etc.). However, if there are several nonlinear loads, such as computer/servers with
switch-mode power supplies, gaming slot machines, LED lighting, motors, or variable fre-
quency drives (VFDs), an HMT should be considered.
Systems with harmonic loads commonly are designed with K-rated transformers in lieu of
an HMT. It is important to note that an HMT will correct the issue while a K-rated trans-
former will only withstand the heating of the transformer core. The installation of a K-rated
transformer is a solution if the system as a whole can withstand the harmonic currents with-
out failures and the only point of concern is the premature failure of the transformer due to
overheating the windings.
Locations
A specific location can make or break transformer selection. Depending on the space
in which the transformer is intended to be installed, certain options may be completely
off the table. For example, if you are in an indoor corrosive environment, the most effec-
tive options are either a nonventilated or cast coil transformer. Conversely, it may be too
costly to use a liquid-filled transformer in a standard indoor application due to potential
vault-construction requirements and oil containment.
Consideration also should be given to the location of transformers in relation to surround-

ing occupancies and replacement. Transformers produce an audible low humming and 122
vibration that can

propagate through
the structure and sur-
rounding spaces. It is
common to furnish a
transformer with vibra-
tion-isolation pads and
flex-metal conduit for
the final connection;
these precautions pre-
vent the vibrations from
entering the building
structure. However, the
hum of the transformer
will still be audible in
surrounding spaces. Figure 3: An example of an exterior pad-mounted liquid-insu-
lated 1,000-kVA transformer is shown. Utilities typically require
outdoor installations as part of their standards for maintenance
Vault room: When designing with and replacement. Additionally, a curb and/or bollards typically
a liquid-insulated transformer for in- are required for physical protection. Image courtesy: NV5
stallation within the building footprint, a vault room is often required by NEC 450.26. Less
flammable liquid-insulated transformers also can fall under the constraints of a vault room,
but this is slightly less common—and when selecting a less flammable liquid-insulated
transformer, the intent is often to avoid construction of a vault room.
The room requirements are outlined in NEC 450, Part III, beginning with 450.41. There are
several requirements outlined in Part III, such as ventilating the vault with outside air via 123
ductwork or a flue, 3-hour fire-resistant construction, a 4-in. concrete floor, and oil con-
tainment for the largest transformer in the room. A typical 3-hour fire-resistant room will
be constructed with 6-in.-thick reinforced concrete, which can add considerable cost and
complexity to a project.
There is an exception to allow for 1-hour fire-resistant construction, with the use of sprin-
klers, carbon dioxide, or halon systems. These rooms typically are designed for utility
transformers and services; in these cases, both the NEC and the utility’s requirements shall
be followed.
Indoors: Dry-type transformers are the most commonly used transformers for indoor
installation, as there are no restrictions on their location. As previously mentioned, if the
transformer is larger than 112.5 kVA with an insulation class below 155°C, a 1-hour fire-re-
sistant room shall be required.
However, this is atypical of most transformers specified today. Additionally, less flammable
oil-filled transformers may be installed indoors with minimal requirements, such as either
an automatic sprinkler system or liquid containment with no combustibles stored inside
the room for transformers less than 35 kV. It is not uncommon to see smaller dry-type trans-
formers installed on a wall-mount assembly, hung from the slab in an open-ceiling room,
even hung from the slab with an open ceiling or above a drop ceiling. Per NEC 450.13(B), a
transformer installed in a hollow building space shall not exceed 50 kVA.
Overcurrent protection: Transformers must be protected like any other component of a

building’s electrical infrastructure. An entire article could be written discussing protection
strategies, sizing, and requirements. These requirements are outlined in NEC 240.21 and 124
450.3. It is common practice to use NEC Table 450.3(B) and provide both the secondary
and primary protection based on this table.
In conclusion, there are many details that come into play when selecting and sizing a trans-
former. Once a specific end use and location have been determined, the selection process
can begin while considering and aggregating the topics outlined in this article. It is im-
portant to note that some end users may have their own requirements in addition to those
listed here.
Stephen Berta is a project consultant at NV5. He has experience in electrical design of

hospitality, gaming, K-5 education, and data centers.
Robert R. Jones Jr. is the associate director of electrical engineering for NV5’s Las Vegas
office. He has experience in multiple market sectors including hospitality, commercial,
medical, and government projects.
125
Designing backup, standby,
and emergency power for high-
Backup power design for a high-performance building is required, and offers
many benefits.
E lectrical engineers must consider many factors when designing backup, standby, and
emergency power systems. Safety, maintainability, code compliance, and economics play
crucial roles in determining the topology of an emergency system for a critical facility. Spe-
cific requirements for emergency power vary based on building-occupancy type, facility use,
and critical function (see Table 1). This article focuses on using backup power as an alterna-
tive source of the main utility in high-performance buildings, with Graton Resort and Casino
in Rohnert Park, Calif., as an example. The property uses three parallel diesel generators as
its backup power source.
What is a high-performance building?

A high-performance building is a building that integrates and optimizes, on a lifecycle basis,
all major high-performance building attributes, which include energy-conservation, environ-
ment, safety, security, durability, accessibility, cost-benefit, productivity, sustainability, func-
tionality, and operational considerations (i.e., schools, libraries, factories, etc.). With energy
codes becoming more stringent concerning energy efficiency and increasing federal incen-
tives for buildings that are eco-friendly, owners/clients would rather go this route when de-
signing their building, which can be cost-effective when the building is complete. One must
take note that the design of a backup power system for a high-performance building isn’t
much different than for a standard building with similar qualities. 126
Designing backup, standby, and emergency power for high-performance buildings
The need for backup power

In the event of power failure from the utility, buildings rely on backup power both for the
safety and health of the public as well as the protection of important business assets that
cannot be lost during a power outage. Unexpected disasters often disrupt the power to hun-
dreds of thousands of people and businesses. Many buildings, such as hospitals, airports,
data centers, water and sewage facilities, fueling stations, and communication and trans-
portation systems, require some sort of alternative power to ultimately save lives during the
event of a crisis. Power loss to a business also can create a significant economic impact. The
longer a business is without power, the greater the economic loss. When these unexpected
situations occur, backup power provides a source to support the equipment loads via unin-
terruptible power supplies, generators, or battery-storage systems.
Requirements
Having the knowledge in backup power design for emergency, legally required standby,
and business critical loads is an important skill for electrical engineers and designers to have
and can sometimes be complex. When designing backup systems, electrical engineers and
designers must assure that the backup power and the building’s electrical systems can han-
dle the critical loads accurately and adequately. They also have to take into account that the
systems need to be code-compliant with their authority having jurisdiction. The International
Building Code (IBC)-2015 dictates the exact requirements for these specific systems, such
as emergency and legally required standby power. NFPA 70-2017: National Electrical Code
(NEC), Article 700.12, and the California Electrical Code (CEC) require emergency power
systems to automatically start within 10 seconds and run at full capacity when there is pow-
er failure. The three key articles from the NEC that must be taken into consideration when
designing backup power systems are Article 700, Emergency Systems; Article 701, Legally
127
Required Standby
Systems; and Arti-
cle 702. Optional
Standby Systems,
which clarify what
one should con-
sider emergency
and standby.
Article 700 in
the NEC defines
emergency pow-
er as “Legally
required and
classed as emer-
gency by munici-
pal, state, federal,
or other codes,
or by any govern-
mental agency
having jurisdic-
tion. It is intend-
Specific requirements for emergency power vary based on build-
ed to automatically supply ing-occupany type, facility use, and critical function.
illumination, power, or both to
designated areas and equipment in the event of failure of the normal supply or in the event
of an accident to elements of a system intended to supply, distribute, and control power and 128
illumination essential for safety to human life.” Examples of emergency loads are:
• Exit signs.
• Egress lighting.
• Elevator lighting and controls.
• Automatic fire alarm systems.
• Sprinkler alarm systems.
• Fire pumps.
• Power and lighting in fire command centers.
• Elevator-car lighting and controls.
Elevator machine room lighting as defined by Article 701, Legally Required Standby Power,
is “Legally required and so classed as legally required standby by municipal, state, federal,
or other codes or by any governmental agency having jurisdiction. These systems are intend-
ed to automatically supply power to selected loads (other than those classed as emergency
systems) in the event of failure of the normal source.” Legally required loads include:
• Stair pressurized fans.

• Elevators machine equipment.
• Emergency/legally required power room-cooling systems and controls.
• Sliding doors for rated openings.
• Elevator machine room cooling systems and controls.
• Smoke control equipment.
• Air compressors for dry-pipe sprinkler systems.
• Clean agent fire suppression systems.
• Miscellaneous equipment. 129
Article 702, Optional

Standby Power, is intend-
ed to supply power to
public or private facilities
or property where life
safety does not depend
on the performance of
the system. These sys-
tems are intended to
supply onsite-generated
power to selected loads
either automatically or
manually. This section
also is considered busi-
ness-critical loads. These loads are typically Figure 1: There is provision for a third generator
on the right so that the client can easily add more
owner-selected equipment.
generators in the future. Image courtesy: NV5
When separating your business critical/optional standby loads from emergency and legally
required standby power, it is always best to identify the owner/client’s emergency require-
ments. There may be times where the business needs require high reliability, thus one must
be able to adapt the backup power system to the client’s needs and still provide the most
appropriate degree of life safety for the occupants of a building in accordance with the code.
Benefits of parallel generation systems

Even though parallel systems can at times seem complex, they have greater advantages with
respect to reliability and redundancy. The backup power source for Graton Resort and Ca- 130
sino uses multiple generator sources paralleled together to provide flexibility in load man-
agement, increased reliability, and the capability of maintenance with little to no disruption.
The generators are paralleled to a common bus of the system in a separate raceway from the
normal power, which serves the emergency and business critical loads, especially, for system
response time and dynamic load response once the generators are in operation.
Redundancy. The redundancy of the multiple generators in Graton’s backup system provides
greater reliability than a single generator unit for critical loads. In the event of a failure of one
generator, the backup loads will be redistributed among the other generators in the system
on a priority basis. The emergency loads that require the highest degree of reliable backup
power quite often account for only a fraction of the overall power generated by the system.
In a parallel system, this means that the most emergency elements will have the redundan-
cy necessary to maintain power even if one of the units goes out. If an N+1 configuration is
adopted, one generator can be offline for maintenance while still serving the required loads.
Furthermore, providing a running spare N+1 generator will increase the reliability of the gen-
erator system from 98% to 99.96% reliability.
One of the primary purposes of redundancy is to eliminate single points of failure. Caution
must be exercised to ensure these single points of failure are not simply moved to another
part of the system. The controls enabling redundancy also must be analyzed to avoid failure
modes that compromise reliability. For example, paralleled generator sets that rely on a single
master control for signals to start and close to a paralleled bus actually replace one failure
point with two, as the master control and the communication link between the master and the
generator sets each represents single points of failure. A well-engineered paralleling system
will have dual hot backup control systems, redundant communication pathways, redundant
battery-select dc power supplies, selectively coordinated breakers and multiple power path- 131
ways, and a well-document-

ed procedure for system
recovery whenever a com-
ponent fails.
With an N+1 parallel gen-

erator system, if at one
point one of the generators
fails or requires mainte-
nance, the individual unit
can be taken off the bus
without affecting the other
generators that are con-
nected to the bus. This
Figure 2: At Graton Casino and Resort, outdoor
provides additional protection that ensures an
paralleling diesel generators within a generator
uninterrupted supply of power for the building’s yard are housed in weatherproof enclosures and
critical circuits. To avoid load sharing among have subbase tanks, step-ep transformers, mufflers,
and fuel containment. Image courtesy: NV5
the generators, it is always best to match the
new paralleled-system generators with the same manufacturer, ratings, and type. It is even
more important to do the same when modifying an existing system. Standardization of the
generator model type enhances maintenance and operations of the generator system.
Efficiency. This is a key factor of a high-performance building. The more efficient the system
is, the more stability you have, which reduces cost and losses. Having the functionality of
three generators in lieu of one larger generator to handle the load makes the system more
versatile and economical. Because loads do not remain at a constant level, the inconsistent 132
power demand can

cause a single larger
generator to run at
loads below 30% of
its maximum capacity,
which could cause wet
stacking and lead to low
fuel efficiency and more
frequent maintenance.
The optimal operational
point for generators is
between 75% and 80%
of its rating. Ensuring
that the generators are
at this optimal point will reduce mainte- Figure 3: The photo shows the generator system con-
trol cabinets of Graton Casino and Resort’s emergency/
nance and fuel costs.
optional standyby system with human-machine interface
(HMI) screen, indicating lights, local breaker controls,
Expandability. When sizing generators to and metering. Image courtesy: NV5
match system load requirements, it often can be difficult to properly project the increases in
load and adequately plan for unanticipated additional requirements. If the load projections
are aggressive, the initial investment for a generator may be higher than necessary. If the
load projections are inadequate, the standby power system may be compromised or ex-
pensive post-installation upgrades may be needed. Graton’s parallel system offers a level of
scalability and modularity that allows for variations in load over time and optimal operation
of the installed units. If the physical space of the generators is planned properly, additional
generators can be added in the future if needed (see Figure 1). 133
System testing. A very important

requirement for emergency/standby
systems, the system-testing proce-
dure ensures that a high-performance
building is safe, secure, durable, and
productive. NFPA 110-2016: Standard
for Emergency and Standby Power
Systems includes Emergency Gener-
ator Testing Requirements for Emer-
gency Power Supply Systems (EPSS),
which sets safety standards to protect
building occupants by making sure
generator-powered backup lighting
will operate as expected. Monthly and
yearly tests are performed on genera-
tor systems whose failure could result
in death or injury. Should a generator
system fail the EPSS monthly test, it
must be repaired immediately and re-
Figure 4: The generator system control single line
tested until the emergency system passes. diagram for Graton Casino and Resort shows three par-
allel generators, a gen bus, a load bus, a utility incomer
breaker, individual gen breakers, and load bus feeder
No-load testing. During no-load testing, breakers. Image courtesy: NV5
the generators are started, synchronized,
and connected to the generator bus without taking on load. This test will not have any im-
pact on the load fed from the utility and can be scheduled to occur at regular intervals (en- 134
gine exercise setting). No-load testing determines whether the engine can start when given
a signal to do so, but will not indicate whether the generator will actually carry the load. An
on-load test can determine if the generator system can carry the critical load reliably.
Onload testing. During onload testing, the load is gradually transferred from the utility to
the generators via closed transition, runs for a period of time, and then is transferred back to
the utility. Running a generator system under actual facility load is the only true way to test its
capability to operate as specified. There are many engine and generator components that
must see real operating conditions to provide confirmation of their ready state. NFPA 110 re-
quires that diesel generators be exercised monthly for 30 minutes at operating temperature
and every 3 years to at least 30% of nameplate load for a period of 4 hours.
Future needs
One last step in the sizing equation has to do with future needs. The general rule is to always
size the system 20% to 25% more than its maximum demand. Power use is not always fixed
and tends to grow over time. Therefore, any generator-set-sizing exercise needs to take
system expansion into consideration. Even with sophisticated software, the final decision on
generator set size needs to be tempered with judgment.
Michael Villalba is a designer with FEA Consulting Engineers. He previously worked for NV5.
He has 3 years of experience working on projects, such as convention centers, casinos, ho-
tels, and photovoltaic farms.
135
Specifying paralleled
generation systems
Learn about important factors related to specifying onsite generation systems,
focusing on paralleled generators.
P rogress in developing countries, the proliferation of distributed generation, and an

ever-increasing focus on power system reliability continue to stimulate demand for
onsite power generation. Grid modernization and technological advances in the indus-
try are enabling these systems to be applied in ways not previously common.
From interruptible service agreements to microgrids, the applications of engine-gener-

ators have diversified significantly in recent years. As such, sound judgment and funda-
mentals on the part of the consulting and specifying engineer are as important as ever.
Paralleled generator systems

Benefits of paralleling generators in onsite power systems are well documented. First,
multiple generators in a system can increase reliability. Even if a system does not have
fully redundant generation, load-prioritization schemes keep critical loads powered
if a generator set fails. Next, generators can operate closer to their optimal load with
demand schemes that take paralleled units offline when they’re not needed. Further,
systems with multiple generators are more scalable, facilitating system expansion and
capital project flexibility. Also, smaller units can have faster delivery times. Finally, par-
alleled units offer greater operational flexibility, facilitating routine maintenance.
Creating thoughtful design criteria helps convert the client’s goals into high-level tech-
136
Specifying paralleled generation systems
nical requirements. These criteria establish design philosophy consensus early in the
project, which can dramatically reduce rework down the road. This document is used
throughout the project and defines the rest of the design process.
First, the design criteria outline the client’s commercial goals. For example, will they
use the generation for interruptible service agreements where the utility provides
financial incentives in exchange for generation support? Will they be operating in a
baseload or peak-shave mode to help minimize demand charges? Or will the system
be focused strictly on emergency or standby power?
Answering these questions is critical for the genset supplier and the design of the elec-
trical distribution system. Also, these considerations define code and environmental re-
quirements to which the project will be exposed. For example, NFPA 110: Standard for
Emergency and Standby Power Systems has specific requirements for things like start
time for emergency power systems. Also, the U.S. Environmental Protection Agency
can have more stringent pollution-control requirements for generators that operate
while normal power is available.
Next, the engineer should establish the critical loads and the reliability required. In
some cases, not all loads are critical. Leveraging this fact can increase the overall re-
liability of the system, especially with paralleled generators. For example, if a single
generator fails, noncritical loads can be shed or not added in the first place. This capa-
bility can essentially establish N+1 redundancy for critical loads, even though the over-
all system isn’t fully redundant. Conversely, if every load connected to the generation
is essential, it’s crucial to know this early so the appropriate design precautions can be
taken. 137
System reliability is a whole

science by itself. Tools such as
the IEEE 3006 Power System
Reliability Standards can be
very handy to study and quan-
tify the reliability of different
design concepts.
In addition to reliability, an-

ticipated future expansion
should be considered. This
can establish spatial con-
straints that may otherwise
not exist. For example, does
the design need to leave
room for an additional gener-
ator or does room need to be
left for additional sections of
switchgear? Does the electri-
cal system need spare capacity for additional Figure 1: This one-line diagram is configuring
in an isolated bus configuration. Image courtesy:
generation? Incorporating these things into
Stanley Consultants
the design from inception will result in a much
cleaner final build-out.
Finally, applicable codes and standards should be compiled. Nothing is worse than 138
finding out that a local code requires something beyond NFPA 70: National Electrical
Code (NEC) requirements after construction has begun. Performing a review of appli-
cable codes and standards at the beginning of the project can help the engineer re-
fresh their memory and highlight pitfalls before design work begins in earnest. These
can act as valuable guides during design. Also, the client’s insurance requirements
should be reviewed to see if, for example, FM Global requirements are more stringent
than the NEC.
Generation capacity
After the design philosophy has been established, the required capacity for the gen-
eration system can be determined. This exercise requires a review of the facility’s load
profile. If the generators will serve an existing system, actual metering data is very
valuable here.
If this data is not available, however, load lists can be compiled to quantify the load.
The designer must consider the expected harmonic content of the system, which
may require generator derating. Also, starting requirements for large motors must be
evaluated to make sure the alternators and exciters can provide the reactive power
required. Note a general rule states that a generator should be sized 2.5 times larger
than the largest motor it is required to start. Generator manufacturers have online soft-
ware tools that can aid this effort.
Depending on the type of system, there are NEC requirements that must be consid-
ered as well. For example, generation systems are exposed to different requirements
when they’re supplying emergency, legally required standby or optional standby loads.
Specifically, when supplying emergency loads, NEC Article 700.4 states, “An emergen- 139
cy system shall have adequate capacity and rating for all loads to be operated simulta-
neously.” Alternatively, Article 702.4 allows for optional standby-capacity calculations
to be made in accordance with Article 220. When demand factors are applied, this can
offer some capacity relief.
While the criteria mentioned above will help the designer ensure enough capacity is
provided, it is worth noting that having too much capacity can be a problem as well.
This can expose the generator to problems such as wet stacking or fuel stagnation.
System topology, overall one-line diagram

Once the required power is set, the overall one-line diagram is developed. This
establishes the overall system topology identifying major pieces of electrical equip-
ment and how they’re connected. Here, the engineer chooses the voltage levels that
will be used. The 480-V generator is extremely common, and this voltage level has
many benefits including the fact that a wide range of contractors and operators are
familiar with it.
Also, generation at the same voltage as most of the loads can result in fewer electrical
distribution components. When low-voltage systems result in bus ampacities of 4,000
to 5,000 amp and higher, installing cabling becomes difficult and incident energies
become very high. In these cases, it may be time to consider medium-voltage systems,
which have lower ampacities for the same amount of power. If multiple generators exist
onsite, it is best to parallel them as opposed to keeping them isolated. This provides
the redundancy benefits described above. Commonly, systems that call for paralleling
two or more generators (G in Figures 1 and 2) do so with electrically operated circuit
breakers (Figures 1 and 2). There are many different ways to connect this switchgear 140
with the utility (U in Figures 1 and 2) and the loads. The best configuration depends
upon the answers to a few critical questions.
Do the generators need to parallel with a utility? This is required when a facility oper-
ates in peak-shave mode, for example. These systems require that the generators and
the utility simultaneously provide power to the loads, so there must be a way to con-
nect the two together. This can be done with a utility breaker connected to the gener-
ator bus. If this is not required, a simpler isolated-bus configuration may be acceptable
(see Figure 1). NEC Article 705 also has requirements for these applications.
Should loads be switched with automatic transfer switches (ATS) or electrically oper-
ated switchgear breakers? ATS are easily procured and designed for a large number
of switching operations. Plus, they may be required in emergency systems where code
requires segregation of emergency loads. That said, performing utility/generator trans-
fer in the paralleling switchgear can maximize factory wiring and minimize the electrical
equipment footprint, making it a cost-effective way of implementing a full-facility back-
up system.
Is a generator bus required? Generator buses connect all the generators together
(Figures 1 and 2). This bus is then connected to load buses via tie breakers. Separate
generator buses have the advantage of allowing generator-paralleling testing without
disturbing the facility’s load. Plus, this seems to be the configuration with which gen-
erator suppliers are most accustomed. This allows the selection of standard designs
that minimize the custom engineering required. However, the generator bus and corre-
sponding required tie breaker(s) will increase cost and space requirements.
141
Are redundant utility sources required? If redundant utility sources are required, the
switchgear will increase in complexity. These sources need to connect to the load bus
in addition to the generators. This can be done with a main-tie-main configuration,
with or without a generator bus. Alternatively, if the two utility sources don’t need to
be able to feed the entire facility load, two separate load buses can be used, each
powered from a utility/generator bus-transfer pair (Figure 2). In this configuration, each
load bus would feed half of the facility load, saving a few tie breakers and some equip-
ment ampacity.
Where do you put the generator breaker? Some manufacturers will put the switching
device (contactor or breaker) on the generator skid itself. In this configuration, gener-
ator sets can essentially be daisy-chained together to connect to the load bus. In this
configuration, the conductors implementing the daisy chain act as the generator bus.
This can be economical for smaller units, but it can result in heavy cabling for larger
systems, which can be cumbersome.
How many generators should be used? Lastly, because it has already been decided to
use parallel generators, the engineer needs to determine how many to use. Should
two units rated 100% each be used, or possibly three units rated 50% each? This is a
question of both reliability and cost optimization. Generator set suppliers can provide
budgetary information on their equipment that’s very valuable here.
Electrical characteristics
After the major pieces of electrical equipment have been chosen, their details can be
determined. Load-flow studies help define the continuous-current ratings required. It is
important to model all of the system-switching configurations to determine the worst- 142
case scenario. For example, if a main-tie-main switchgear lineup is to operate at full

load from one utility source, then a single end must be able to handle the full facility
load. Also, future expansion should be considered.
Next, a short-circuit study is performed. This study establishes the short-circuit ratings
of the equipment, such as interrupting current and bus bracing. Again, when dealing
with multiple generators and multiple utility sources, be sure to consider each possible
switching configuration. Note that the line-ground fault current available from gener-
ators can be greater than their 3-phase fault current, so a complete short-circuit study
must consider different fault types. This results from the fact that generators’ zero-se-
quence impedance is lower than their positive-sequence impedance. This is especially
true in 2/3 pitch generators, which have very low zero-sequence impedance.
When paralleling different sources, available fault-current magnitudes can be quite

high. When these sources are paralleled for long periods of time, the connected
equipment should be rated to handle the total amount. Disagreement exists, however,
when sources will only be paralleled for short amounts of time.
For example, consider systems fed by two redundant utility sources and paralleled
generators. One school of thought here is that the equipment does not need to be
designed to take into account the total fault current because both sources are rarely
paralleled. The competing thought, however, is that when these sources are paralleled,
switching is taking place, increasing the likelihood of a fault. The project team should
carefully consider their approach and gain consensus among stakeholders.
System grounding is very important for paralleled generators, especially when they 143
can parallel with

the utility. Genera-
tors’ fault capability
must be considered
when choosing the
system-grounding
type. For example,
because of the low
generator zero-se-
quence impedance
cited above, medi-
um-voltage gener-
ators are often not
braced to handle
their own line-
ground fault current
when grounded sol-
idly. Therefore, some
type of impedance
grounding is re-
Figure 2: This one-line diagram shows a system con-
quired. Common types are low-resistance,
nected in a transfer-pair configuration with two utility
high-resistance, and reactance grounding. sources. Image courtesy: Stanley Consultants
Low-resistance grounding limits the fault current to hundreds of amps, high-resistance

grounding limits the fault current to tens of amps, and reactor grounding typically lim-
its the fault current to around 60% of a 3-phase fault current. Higher impedances have 144
the advantage of less line-ground fault current and help reduce circulating current;
however, caution should be exercised when generators are paralleled with other sol-
idly grounded sources like wye-windings of utility transformers. Too high a grounding
impedance on a generator connected to a solidly grounded source will cause the line-
ground fault current in that source to rise. This may exceed the equipment’s capability.
For this reason, reactor grounding is often selected when generators are paralleled
with solidly grounded components.
Finally, when deciding on system-grounding types, the varying levels of fault currents
resulting from different switching configurations should be considered. This can cause
problems with protective relaying when the available fault current varies over too wide
a range.
Protection, control, and metering

At this stage, additional details defining how the system is metered, controlled, and
protected can be defined. Two primary approaches are employed here. Some engi-
neers will performance-specify the major aspects of their system and let the generator
supplier work out the details. This is efficient from an engineering effort standpoint
and works well for typical, off-the-shelf systems.
On projects with more customization or integration with existing systems, more detail
may be required from the specification engineer. When this is the case, a protection,
control, and metering one-line diagram may be produced in lieu of a prescriptive spec-
ification. After all, a picture is worth a thousand words.
This detailed one-line diagram shows the major protection and control equipment 145
required. Some primary examples are the protective relaying, the master control cab-
inet, or a manual synchronization panel (if required). It also shows what signals are
exchanged among these devices.
For example, the master control cabinet may need to know the breaker statuses from
the paralleling switchgear or voltages from different points in the system. This requires
that breaker auxiliary contacts and potential transformer circuits be wired to the cabi-
net. Other examples include contacts from lockout relays that trip in the event of a bus
fault. The master controls need these signals so they don’t close the generators on a
faulted bus.
Finally, these types of drawings are especially important when the generator set ven-
dor is not the same as the paralleling switchgear vendor. Detailed drawings can help
coordinate the two.
Note that when paralleling with the utility, a number of additional protective-relaying
elements are required. For example, some utilities require that a distributed-genera-
tion system be provided with anti-islanding that shuts down generation when a seg-
ment of the utility circuit has opened. This helps prevent the generation from powering
a local area of the utility’s system, which can be hazardous to linemen.
Additionally, reverse real and reactive power protection are often required to ensure
the utility isn’t motoring the generator and that the onsite generation isn’t exporting
more power to the utility than agreed upon. Directional overcurrent protection can
also be a valuable tool when generator systems are paralleled with the utility. These
schemes help facilitate coordination in these systems with multidirectional power flow. 146
To take it one step further, three-line diagrams may also be required, which further de-
fine the protection system. At least a partial three-line diagram can be useful in detail-
ing ground-fault protection. For example, this helps answer questions like:
• Should the neutral be switched when the load is transferred?

• Will all of the generators’ neutrals be bonded at a single point in the switchgear?
Power systems with paralleled generators can provide many benefits to facility own-
ers, both from an economic and reliability standpoint. To realize these benefits, it is
important that the consulting and specifying engineers use the tools at their disposal
and apply sound engineering principals to the application. This is especially true with
the increasing diversity of projects and the complexity that can come from paralleling
generators.
Joseph Thornam is a principal electrical engineer at Stanley Consultants.
Stanley Worcester is the chief electrical engineer at Stanley Consultants.
147
Your questions answered: Emergency,
standby, and backup generators
Presenters from the critical power webcast on Dec. 11, 2018, answer
questions left unaddressed during the live presentation.
W hen designing generator systems, electrical engineers must ensure that genera-
tors and the building electrical systems that they support are appropriate for the
specific application. Whether providing standby power for health care facilities or power
for other facility types, engineers must make decisions regarding generator sizing, load
types, whether generators should be paralleled, fuel storage, switching scenarios, and
many other criteria.
Presenters from the webcast on critical power: emergency, standby, and backup genera-
tors respond to questions not addressed during the live event.
Presenters:
Danna Jensen, PE, LEED AP BD+C, senior vice president, WSP USA, Dallas
Kenneth Kutsmeda, PE, LEED AP, engineering design principal, Jacobs, Philadelphia
Question: What are some paralleling challenges and solutions for

large facilities?
Jensen: Large facilities have larger load demands. And when dealing with level 1
systems where a certain amount of the loads must be back up and running in a little as
10 seconds, it can be challenging to overcome this requirement with larger engines. It
takes longer to parallel two larger engines as opposed to smaller ones, but the smaller 148
Your questions answered: Emergency, standby, and backup generators
ones my not support your

load. One solution to this is
to have multiple paralleling
switchgear with a tie break-
er in between. This allows
for faster response times
and larger loads.
Question: NFPA 110:

Standard for Emer-
gency and Standby
Power Systems re-
quires a load bank
testing of generator systems. How do you avoid having the generator
system overload if the normal power fails during the test?
Kutsmeda: Use what is called a “load bank dump signal.” Add an under-voltage
relay on the incoming feeder that senses loss of utility power and sends a signal to dump
the load bank load.
Question: Are there any special considerations to keep in mind if

one would like to be able to have load-shedding capabilities? Do
you need a certain type of generator control panel to have this ca-
pability?
149
Jensen: Load shedding can be done in a number of different ways. They can be ac-
complished via the paralleling switchgear controls directly to the automatic transfer
switch (ATS), or by distribution breakers, or through another set of relays themselves. The
most important aspect of load shedding is to fully test and commission the system to
ensure it is programmed and operates as intended.
Question: What are some recommendations for generators feeding

uninterruptible power supply (UPS) systems? Discuss UPS compatibil-
ity with generators and de-rating.
Kutsmeda: Older UPS system had very high harmonics that required you to oversize
the generator to compensate. Newer UPS system have filters to minimize harmonics so
that is no longer required. You’ll want to make sure the UPS has very low input current
distortion (5% or lower). With that you can put about 75% UPS power onto the generator
without having to oversize.
Another thing to avoid is large UPS systems with very low load. At low loads, the UPS put
capacitance back onto the system, which drives the power factor to leading. On the UPS
end, make sure the UPS ramping function is turned so that it soft load the generator. You
also can reduce the battery charging current while the system is on generator to reduce
load.
Question: How does the CAT scan momentary load play out sizing of
the generator?
150
Jensen: Any imaging equipment has a high in-rush load and are very sensitive to
voltage drop (typical manufacturers say their equipment must be fed from “clean” power
with only +/-2% voltage drop. This could lead to oversizing of generators to be able to
maintain such a strict tolerance. Another option is to include a UPS with power condi-
tioning to ride through any of the start-up and voltage dips on a generator system. Use
a manufacturer’s sizing software and plug in the appropriate parameters in order to get
the best recommended system size.
Question: How does the reliability of a Tier 4 genset fare with critical
facilities? Or is Tier 2 preferred in critical systems?
Kutsmeda: EPA Tier 2 and Tier 4 are determined by the emission and runtime re-
quirements. If a generator is only used to provide power with the loss of utility power
and it operates less than 100 hours a year for maintenance/testing, then it can be Tier 2.
If the generator provides power for peak shaving, storm avoidance or maintenance and
testing exceed 100 hours per year, then the generator must be a Tier 4.
There are, however, two types of Tier 4 generators. Tier 4 certified means the SRC are
installed and certified in the factory with the engine. Tier 4 certified requires shutdown
of the generator with loss of urea or issue with the SCR. Tier 4 compliant means the SCR
system is installed in the field. Tier 4 compliant does not require the generator to be
shutdown with loss of urea. Because of this Tier 4 compliant is more widely used for criti-
cal applications if emissions reduction is required.
151
Question: How do you determine how much spare capacity we have

on a generator?
Jensen: Once a system is up in operation, an easy way to determine the remaining

capacity is to meter the normal side of the transfer switches for an extended period. The
codes require at last 30 days of metered load when using that to determine spare capac-
ity. A good option is to include a power metering system on all transfer switches so this
data is always available for the plant engineer to pull up and have a good handle on the
overall capacity of the system.
Question: Referring to NFPA 70: National Electrical Code (NEC) Arti-

cle 700: Emergency and 2 hours onsite fuel storage: How do you de-
sign for this if using natural gas? What about dual-fuel with propane?
Kutsmeda: If the natural gas source is reliable, some authorities having jurisdiction
(AHJs) will allow natural gas because it could potentially be an indefinite source of fuel
(more than 2 hours). Other options are using propane tanks or a dual fuel generator,
which starts using diesel and then transfers to natural gas. The dual fuel would also have
the 2 hours of fuel stored onsite.
Question: What ancillary or generator accessories require power, and

what branch of the emergency power supply system (EPSS) should
they be powered from?
152
Kutsmeda: Block heaters, jacket water heaters, battery charger, fuel pumps, and lou-
vers. For critical facilities (NEC 708), those loads mush be connected to the critical power
system. For emergency systems, they should be connected to the emergency distribution.
Question: Expand on segregation of emergency loads—is a separate

room needed for transfers switches only?
Jensen: NPFA 110 requires the emergency power supply (EPS) to be in a separate
dedicated 2-hour room for Level 1 systems. Other EPSS equipment is also permitted in
this room (NFPA 110-7.2.1.2). NFPA 110 also requires the EPSS to be in a separate room
from the normal power service entrance equipment when the equipment is rated over
150 V to ground and equal to or greater than 1,000 amps. So for systems of this size/volt-
age and greater, the transfer switches must be in a separate room from the normal pow-
er service equipment, but they can share the same room as other EPSS equipment, such
as the paralleling switchgear and emergency power distribution boards.
Question: Is there a code requirement for health care occupancies

served by single engine systems to be provided with means for con-
necting a portable/roll up generator in event of failure?
Jensen: Now there is. New for NFPA 70-2017, Article 700.3(F) states that a temporary
source of power for maintenance or repair of the alternate source of power is required
and per NFPA 70 Article 517.26, the life-safety branch must meet the requirements of
Article 700. A health care occupancy must also have the portable connection.
153
Question: NFPA 110 prohibits interconnection of sources. Does this

include banning closed-transition transfer?
Jensen: No. NFPA 100-6.2.3 says that mechanical interlocking or an approved alter-
nate method shall prevent the inadvertent interconnection of the primary power supply
and the EPS, or any two separate sources of power. This requirement does not preclude
the use of closed transition switches, but rather requires that if these type of switches are
used, the proper protection such as reverse power relays are used.
Question: What type of software should be used for the design and
calculation of a power distribution system?
Jensen: All of the major manufacturers have their own sizing program. It is best to use
the manufacturer that is actually being chosen for the site as they have the best insight
as to how their engine/alternator will react to the different type of load parameters.
Question: What are some good products to comply with NEC sec-
tion 700.3(F)?
Jensen: All of the major engine and switchgear manufacturers now make an integrat-
ed cabinet that works for quick and easy connection of a temporary roll up generator, or
they can be custom-built by the electrical contractor. The most important consideration
is to plan for the appropriate space where it is accessible by a roll-up generator. And
plan for the proper protection within the switchboard it connects to in order to avoid any
revers power back on to the utility system or other power source. 154
Question: Have you any knowledge of retrofitting 2600 kW generator

to Tier IV? What does it entail?
Kutsmeda: I have added SCR and particulate filters to an old two-stroke diesel en-
gine to reduce emissions. It involved removing and reconfiguring the exhaust and muf-
fler sections to be able to install the SCR and filters. SCR will have a urea system that
can either be a tote system or permanent tank. The urea is temperature sensitive so may
need to be heat traced if in an exposed generator room.
In this case, there was not enough room between the exhaust on the engine and where
the exhaust left the building. Therefore, the exhaust had to be turner so that is went
south and then turned 180 deg north and went through the SCR and out the building.
Back pressure was a big issue in this case.
Question: When it comes to generator sizing tips, loading the sys-

tem with the largest motors first, does this “first” mean before all
other motors or all other smaller loads? For example, in a hospital
project largest motors are typically part of the equipment branch and
won’t be loaded until after life sand critical loads are connected.
Jensen: This is correct. In a hospital application, it may not be applicable to load the
largest motor loads first. This would be more of an option in an industrial plant where
you can load the largest motors first so that the rest of your load does not see the large
voltage dip when the motors are added to the system.
155
Electrical & Power
Thank you for visiting the Electrical & Power eBook!
If you have any questions or feedback about the contents

in this eBook, please contact CFE Media at
customerservice@cfemedia.com
We would love to hear from you!
Sponsored by

CFE Schneider Electrical&amp; Power

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

CFE Schneider Electrical&amp; Power

Uploaded by

Copyright:

Available Formats

Electrical

The role of NFPA 110 and its

When and how to design parallel

Data center design

Codes do not have standardized

required system type as

TLC Engineering &

preclude the installa-

Consideration should also be given to the needs of first responders in an emergency.

Coordination requirements  The role of NFPA

Understanding the above coordination requirements and AHJ interpretation is critical

proportional impact on the

Specific initial acceptance-testing requirements for parallel systems (Chapter 7.13.4.1.3)

The role of NFPA 110

 When and how

Systems capable of paralleling generators inherently provide system-growth capability

The role of NFPA 110

TLC Engineering &

 When and how

The role of NFPA 110

TLC Engineering &

 When and how

add to an existing system. Additionally, when connecting to existing distribution sys-

generator packages that can be

should be clearly defined and made 20

TLC Engineering &

Data center design

platform other codes

TLC Engineering &

When and how

N emours Children’s Hospital (NCH) in Orlando, Florida is a renowned, state-of-the-art

have come from more than 70 different countries.

tricity for equipment to func-

TLC Engineering for Architecture is a national company with expertise

Delivering the solution

To meet this goal, the team developed a 3-tiered

The role of NFPA 110

TLC Engineering &

When and how

Justin Powell of Graybar agreed: “There’s a personal touch, a personal relationship.” 30

The role of NFPA 110

TLC Engineering &

When and how

The role of NFPA 110

TLC Engineering &

When and how

tem integrators, connecting over 1 billion devices.

The role of NFPA 110

Data center design

The role of NFPA 110

TLC Engineering &

When and how

equipment rating. In the event of a component failure or maintenance in one system,

TLC Engineering &

A commonality between the

TLC Engineering &

This configuration has a number of impacts to the distribution:

• Backup power generation: This topology follows the

• Deployment: Each 3M2 system can be designed to accommodate a discrete IT 40

block. Expansion within a deployed 3M2 system is exceptionally challenging and

balanced against smaller

This configuration has several impacts to consider:

TLC Engineering &

This configuration has a number of impacts on the distribution:

CFE Schneider Electrical& Power

CFE Schneider Electrical& Power