IEEE SA
STANDARDS
ASSOCIATION
IEEE Recommended Practice for
the Design of DC Power Systems
for Stationary Applications
IEEE Power and Energy Society
Developed by the
Energy Storage and Stationary Battery Committee
IEEE Std 946™-2020
(Revision of
IEEE Std 946-2004)
+ IEEE
IEEE Std 946™-2020
(Revision of IEEE Std 946-2004)
IEEE Recommended Practice for
the Design of DC Power Systems
for Stationary Applications
Developed by the
Energy Storage and Stationary Battery Committee
of the
IEEE Power and Energy Society
Approved 30 January 2020
IEEE-SA Standards Board
Abstract: Recommended practices for the design of de power systems for stationary applications
are provided in this document. The components of the de power system addressed by this
document include lead-acid and nickel-cadmium storage batteries, static battery chargers, and
distribution equipment. Guidance in selecting the quantity and types of equipment, the equipment
ratings, interconnections, instrumentation and protection is also provided. This recommendation is
applicable for power generation , substation, and telecommunication applications.
Keywords : auxiliary, backup, battery, battery charger, charger sizing, control , cross-tie, de, direct
current, distribution, duty cycle, generating station, ground detection, IEEE 946™ , instrumentation,
nuclear, panels, protection coordination, rectifiers, reserve, selective protection, short-circuit,
substation, telecommunication
The Institute of Electrical and Electronics Engineers, Inc.
3 Park Avenue, New York, NY 1001 6-5997, USA
Copyright© 2020 by The Institute of Electrical and Electronics Engineers. Inc.
All rights reserved . Published 23 September 2020. Printed in the United S tates of America .
IEEE is a registered trademark in the U.S. Patent & Trademark Office, owned by The Institute of Electrical and Electronics Engineers,
Incorporated.
National Electrical Safety Code and NESC are registered trademarks and service marks in the U.S. Patent & Trademark Office, owned by
The Institute of Electrical and Electronics Engineers, Incorporated .
National Electrical Code, NEC, NFPA 70, and NFPA 70E are registered trademarks in the U.S. Patent & Trademark Office, owned by the
National Fire Protection Association.
PDF:
ISBN 978-1-5044-6454-3
ST024059
Print:
ISBN 978-1-5044-6455-0
STOP0 24059
IEEE prohibits discrimination, harassment, and bullying.
For more information, visit https://www.ieee.org/aboutlcorporatelgovemance/p9-26.html.
No part of this publication may be reproduced in any form, in an electronic retrieval system or otherwise, without the prior written permission
of the publisher.
Important Notices and Disclaimers Concerning IEEE Standards Documents
IEEE Standards documents are made available for use subject to important notices and legal disclaimers.
These notices and disclaimers, or a reference to this page (https://standards.ieee.org/ipr/disclai1ners.html),
appear in all standards and 1nay be found under the heading "Important Notices and Disclaimers Concerning
IEEE Standards Documents."
Notice and Disclaimer of Liability Concerning the Use of IEEE Standards
Documents
IEEE Standards documents are developed \Vithin the IEEE Societies and the Standards Coordinating
Com1nittees of the IEEE Standards Association (IEEE SA) Standards Board. IEEE develops its standards
through an accredited consensus development process, \Vhich brings together volunteers representing varied
viewpoints and interests to achieve the final product. IEEE Standards are documents developed by volunteers
with scientific, academic, and industry-based expertise in technical working groups. Volunteers are not
necessarily members of IEEE or IEEE SA, and pa1ticipate without co1npensation from IEEE. While IEEE
ad1ninisters the process and establishes 1u les to promote fairness in the consensus development process, IEEE
does not independently evaluate, test, or verify the accuracy of any of the infonnation or the soundness of any
judgments contained in its standards.
IEEE does not warrant or represent the accuracy or completeness of the material contained in its standards,
and expressly disclaims all warranties (express, implied and statutory) not included in this or any other
doctunent relating to the standard, including, but not limited to, the warranties of: merchantability; fitness
for a pa1ticular purpose; non-infringement; and quality, accuracy, effectiveness, currency, or completeness
of material. In addition, IEEE disclaims any and all conditions relating to results and \Vorkmanlike effort. In
addition, IEEE does not \Varrant or represent that the use of the material contained in its standards is free from
patent infringement. IEEE Standards documents are supplied "AS IS" and "WITH ALL FAULTS."
Use of an IEEE standard is \vholly voluntary. The existence of an IEEE Standard does not imply that there
are no other ways to produce, test, 1neasure, purchase, market, or provide other goods and services related to
the scope of the IEEE standard. Furthermore, the viewpoint expressed at the time a standard is approved and
issued is subject to change brought about through developments in the state of the art and comments received
from users of the standard.
In publishing and making its standards available, IEEE is not suggesting or rendering professional or other
services for, or on behalf of, any person or entity, nor is IEEE undertaking to perform any duty owed by any
other person or entity to another. Any person utilizing any IEEE Standards document, should rely upon his or
her own independent judgment in the exercise of reasonable care in any given circumstances or, as appropriate,
seek the advice of a competent professional in determining the appropriateness of a given IEEE standard.
IN NO EVENT SHALL IEEE BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO: THE
NEED TO PROCURE SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
OTHERWISE) ARISING IN ANY WAY OUT OF THE PUBLICATION, USE OF, OR RELIANCE
UPON ANY STANDARD, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE AND
REGARDLESS OF WHETHER SUCH DAMAGE WAS FORESEEABLE.
3
Translations
The IEEE consensus development process involves the revie"v of documents in English only. In the event that
an IEEE standard is translated, only the English version published by IEEE is the approved IEEE standard.
Official statements
A statement, written or oral, that is not processed in accordance \Vi th the IEEE SA Standards Board Operations
Manual shall not be considered or inferred to be the official position of IEEE or any of its committees and shall
not be considered to be, nor be relied upon as, a formal position of IEEE. At lectures, symposia, seminars,
or educational courses, an individual presenting information on IEEE standards shall make it clear that the
presenter's views should be considered the personal views of that individual rather than the formal position of
IEEE, IEEE SA, the Standards Committee, or the Working Group.
Comments on standards
Com1nents for revision of IEEE Standards documents are welcome fro1n any interested party, regardless of
membership affiliation with IEEE or IEEE SA. Hov;ever, IEEE does not provide interpretations, consulting
information, or advice pertaining to IEEE Standards documents.
Suggestions for changes in documents should be in the form of a proposed change of text, together \Vith
appropriate supporting comments. Since IEEE standards represent a consensus of concerned interests, it is
important that any responses to comments and questions also receive the concurrence of a balance of interests.
For this reason, IEEE and the 1nembers of its Societies and Standards Coordinating Committees are not able to
provide an instant response to comments, or questions except in those cases where the matter has previously
been addressed. For the same reason, IEEE does not respond to interpretation requests. Any person who would
like to participate in evaluating comments or in revisions to an IEEE standard is welcome to join the relevant
IEEE working group. You can indicate interest in a working group using the Interests tab in the Manage Profile
& Interests area of the IEEE SA my Project syste1n . An IEEE Account is needed to access the application.
Com1nents on standards should be submitted using the Contact Us form.
Laws and regulations
Users of IEEE Standards documents should consult all applicable Jaws and regulations. Compliance \Vith
the provisions of any IEEE Standards document does not constitute compliance to any applicable regulatory
requirements. Implementers of the standard are responsible for observing or referring to the applicable
regulato1y requirements. IEEE does not, by the publication of its standards, intend to urge action that is not in
co1npliance with applicable laws, and these documents may not be constn1ed as doing so.
Data privacy
Users of IEEE Standards documents should evaluate the standards for considerations of data privacy and
data ownership in the context of assessing and using the standards in compliance with applicable laws and
regulations.
Copyrights
IEEE draft and approved standards are copyrighted by IEEE under US and international copyright laws. They
are 1nade available by IEEE and are adopted for a wide variety of both public and private uses. These include
4
both use, by reference, in Jaws and regulations, and use in private self-regulation, standardization, and the
promotion of engineering practices and methods. By making these documents available for use and adoption
by public authorities and private users, IEEE does not \Vaive any rights in copyright to the documents.
Photocopies
Subject to payment of the appropriate licensing fees, IEEE w ill grant users a limited, non-exclusive license to
photocopy portions of any individual standard for company or organ izational internal use or individual, noncommercial use only. To arrange for payment of Iicensing fees, please contact Copyright Clearance Center,
Customer Service, 222 Rosewood Drive, Danvers, MA 01923 USA;+ 1 978 750 8400; https: //www.copyright
.com/ . Permission to photocopy portions of any individual standard for educational classroom use can also be
obtained through the Copyright Clearance Center.
Updating of IEEE Standards documents
Users of IEEE Standards documents should be aware that these documents may be superseded at any time
by the issuance of ne"v editions or may be amended from time to time through the issuance of amendments,
corrigenda, or errata. An official IEEE document at any point in time consists of the current edition of the
document together with any amendments, corrigenda, or errata then in effect.
Every IEEE standard is subjected to review at least every 10 years. When a document is more than 10 years old
and has not undergone a revision process, it is reasonable to conclude that its contents, although still of some
value, do not \Vholly reflect the present state of the art. Users are cautioned to check to determ ine that they have
the latest edition of any IEEE standard.
In order to determine whether a given document is the current edition and whether it has been amended through
the issuance of amendments, corrigenda, or errata, visi t IEEE Xplore or contact IEEE. For more information
about the IEEE SA or IEEE's standards development process, v isit the IEEE SA Website.
Errata
Errata, if any, for all IEEE standards can be accessed on the IEEE SA Website. Search for standard number and
year of approval to access the web page of the publ ished standard. Errata links are located under the Additional
Resources Deta ils section. Errata are also available in IEEE Xplore. Users are encouraged to periodically
check for errata.
Patents
IEEE Standards are developed in compliance \Vi th the IEEE SA Patent Policy.
Attention is called to the possibility that implementation of this standard may require use of subject 1natter
covered by patent rights. By publication of this standard, no position is taken by the IEEE with respect to the
existence or validity of any patent rights in connection there\vith. If a patent holder or patent applicant has
filed a statement of assurance via an Accepted Letter of Assurance, then the statement is listed on the IEEE
SA Website at https://standards.ieee.org/about/sasb/patcom/ patents.html. Letters of Assurance may indicate
whether the Submitter is willing or unwilling to grant licenses under patent rights \Vithout compensation
or under reasonable rates, \Vith reasonable tenns and conditions that are demonstrably fi·ee of any unfair
discrimination to applicants desiring to obtain such licenses.
Essential Patent Claims may exist for which a Letter of Assurance has not been received. The IEEE is not
responsible for identifying Essential Patent Claims for\vhich a license may be required, for conducting inquiries
5
into the legal validity or scope of Patents Clai1ns, or determining whether any licensing terms or conditions
provided in connection with submission of a Letter of Assurance, if any, or in any licensing agreements are
reasonable or non-discriminatory. Users of this standard are expressly advised that detennination of the
validity of any patent rights, and the risk of infringement of such rights, is entirely their O\Vn responsibility.
Further information may be obtained from the IEEE Standards Association.
IMPORTANT NOTICE
IEEE Standards do not guarantee or ensure safety, security, health, orenvi ronmental protection, or ensure against
interference with or from other devices or networks. IEEE Standards development activities consider research
and information presented to the standards development group in developing any safety recommendations.
Other information about safety practices, changes in technology or technology implementation, or impact
by peripheral systems also may be pertinent to safety considerations during implementation of the standard.
Implementers and users of IEEE Standards documents are responsible for determining and complying \Vith
all appropriate safety, security, environmental, health, and interference protection practices and all applicable
laws and regulations.
6
Participants
At the time this recommended practice \vas completed, the DC Power Working Group had the following
membership:
Haissam Nasrat, Chair
Richard Hutchins, Secretary
A1nber Aboulfaida
Curtis Ashton
Robert Beavers
Robert Beck
Christopher Belcher
Steven Belisle
Duane Brock
Ja1nes Buniak
Tho1nas Carpenter
Larry Carson
Murad Daana
Rajesh Dhiman
Robert Feisley
Kevin Fellhoelter
David Frankli n
Ali Heidary
David Hood
Wayne Johnson
Roger Kang
Tho1nas Keels
Yves Lavoie
Rufus Lav;hom
Jose Marrero
Tania Martinez- Navedo
Stephen Mccluer
Matthew McConnell
Daniel McMenamin
Jaines Midolo
Sepehr Mogharei
Tho1nas Mulcahy
Bansi Patel
Art Sa lander
Surendra Salgia
Christopher Searles
Joseph S tevens
Tho1nas S tomberski
Kurt Uhli r
Gustavo Varela
Lesley Varga
Stephen Vechy
Jason \.Vallis
Donald \.Vengerter
The following members of the individual balloting committee voted on this recommended practice. Balloters
may have voted for approval, d isapproval, or abstention.
A1nber Aboulfaida
Ali Al Av;azi
Steven Alexanderson
Curtis Ashton
Gary Balash
Rados lav Barac
Tho1nas Barnes
Robert Beavers
Robert Beck
Mark Bo\vman
Jon Brasher
Duane Brock
Jeffrey Brogdon
Chris Brooks
De1netrio Bucaneg Jr.
David Bums
\Vi ll ian1 Bush
\Vi ll ian1 Cantor
Paul Cardi nal
Tho1nas Carpenter
Suresh Channarasappa
Michael Chirico
Randy Clelland
Bryan Cole
Matthew Davis
Ray Davis
John Disos\vay
Gary Donner
Michael Dood
Edgar Dullni
Kevin Fellhoelter
Robert Fletcher
Rostys la\v Fostiak
Dale Fredrickson
David Giegel
Mietek Gli nko\vski
Jalal Gohari
Joseph Gravelle
Randall Groves
Aj it G\val
Ali Heidary
Lee Herro n
Werner 1-loelzl
Robert Hoerauf
Richard Jackson
Anil Jan1es
GezaJoos
Innocent Kan1,va
Peter Kelly
Yuri Khersonsky
Hem1ann Koch
Boris Kogan
Jim Kulchisky
Saumen Kundu
Mikhail Lagoda
Chung-Yiu Lan1
Daniel Lan1bert
Rufus La,vhom
Tin1othy Lensmire
Albert Livshitz
Jon Loeliger
Debra Longtin
7
Jose Marrero
Hugo Marroquin
Michael May
Stephen McCluer
Williain McCoy
James McDo,vall
Daniel McMenan1in
Steven Meiners
Larry Meisner
James Midolo
Sepehr Mogharei
Daleep Mohla
Thomas Mulcahy
Jerry Murphy
Haissam Nasrat
Dennis Neitzel
Arthur Neubauer
Michael Ne,vn1an
Nick S. A Nikjoo
Joe Ni n1s
James O'Brien
Lorraine Padden
Bansi Patel
Anthony Picagli
John Polenz
Thomas Proios
Robert Rallo
Jan Reber
Tin1othy Robirds
Charles Rogers
Thomas Rozek
Ryandi Ryandi
Art Sa lander
Surendra Salgia
Steven Sano
Bartien Sayogo
Robert Schuerger
Christopher Searles
Robert Seitz
Nikunj Shah
Devki Sham1a
David Sn1ith
Jeremy Sn1ith
Ralph Stell
Gary Stoedter
Thomas Ston1berski
K. Stump
Sercan Teleke
Michael Tho1npson
Wayne Ti1run
James Van De Ligt
Lesley Varga
Gerald Vaughn
Stephen Vechy
John Vergis
Donald Wengerter
Kenneth White
Hughes Wike
Jian Yu
When the IEEE-SA Standards Board approved this recom1nended practice on 30 January 2020, it had the
following 1nembership:
Gary Hoffman, Chair
Vacant Position, Vice Chair
Jean-Philippe Faure, Past Chair
Konstantinos Karachalios, Secreta1J1
Ted Burse
Doug Edv;ards
J.Travis Griffith
Grace Gu
Guido R. Hiertz
Joseph L. Koepfinger*
John D. Kulick
David J. Lav;
Hov;ard Li
Dong Liu
Kevin Lu
Paul Nikolich
Dainir Novosel
Jon \.Valter Rosdahl
*Me1nber Eme ritus
8
Dorothy S ta nley
Melunet Ulema
Lei \Vang
Sha \.Vei
Philip B. \.Vinston
Daidi Zhong
Jinf,ryi Zhou
Introduction
This introduction is not part ofTEEE Std 946-2020, IEEE Reco1n1nended Practice for the Design of DC Po\ver Syste1ns
for Stationary Applications.
DC power systems continue to play a vital role in generating station, substation, and telecom controls and
providing backup for emergencies. This recommended practice fulfils a need within the industry to provide
common or standard practices for the design of de power systems. The design features are applicable to all
installations and systems capacities.
The original issue of IEEE Std 946 was pub Iished in 1985 with the title IEEE Recommended Practice for
the Design of Safety-Related DC Power Systems for Nuclear Power Generating Stations. The I 992 revision
changed the title to apply to all generating stations, wh ile including specific guidance and a detailed bibliography
of nuclear design reference standards. This revision makes a general update to reflect the most recent industry
practices as well as substantial additions to annexes. Tn addition to po\ver generation applications, this
recommended practice covers de po,ver system design in substations and telecommunication applications.
Some discussions and illustrative figures have been retained as they offer a constiuctive co111parison to designs
\Vithout having to resort to additional standards.
This reco!lll11ended practice \Vas prepared by a Working Group that is part of the Energy Storage and Stationary
Battery Committee and was sponsored by the Energy Development and Po,ver Generation Committee of the
IEEE Po,ver and Energy Society.
Note that IEEE Std 1818TM and IEEE Std 946 are complementary documents, developed by independent
\VOrking groups. 1
'Information on references can be found in Clause 2.
9
Contents
'I . Overview ................................................................................................................................................... '12
'I . 'I Scope .................................................................................................................................................. '12
1.2 Purpose ................................................................................................................... ............................ 12
'1 .3 Exclusions .......................................................................................................................................... '13
2. Normative references ................................................................................................................................ '13
3. Definitions ................................................................................................................................................. '1 4
4. Organ ization of this recommended practice .............................................................................................. 15
5. Description and operation ......................................................................................................................... '15
5. 'I General ............................................................................................................................................... '15
5.2 System design considerations ............................................................................................................. 16
6. Batteries .................................................................................................................................................... 'I 8
6. 'I Number of battery strings ................................................................................................................... '1 8
6.2 Determination of battery duty cycle and battery size (capacity) ......................................................... 19
6.3 Installation design............................................................................................................................... 2'1
6.4 Maintenance, testing, and replacement ............................................................................................... 22
6.5 Qualificat ion, relevant codes, and standards ....................................................................................... 23
'7. Battery chargers ........................................................................................................................................ 23
'7. 'I Number of chargers ............................................................................................................................ 23
'7.2 Load sharing between paralleled chargers .......................................................................................... 24
'7.3 Determination of rated output ............................................................................................................. 24
'7.4 Installation design............................................................................................................................... 26
'7.5 Output characteristics ......................................................................................................................... 26
'7.6 Qt1alificat ion ....................................................................................................................................... 29
8. Distribution system ................................................................................................................................... 30
8. 'I System layout ..................................................................................................................................... 30
8.2 Distribution panels ............................................................................................................................. 35
8.3 Available short-circuit current ............................................................................................................ 35
8.4 Protective device description and rating ............................................................................................. 3'7
8.5 Voltage ratings for loads ..................................................................................................................... 38
8.6 Qt1alificat ion ....................................................................................................................................... 39
9. DC power system instrumentation, controls, and alarms ........................................................................... 40
9. 'I General ............................................................................................................................................... 40
9.2 DC power system ground detection .................................................................................................... 43
9.3 DC bus undervoltage alarm ................................................................................................................ 45
9.4 Special loading considerations ........................................................................................................... 45
9.5 Design features to assist in battery testing .......................................................................................... 46
9.6 Cross-tie between buses ..................................................................................................................... 46
I0. Protection against electrical noise, lightning, and switch ing surges ........................................................ 46
I0.1 Electromagnetic interference (EM!), radio-frequency interference (RFI) ........................................ 4'7
I0.2 Lightning and S\Vitching surges ........................................................................................................ 4'7
II. Spare equipment ...................................................................................................................................... 48
Annex A (informative) Bibl iography .............................................................................................................. 49
10
Annex B (informative) Batte1y charger sizing- Sample calculations ............................................................ 51
Annex C (informative) Battery available short-c ircuit current-Sample calculations ................................... 56
Annex D (informative) Battery charger and de po"ver system, short-circuit current contribution .................. 59
Annex E (informative) Effect of unintentional grounds on the operation of de power systems ...................... 62
Annex F (infonnative) Telecommunication-specific considerations .............................................................. 65
Annex G (informative) Load sharing of chargers ........................................................................................... 70
Annex H (informative) Center tapped battery design considerations ............................................................. 71
Annex I ( infom1ative) Additional batte1ies in nuclear power generation applications ................................... 72
II
IEEE Recommended Practice for
the Design of DC Power Systems
for Stationary Applications
1. Overview
1.1 Scope
This recommended practice provides guidance for the design ofstationary de po"ver systems. The components
of the stationary de power systen1 addressed by this recommended practice include the following:
Storage batte1ies
Static battery chargers/rectifiers (including sizing)
Distribution equipment
Protection equipn1ent
Control equipment
Interconnections
Instrun1entation
Guidance for selecting the quantity, types, and ratings of equipment is also provided.
The considerations of each of these different components and the issue of load voltage and other load specifics
are discussed in terms of their effect on the design of the \Vhole system. Guidance on short-circuit calculation
and contribution of different de po"ver system components is also offered to improve reliability, perfon11ance,
and safety of the installation.
1.2 Purpose
The purpose of this docun1ent is to provide the user "vith infon11ation and recommendations concerning sizing
and designing de po"ver systems in stationary applications.
While the recommended practices in this document apply to de po,ver systems in substations, additional
guidance for substations is provided in IEEE Std 1818.2
2 Information
on references can be found in Clause 2.
12
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
1.3 Exclusions
Electrically self-contained ac-ac equip1nent and the following co1nponents of the de po>ver syste1n, with the
exception of how they influence the de power system design, are specifically excluded fro1n the scope of this
recorru11ended practice:
The ac power supply to the battery chargers/rectifiers
Photovoltaic, wind, and other alte1native de power source systen1 designs
Loads served by dedicated engine starting battery systen1s
Applications requiring de voltage supply above l 000 V nominal
Motor generator sets
Battery technologies other than lead-acid (L-A) and nickel-cadn1iu1n (Ni-Cd)
Separate syste1ns are usually recon1n1ended for the following special service applications, and are not within
the scope of this document:
Engine (cranking) starting
Emergency lighting
Fire detection and annunciation
Fire protection actuation
2. Normative references
The following referenced documents are indispensable for the application of this document (i.e., they must
be understood and used, so each referenced docun1ent is cited in text and its relationship to this docun1ent is
explained). For dated references, only the edition cited applies. For undated references, the latest edition of the
referenced document (including any amendn1ents or co1rigenda) applies.
IEEE Std 308TM, IEEE Standard C1iteria for Class lE Power Systen1s for Nuclear Power Generating Stations.3•4
IEEE Std 344™, IEEE Standard for Seismic Qualification of Equip1nent for Nuclear Power Generating
Stations.
IEEE Std 450TM, IEEE Recom1nended Practice for Maintenance, Testing, and Replacen1ent of Vented LeadAcid Batteries for Stationary Applications.
IEEE Std 484TM, IEEE Recom1nended Practice for Installation Design and Installation of Vented Lead-Acid
Storage Batteries for Stationary Applications.
IEEE Std 485TM, IEEE Recomn1ended Practice for Sizing Lead-Acid Batteries for Stationary Applications.
IEEE Std 649™, IEEE Standard for Qualifying Class lE Motor Control Centers for Nuclear Power Generating
Stations.
The IEEE standards or products refe rred to in Clause 2 are trademarks owned by The Institute of Electrical and Electronics Engineers,
Incorporated.
' IEEE publications are avai lable from The Institute of Electrical and Electronics Engineers (https://standards. ieee.org/).
3
13
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
IEEE Std 650TM, IEEE Qualification of Class lE Static Battery Chargers, Inverters, and Unintenuptible Power
Supply Systems for Nuclear Power Generating Stations.
IEEE Std 666TM, IEEE Design Guide for Electric Power Service Systems for Generating Stations.
IEEE Std 693-2005TM, IEEE Recommended Practice for Seismic Design of Substations.
IEEE Std 979™, TEEE Guide for Substation Fire Protection.
IEEE Std 1106TM, fEEE Recommended Practice for Installation, Maintenance, Testing, and Replacement of
Vented Nickel-Cadmium Batteries for Stationary Applications.
IEEE Std 11 15™, IEEE Recommended Practice for Sizing Nickel-Cadmium Batteries for Stationary
Applications.
IEEE Std 1187™, TEEE Recommended Practice for Installation Design and Installation of Valve-Regulated
Lead-Acid Batteries for Stationary Applications.
IEEE Std 1188™, TEEE Recommended Practice for Maintenance, Testing, and Replacement of Valve
Regulated Lead-Acid (VRLA) Batteries for Stationary Appl ications.
IEEE Std l l 89TM, TEEE Guide for Selection of Valve-Regulated Lead-Acid (VRLA) Batteries for Stationary
Applications.
IEEE Std 1375™, TEEE Guide for the Protection of Stationary Battery Systems.
IEEE Std 1491™, TEEE Guide for Selection and Use of Battery Mon itoring Equipment in Stationary
Applications.
IEEE Std I 578TM, IEEE Recommended Practice for Stationary Battery Electrolyte Spill Containment and
Management.
IEEE Std 1635™, TEEE/ASHRAE Guide for the Ventilation and Thermal Management of Batteries for
Stationary Applications.
IEEE Std 1818TM, TEEE Guide for the Design of Low Voltage Auxiliary Systems for Electric Power
Substations.
NEMA PE 5, Utility Type Batte1y Chargers.5
NFPA 70®, National Electrical Code® (NEC®). 6
NFPA 70E®, Standard for Electrical Safety in the Workplace.
3. Definitions
For the purposes of this document, the follo\ving terms and definitions shall apply. For terms not defined in
this clause, IEEE Std 1881 [820] , IEEE Standard Glossary of Stationary Battery Terminology [B20], and the
IEEE Standards Dictionary Online shall be consulted. 7
5NEMA publications are available
from the National Electrical Manu facturers Association (https://www.nema.org/).
NFPA publications are published by the National Fire Protection Association (https://www.nlpa.org/).
1
IEEE Standards Dictio11a1y Online is available at: http://dictionary. ieee.org. An IEEE Account is required for access to the dictionary,
and one can be created at no charge on the dictionary sign-in page.
6
14
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
battery cap acity: The quantity of electrical energy, measured in ampere-hours (Ah) or watt-hours (Wh),
produced by a batte1y during discharge.
batte ry charger: A device to restore and maintain the charge of a secondary batte1y. For easier reading,
" charger" is used throughout this reco1nmended practice to refer to battery charger or a rectifier connected to
a battery.
battery state of charge: The stored or remaining capacity in a battery expressed as a percentage of its fully
charged capacity.
duty cycle: The sequence of loads a battery is expected to supply for specified time periods.
nominal battery voltage: The value ass igned to a battery of a given voltage class for the purpose of convenient
designation. The operating voltage of the system may vary above or below this value.
N+X: Parallel redundancy to ensure that the system is ah,vays available. N is the minimum required number of
modules/systems. X is the variable referring to extra units needed for reliable operation.
4. Organization of this recommended practice
Stationary de power systems appear in many applications and industries. All have certain commonalities, \Vhile
some have some unique requirements. It should be noted that these commonalities or unique requirements are
derived from variance in environmental conditions, reliability expectations, and importance of application.
That translates to specific feature requirements or technology differences (thyristor versus high-frequency
switched mode chargers, or lead-acid versus Ni-Cd batteries) that can provide an engineering approach to the
selection/ design of de power systems. For example, a substation charger can be used in a telecom application
and vice versa as long as it can meet the requirements.
Describing every application is beyond the scope of this document, therefore the three dominant applications
are generation, substations, and telecommunications. Large telecommunication carriers may have their O\Vn
internal de power system standards that examine de po"ver systems and their requirements in detail. This is
also beyond the scope of this document, but is worthy of a mention. In this recommended practice, each section
includes subparagraphs reserved for these three applications when there are unique requirements. For other
industrial applications, one can use the recommendations-in part or as a whole-of one of the three dominant
applications. For example, substation application recommendations may be used, \Vhere applicable, for the
design of an industrial process control de power system. It is not the intent of this recommended practice
to exclude other industrial applications. Lead-acid and nickel-cadmium batteries are the types of batteries
primarily used in these applications. Some other battery technologies may be used but are not fully addressed
in this document.
5. Description and operation
D C power systems provide reliable po\ver to critical loads. Examples of critical loads include auxiliary
motors, circuit breakers and S\vitchgear, relays, solenoids, SCAD A, telecommunications equipment, inverters,
e1nergency lighting equipment, fire suppression equipment, etc.
5.1 General
A de system normally consists of one or 1nore battery strings, one or more batte1y chargers/rectifiers and one
or more distribution panels. lf a battery isolation/protective device is used, refer to the battery protection
guidelines of IEEE Std 1375. Refer to simplified typical connection in Figure I for the line of demarcation that
limits the application of IEEE 946 versus IEEE 1375.
15
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
IEEE 946
I
DC Bus
Battery
C harger
-
,,---..._
IEEE 1375
,,---..._
Battery
,,---..._
I+
-
-
To loads
T
Figure 1-IEEE 946 versus IEEE 1375 line of demarcation
In nom1al operation, the battery and battery charging system are both connected to the loads through a common
bus or via a de distribution panel. Therefore, they operate as parallel sources. The battery charging system
applies voltage and supplies current to the battery in order to maintain a full state of charge in the battery. The
charger also generally supplies the continuous load and/or other loads as specified.
Ifthe load exceeds the maximun1 current rating of the battery charging system, the battery charging system output
voltage \Vil! drop, causing all current in excess of the battery charging system rating to be supplied by the battery.
In the event of a failure of the ac po\ver supply to the battery charger (in case the charging system consists of
only one charger), a battery charger failure, or the battery charger being removed fron1 service, the battery
should supply all the power required by the load(s) for some specified periods of time. This is comn1only
referred to as the "duty cycle."
5.1.1 Power generation
Specific design guidance for de power systems for nuclear generating plants are discussed fully in numerous
design standards listed in Annex A.
5.1.2 Substation
For additional guidance refer to IEEE Std 1818.
5.1.3 Telecommunications
Telecommunication installations require de po\ver for aln1ost all equipment, as only a small percentage of
telecommunications equipment is ac-po\vered. Therefore, de po\ver systen1s are required for normal and
backup powering of nlost telecommunications installations. Refer to Annex F for the specific considerations
in telecomn1unication applications.
Telecommunication loads in substations commonly used for telemetry and telecontrol generally operate at lo"ver
voltage (e.g., -48 V de) than the main substation de po\ver system voltage (e.g., 125 V de). A dedicated de po"ver
system might be needed to feed such loads. In cases "vhen the power consumption is relatively lo\v, they can be
fed by the substation de po"ver systen1 through de-de converters (e.g., 125 V de to 48 V de converters).
5.2 System design considerations
In addition to the load's requirements, the design should accommodate any regulatory agency requirements,
safety, or other requirements. The design should also address factors such as reliability, design philosophy,
maintenance, testing, ventilation, floor and "vall loading, and space limitations, etc.
16
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
The equipment to be connected to the battery and batte1y charger \Viii govern the minimum and 1naxi1num
voltage operating voltage range of the de po\ver system.
The designer should consider a single system or multiple systems based on the voltage, current, and redundancy
requirements of the components. For example, if a communication system requires 48 V de input and the de
bus is 125 V de, then consider whether the communication equipment would be either supplied by its O\Vn
battery and charger or by a de-de converter fed from the main 125 V de system.
A recommended practice is to create a diagram at the start of the design process sho"ving the battery or
batteries, charger(s), de panel(s), and all connected loads. Consideration should also be given for future growth.
Redundant de po"ver systems may also be considered. Redundancy may provide flex ibility for maintenance,
testi ng, or replacement in the event of equipment fai lure or the need to upgrade in the future. The abil ity to
connect other de power systems may also aid in maintenance activities.
Although not recommended, center-tapped battery designs have been utilized in special applications. Tn most
cases, a center-tapped battery design should be considered as the least advantageous design, as it may result
in issues such as a more compl icated ground fault detection, battery state of charge imbalance, cell voltage
imbalance, load imbalance, improper autonomy, etc. Designing separate systems for different voltages is
recommended in lieu of center tapped battery. See informative Annex H for more details on center tapped battery.
When special loads such as inverters and de-de converters are connected to a de power system, considerations
should be made to understand their potential effect on the de bus in terms of charger regulation, current and
voltage ripple, transient behavior, or other disruptive interactions.
5.2.1 Power generation
The most common nom inal system voltages utilized in a power generation de power system are 24 V de, 48 V
de, 125 V de, and 250 V de. The following de voltage values are commonly seen on equipment within a plant
and are provided for illustrative reference only:
250 V de
-
Motors for emergency pumps
-
Large valve operators
-
Large inverters
125 V de
-
Motors and valve actuators
-
Control power for relay logic circuits
-
Opening and closing of switchgear circuit breakers
-
Smaller inverters
-
Field flashing
48 V de, 24 V de, 12 V de
-
Com1nunication systems
-
Specialized instrumentation and controls
Some battery technologies (e.g., lead-acid) 1nay exhibit a momentary voltage dip phenomenon known as the
"coup de fouet" \vhen supplying high initial inrush cu1Tents. The coup de fouet effect value is provided by the
manufacturer and should be considered \Vhen sizing battery capacity in order to reduce load disturbances or
shutdo\vns. The coup de fouet beco1nes greater as a batte1y ages.
17
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
Once the voltage li1nits and design parameters are established, the size of the batte1y (number of cells and
capacity) can be established as outlined in applicable best practices or standards, e.g., IEEE Std 485 and IEEE
Std 1115 . lt may be useful to size the battery such that the design duty cycle can be met \Vith a reduced ntunber
of cells (N-1, N-2).
When determining system voltage in facilities wi th long cable runs, voltage drop should be considered. In
order to evaluate the voltage drop across a cable, Ohm's law shall be applied using the maximum current times
the cable resistance provided by the manufacturer.
Legacy equipment or available equipment may constrain the choice of de voltages.
Considerations should be taken for special operation conditions. For example, the capabilities ofa black start or
system restoration plan may require more than one attempt to restore the station ac service. This can be the case
when the main electrical grid power is unavailable, and electrical power is provided by a backup generator.
5.2.2 Substation
Telecommunication loads in substations commonly used for telemetry and telecontrol generally operate at
lo\ver voltages (e.g., 24 V de or -48 V de) than the main substation de system voltage (e.g., 125 V de). A
dedicated de system might be needed to feed such loads. Tn cases where the power consumption is relatively
lo\v, it can be fed by the main substation 's de system through de-de converters (e.g., 125 V de to 24 V de or
-48 V converters). In such cases, considerations for additional needs, such as required incremental power,
voltage regulation, and high-frequency noise should be made when designing the main substation's de system.
For additional guidance refer to IEEE Std 1818.
5.2.3 Telecommunication
Nominal system voltages for telecommunications de po"ver plants are most commonly 48 V de, 24 V de,
and 12 V de. Minimum and maximum telecommunication equipment operating voltage levels are defined in
ANSI/ATIS-0600315.2013 [82] and ANSI/ ATTS 0600315.01.2015 (83]. It is recommended that chargers be
equipped w ith filtered outputs to the appropriate level as charted in NEMA PE 7 [828].
6. Batteries
Stationary batteries are used to supply de power to specified de loads \Vhen the source of ac power to the
battery charger has been disrupted or is insufficient to support the loads. Refer to appropriate IEEE standards
or the battery manufacturer's guidance for the proper selection of batte1y type to be used in the application.
Correct battery selection is essential for reliability, useful life, cost, and 1naintenance planning. Factors such as
operating te1nperature, duty cycle, battery life, and deep cycling should also be considered.
6.1 Number of battery strings
The number of battery strings in an independent de power system should be considered at the design stage.
More than one battery string for capacity or redundancy should be considered to help ensure compliance or
reliability requirements.
6.1.1 Power generation
Some examples of how to improve the reliability and protection of critical loads include the following:
When loads are divided into t\VO or more independent systems, each independent system should be
provided with its O\Vn de power system.
18
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
When maximum de power requirements exceed the capacity of one battery string, the system designer
should consider either of the follo\ving:
-
The use of parallel battery strings.
-
The use of two independent de power systems, each \Vi th its own battery. In this case, selectivity
and coordination of de buses and protective devices should be implemented.
When modifications to the faci lity cause the de po"ver demand to exceed the ex isti ng de power system's
capacity, the installation of a new independent system should be considered. Other alternatives include
the following:
-
Replacing the existing de power system (battery, battery charger, and possibly an addition to the
main d istribution bus/board) with a larger capacity system. (Note that replacing the existing battery
wi th a larger battery may require the replacement of the main distribution.)
-
Installation of a parallel battery. (Note that this may also require additional considerations such as
match ing existing battery capacity, recharging duration, replacement of the main distribution, etc.)
When de power system independence is required and multiple systems are installed, each system
should be powered from separate and/or independent sections of the ac power system ifan alternate ac
source is available (refer to 8.1.1.1 ).
When maintenance and/or emergency reasons require isolating one battery string from another,
individual disconnects or a c ircuit breaker should be recommended in the design.
When maintenance and/or emergency reasons require isolating one battery string from another, the
loading should be reviewed to ensure that the remain ing one battery string has adequate capacity to
operate the worst case loading scenario.
Addi tional power generation applications cases include specific guidance for center tapped battery design
considerations used in a power generating station (see Annex H) and for the number of batteries to be used in a
nuclear generati ng station (see Annex I).
6.1.2 Substation
For additional guidance refer to IEEE Std 1818.
6.1.3 Telecommunications
Parallel strings are very common in telecommunications in order to meet the required backup ti mes as needed
for both the application and redundancy.
6.2 Determination of battery duty cycle and battery size (capacity)
To size a battery correctly, it is important to know the follo\ving:
The number of loads (if1nore than one load)
The load size and duration
The loads sequence in case of multiple loads
The desired design minimum voltage for any de load when the batte1y reaches its end of discharge point
The lowest anticipated temperature the battery electrolyte should encounter
Refer to IEEE Std 485 for lead-acid and IEEE Std 11 15 for nickel-cadmium.
T he overall duration (total batt.e ry discharge time) of the duty cycle cannot be less than the estimated time
interval necessary to restore the charger output to the battery and connected de loads. This estimated time
19
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
interval is determined by engineering judgment, which is greatly influenced by operating experience, and by
the quantity, reliability, and flexibility of the specific off-site po\ver sources and on-site power sources. The
duty cycle may also be influenced by the accessibility of the site.
In theo1y, a minimtun sizing scenario would only require the batte1y to supply de po\ver to the load for
approximately 1 min (the time needed between the loss of ac power and the loading of an operational standby
power source), assuming that after such time, the charger output and de loads \Vould return to normal. In
practice, however, the duration of the battery duty cycle is generally estimated to be any"vhere from 5 min
to 72 h. The selected time depends upon the overall design requirements. It may consider what to do if the
standby power source fails to automatically restore po"ver.
6.2.1 Power generation
The types of de loads encountered in a generating station can include, but are not lim ited to, any or all of the
following:
Annunciator system
Inverter
Emergency lighting
Emergency lube oil and seal oil pumps
Engine starting and control
Generator field flashing
Fire detection and actuation
Main generator output breaker control
Offsite power recovery
Relays and solenoids
Standby or black-start power source starting and control
Switchgear breaker operation, including spring charging motors
Communication systems
Squib valves
Motorized valves
6.2.2 Substation
Most substations have a duty cycle of8 h or more. Substations located in areas that are remote and/or difficult
to access by field service personnel may require more ti1ne to respond and therefore may require a longer duty
cycle. Develop a load profile in accordance \vith appropriate IEEE sizing standards (i.e., IEEE Std 485 or
IEEE Std 11 15) to help ensure correct battery sizing for the application.
A typical substation duty cycle needs to account for the follo\ving types of load:
Continuous loads, e.g., relays, co1nmunications, security, and monitoring
Non-continuous loads, e.g., emergency lights
Larger 1nomentary loads, e.g., breaker coils and motor-operated switches
For additional guidance refer to IEEE Std 1818.
20
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
6.2.3 Telecommunication
For traditional telecommunications sites, a minimum of 3 h to 4 h of battery backup is typically provided for
sites with permanent on-site, auto-start, and auto-transfer standby engine-alternators. A min imum of 8 h is
typically provided for sites not served by a standby engine-alternator. In order to maintain the battery backup
for the most critical loads, non-critical loads are sometimes shed.
The battery is generally sized based on a constant load current during a predefined period of time. However,
if the load is constant power type in wh ich the current rises as the battery voltage drops, then th is increased
current has to be considered in the battery sizing. Typical nominal de bus voltages are + 12 V, +24 V, -48 V,
380 V, and 575 V.
In general, there may be regulatory rules or guidelines that determine the amount of backup time for traditional
voice service. The most common of these rules requires a minimum of 8 h of battery backup for sites not
backed up by an on-site permanent auto-start, auto-transfer engine-alternator. For sites with a permanent onsite auto-start, auto-transfer engine alternator, the rule is typically 3 h of battery backup plus travel time to
the site, or simply a straight minimum of 4 h of battery backup. Non-voice loads (such as broadband data
and video) may be shed in some sites after a certain period of time, or after a certain voltage is reached in the
discharge in order to extend the operation of more essential loads. Typically, this time could range from 5 min
to several hours and is company specific. When this is done, it is typically done off of a timer activated by a
commercial ac fai lure or "battery on discharge" alarm, or the timer starts \Vhen a certain discharge voltage is
reached.
Most telecommunication de power systems (commonly known in the telecom industry as a "de plant")
designed for three or more hours of battery reserve use a battery designed for long-duration discharge. If the
site is powered by renewable energy or has poor grid qual ity, a battery designed for cycl ing duty should be
considered.
6.3 Installation design
The design of each battery installation, signage, battery room design, and access ("vhen applicable) should be
in accordance with the appropriate IEEE standards and local codes. IEEE standards typically include but are
not I imited to the following:
IEEE Std 484 - Installation of vented lead-acid (VLA) batteries
IEEE Std 1187 - Installation of valve regulated lead-acid (VRLA) batteries
IEEE Std 1106 - Installation, maintenance and testing of nickel-cadmium batteries
IEEE Std 1491 - Battery monitoring
IEEE Std 1578 - Spill containment
IEEE Std 1635 - Ventilation and thermal management
IEEE Std 979 - Substation fire protection
Codes commonly enforced in the US include the following:
National Electrical Safety Code® (NESC®) (Accredited Standards Committee C2) [84]
National Electrical Code® (NEC®) (NFPA 70®)
NFPA 70E®, Standard for Electrical Safety in the Workplace
21
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
A controlled environment is recommended for all batte1y installations. Operating outside the manufacturer's
recom1nended temperature range \vill impact the pe1formance and lifespan of the battery.
Factors to consider in designing the optimum environment for a battery system may include the following:
Airborne contaminants, such as corrosive coastal environments (e.g., salt, humidity)
Particulates
Seismic requirements (refer to TEEE Std 693-2005)
Natural disaster exposure (e.g., floods, tsunamis, wildfires)
Adequate hydrogen ventilation
Spill containment
Access, egress, and safety of personnel
Occupancy
Security requirements
Fire detection and suppression
Heat ventilation
Space layout
Clearance requirements (NESC (84], NEC) and additional space needs for large cell lifting devices
Different battery types can vary substantially not only in physical dimensions, but also in the weight of the
filled containers. Design should consider rack size and type (tiered, stepped, or step-tier) and limitations on
number of tiers to limit the height. With very large batteries, lifting devices may be required to place containers
on tiered racks. The design needs space to accommodate both installation and maintenance requirements of
the battery. If racks with more than two tiers are selected, there may be a temperature gradient between top-tier
and bottom-tier cells, \Vhich can affect loading of the cells and impact battery life.
Applicable national/local codes and standards should be researched and analyzed to improve compl iance as
required. Spill containment systems may be required in certain jurisdictions. Check with the authority having
jurisdiction (AHJ) responsible for the particular jurisdiction \Vhere the site is being designed for installation if
in doubt.
6.3.1 Substation
For additional guidance refer to IEEE Std 1818.
6.4 Maintenance, testing, and replacement
Batteries should be maintained, tested, and replaced in accordance \Vith IEEE Std 450 (VLA), IEEE Std 1188
(YRLA), IEEE Std 1106 (Ni-Cd), and/or IEEE Std 1184 [B 18] for UPS batte1y replace1nentdetermination and
local regulations.
The de supply may be equipped with switching devices to facilitate the removal of a battery string from service
for the purposes of offline maintenance and testing. Taking a batte1y offline may require prior connection of
a temporary battery in order to provide continuous backup po\ver supply to the station. Refer to Clause 8 for
more guidance on additional safety isolation devices needed for this operation.
22
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
6.4.1 Substation
For additional guidance refer to IEEE Std 1818.
6.5 Qualification, relevant codes, and standards
Qualification ofbatteries may be required depending on the application specifics.
Seismic and/or environmental qualification of battery installations is based on the appropriate local codes
where the battery will be installed or on good engineeringjudgment based on the application and criticality of
the installation.
For seismically active areas, the battery cabinets or racks, and the batteries should be seism ically qualified in
accordance with the seismic requirements of the International Building Code (!BC).
6.5.1 Power generation
Nuclear power plants have addi tional requirements based on their licensing commitments. Refer to IEEE Std
535 (B 12).
6.5.2 Substation
Transmission entities might have additional requirements based on North American Electric Reliability
Corporation (NERC). Other regions may have similar regulations.
6.5.3 Telecommunication
Equipment used in telecommunications applications by many large North American telecommunication
companies is required to be Network Equipment- Building System (NEBS) compl iant. See Annex F for further
details.
7. Battery chargers
The battery chargers are used to restore electrical energy in the stationary batteries and to supply power to de
loads during normal operation.
7 .1 Number of chargers
At a m inimum, one battery charger should be provided for each battery. Additional battery chargers should be
considered if increased operation flexibi lity, redundancy, or capacity is desired and/or required. For example,
instead of using one very large charger, one can consider using two identical smaller size chargers in parallel.
As long as the constant loads are fully supported by one charger, util izing two chargers provides flex ibility for
routine maintenance and in case of a fa ilure of one of the chargers. The continuous load would be maintained
wi thout affecti ng the battery, but the recharge ti me would be increased. Constant load would be maintained
wi thout affecting the battery. The use of modular chargers in an N+x configuration (i.e., one or more additional
chargers than are necessary to support the critical load) may provide greater reliability than a single charger,
but less than tvvo fully redundant chargers.
7.1.1 Substation
For additional guidance refer to IEEE Std 1818.
23
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
7.1.2 Telecommunication
Typically, N+ 1 or greater rectifier redundancy is used in telecommun ication stations.
7.2 Load sharing between paralleled chargers
When more than one charger is connected in parallel, load sharing betvveen chargers should be considered.
It is not recommended that chargers of different designs and/or technologies be operated in parallel. Refer to
Annex G for more details about load sharing of chargers.
7.2.1 Telecommunication
Parallel operation is the normal mode of operation for telecommunication chargers. For modem rectifiers,
a microprocessor controller gives them all an exact voltage to synchronize, and because of the precision of
their output regulation, they may all share the load equally. If the controller fai ls, the rectifiers should be
equipped with circuitry that will put them at a "fallback" default output voltage (typically very near, if not
the same as the float voltage at 25 °C). In most systems, this "fallback" voltage is set through the controller,
which the min iature microprocessor in each rectifier remembers in case of controller failure. So, even when
the controller fails, a modern telecommunication rectifier system may still share the load equally among the
rectifiers because all rectifiers are programmed with the exact same "fallback" voltage.
7.3 Determination of rated output
7 .3.1 General
The battery charger output current should be sized to feed the connected load wh ile recharging a fully
discharged battery to reach a level greater than 90o/o of its capacity within a predetermined recharging time.
This estimated time is a combination of engineering, judgment, and/or regulatory requirements, and can be
influenced by operating experience as \Veil as the quantity, rel iability, and flexib ility of other power sources,
e.g., on-site or off-site generators, auxiliary network po"ver sources, or redundant de power systems (batteries
and battery chargers).
Equation (1) should be used to calculate the size of the charger as follo"vs:
(I)
\Vhere
C
e
t
IL
K
le
is the ampere-hour removed from the battery as calculated from the battery sizing calculation
based on the actual duty cycle. For other considerations (7.3.3) or if this value is not kno\vn, then
use the published ampere-hour rating of the battery based on the expected discharge ti111e duration
is the recharging efficiency factor. Suggested values, "vhich may be overridden by the battery
manufacturer, are given in Table 1
is the recharge time in hours, usually betvveen 6 h and 24 h. Recharge time may be impacted based
on the minimu111 or 111axi111um charging current. Consult the battery manufacturer
refers to all constant loads and the duty cycle of non-continuous (not transient loads) currents
expected to be supported by the battery charger "vbile recharging or floating the battery (in amperes).
If the load is a mixture of constant cun·ent and constant po"ver type and the average voltage is not
kno\vn, then it is recommended using the mini111um voltage to determine the equivalent constant
current load of the constant power load part
refers to any factors such as but not exclusive to factors related to future load gro\vth, temperature,
altitude, etc., desired by the specified charger. If no other factor is needed, then k = 1
is the minimum charger output current in amperes
24
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
Table 1-Typical battery recharge factors
Battery typ e
Suggested -e- value
VLA
1.1
VRLA
1.15
Vented Ni-Cd
1.3
Pa1tiall y reco1nbinant Ni-Cd
1.14
Special considerations include the following:
For some battery technologies, such as VRLA, excessive charging current might affect their expected
life. Therefore, if the sum of all the loads is used to size the charger and those loads do not operate
concurrently, then the charger will have too much current available for charging the battery. Tf that
current is higher than the manufacturer recommendations, additional controls, such as charger current
limiting, may be required to protect the battery.
Recharge time factor "t" in constant voltage charging: Due to the load voltage range limitation,
chargers may operate in constant voltage charging rather than in constant current charging. In this case,
the battery is recharging at a lower rate and \Viii take much more time to reach its full capacity. Hence,
it is recommended that "t" is fixed to a value (bet\veen 6 hand 24 h, e.g., 8 h) depending on the site
c1iticality and conditions.
Sample calculations ofbattery charger rating are given in Annex B.
7.3.2 Power generation
In addition to the information in 7.3.1 , in po,ver generation applications, the charger shall have a minin1um
output cu1Tent (12) to be capable of simultaneously supplying the constant load (IL) and the non-continuous
load (ILN)· The follo,ving equations apply:
(2)
(3)
\Vhere
13
Jc
12
f uv
13
is the higher value of12 or le
is the charging current calculated in the above paragraph
is the charger minimum output current
is the largest combination of non-continuous loads (e.g., as defined in 4.2.2 of IEEE Std 485-2020)
that \VOuld likely be connected to the bus simultaneously during normal plant operation, including
periodic testing of de components such as emergency lighting and emergency oil pumps
is the final charger output current value for this application
7 .3.3 Substation
In cases \Vhere substations are located in areas that may require more time to respond, it is recommended to
consider using the battery nominal Ah capacity for charger sizing instead of the removed Ah.
Refer to IEEE Std 1818 for more guidance.
25
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
7.3.4 Telecommunication
Modern telecommunication charging is typically done with constant voltage charging, where rectifiers are
connected in parallel \Vith the batteries and loads.
Telecommunication rectifier sizes are typically given in nominal rated de output Amperes (e.g., 50 A, 200 A,
etc.) for typical nominal 48 V de or 24 V de rectifiers. While some of these rectifiers may put out slightly more
current (if the current limit is set to allow it) than the nominal Ampere rating, the nominal rated output number
is used for rectifier sizing calculations. Telecommunication rectifiers may also be rated in constant-power
watts, wh ich means that their output varies by voltage. For sizing purposes, this number is typically converted
to Amperes at the expected battery float voltage. For example, a 3500 W rectifier floating 24 series-connected
1.215 specific gravity (s.g.) cells at an average of 2.20 V/cell "vould be an approximately 66.3 A rectifier
(3500 W/52.8 V).
7.4 Installation design
7 .4.1 General
A controlled environment is recommended for all charger installations. Additionally, installation design
should consider the follow ing:
Corrosive atmospheric conditions
Nearby industrial processing sites
-
Coastal environments
-
Particulates from nearby industrial processing sites, dust, and dirt
Environmental conditions that would accelerate component degradation, e.g., sunlight exposure, high
ambient temperature
Humidi ty
Seismic requirements required by authority havingjurisdiction (refer to IEEE Std 693-2005)
Natural disaster exposure, e.g., floods, tsunamis, wildfires
Accessibility for maintenance of components, adequate space for egress and safety for personnel
Security requirements, e.g., physical access, cybersecurity
Requirements for replacement under live service
7.4.2 Substation
In addition to the information in 7.4.1 , working clearance per AHJ such as the NESC (84] or NEC in the US
should be considered.
7 .5 Output characteristics
All charging sources should meet the requirements of applicable standards such as NEMA PE 5, and the
following performance characteristics. A connected battery may need additional requirements such as:
Battery maximum charging current and voltage
Max imum current and voltage ripple
Temperature voltage compensation
Other characteristics in order to maximize its service life under the operating conditions
26
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
Connected de loads may also have additional requirements such as:
Maximum and minimum voltage levels
Maximum noise and ripple levels
Inrush currents, etc.
Certain site conditions may require the charger to perform \Vith or without the battery connected \Vith
continuous and non-continuous loads, but not momentary loads.
7.5.1 Ripple and transients
Output ripple, transients, electromagnetic interference (EM!), and radio frequency interference (RFI) should
be limited to:
Help extending battery life
Avoid impact on sensitive loads
Comply w ith the applicable electromagnetic compatibility (EMC) regulations, such as the Federal
Communications Commission (FCC), !EC 61000-4 series [87], and IEEE Std 650 standards
Ripple frequency and current magnitude are the components of concern to batteries. The lower the frequency
and the higher the amplitude of the current ripple, the more damage is being done to the battery. Chargers
compliant to international standards, w ith properly functioning filter circuits, produce ripple at such lo"v
magnitudes as to be inconsequential to battery life. Almost all damaging ripple current to batteries is due to
connected ac inverters and other converters.
Ripple values are usually measured at rated charger voltage and current conditions into a resistive load \Vith
a connected battery. The maximum voltage and current ripple should be specified based on the connected
load(s) ripple limitations.
NOTE- Although inverter reflected ripple effect on the de bus is not covered by th is document, special considerations
1nay be needed to reduce the total ripple content to a level to help ensure correct load operation and to help extending
battery Ii fe.
7 .5.1.1 Power generation
Chargers being equipped with filtered outputs to the appropriate level as charted 1n NEMA PE 5 1s
recom1nended.
7 .5.1.2 Substation
Same as the information provided in 7 .5. 1.1.
7.5.1.3 Telecommunication
Historically, telecom1nunication de syste1ns provided po\ver to equipment that served analog voice circuits.
For this reason, it was particularly important to keep electrical transients and noise to a 1ninimum in the voice
band bet\veen 20 Hz and I 0 kHz. Because de ripple in older silicon controlled rectifier (SCR) and ferroresonant
chargers \Vith 50 Hz ac or 60 Hz ac input was primarily of the primary frequency and lower order harmonic
multiples, this unfiltered ripple fell right into the voiceband. For this reason, telecommunication rectifiers
have historically been well filtered, producing minimal ripple to the batte1y and load (some older rectifiers
were not as well-filtered as others, and thus needed the battery to provide additional filtering to negate the lowfrequency ripple for the "talk batte1y." However, all modern telecommunication rectifiers are well-filtered).
27
IEE E Std 946-2020
IEE E Recommended Practice for the Desig n of DC Power Systems for Stationary Applications
It is recommended that chargers be equipped \Vith filtered outputs to the appropriate level as charted in NEMA
PE 7 [B28].
7.5.2 Operation without a connected battery
For some designs or operation needs, disconnecting the battery from the de po"ver system for maintenance
may be appropriate. Chargers should be capable of supplying the small incremental and continuous loads
wi thin the charger's rated capacity.
NOTE- Due to the dynarnic response tirne of chargers, they are not usua lly capable of supplying the large inrush currents
requ ired by sorne loads. T herefore, it is not reco1n 1nended to use chargers \Vithout a connected battery. Other operation
techn iques, such as using a temporary battery or transferring the load to other de po\ver systems, should be considered.
Also, a deterioration of the voltage regulation and output r ipple may be experienced \Vhen the battery is
disconnected. If the increase in voltage regulation or r ipple cannot be tolerated, the maximum allowable
values should be specified.
A charger with improved filter c ircu it may further reduce the voltage and current ripple magn itude \Vhen
battery is d isconnected.
7.5.2.1 Power generation
Operation wi thout a connected battery is not recommended. Caution is to be taken if the de power system
is a source for the protective relays and c ircu it breaker tripping. Disconnecting the battery may affect the
protection system.
7 .5.2.2 Substation
Operation without a connected battery is not recommended. Breakers may operate at any time, particularly
during a fault. A charger may not have the instantaneous response time needed to respond to a sudden large
current demand, resulting in possible breaker misoperation.
7.5.2.3 Telecommunication
Typically, telecommun ication loads do not include excessive transient loads; hence modern telecommunication
rectifiers can operate w ithout a connected battery due to their fast response time and inherent high output fi lter.
7.5.3 Remote voltage sensing and battery temperature compensation
When chargers and batteries are located at a distance from each other, a significant voltage drop may occur
in cables. Re1note sensing leads should be considered to he lp ensure proper voltage regulation to the battery
terminals. T he loss of remote sensing shall not cause the charger voltage to increase.
When sensors at the battery indicate that operating tem perature has deviated outside the acceptable te1nperature
range, te1nperature compensation can adjust voltage in proportion to the temperature shift. Te1nperature
co1npensation can extend the life of a battery and when the batte1y is overheating, can mitigate and possibly
prevent thermal runav;ay and fire. Temperature compensation might be an option on some chargers and is
recom1nended for use \Vith lead-acid and nickel-cadmium batteries.
A temperature compensation sensor installed on the negative terminal ofthe pilot cell or block is recommended.
M ini1num and maximum voltage compensation levels should be considered to help ensure proper load
operation.
In installations with parallel chargers, any remote voltage and temperature sensing should be located at the
same point.
28
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
7.5.3.1 Telecommunication
The rectifier controller can set the output voltage of the rectifiers based on internal sense, or external sense.
Internal sense is when the output voltage setting is matched to the common "hot" de output bus voltage in that
bay or shelf relative to the grounded return bus in that bay or shelf. This is commonly used in small de plants
where the batteries are very near the rectifiers. In larger de plants \Vhere the batteries are further away, there is
usually more voltage drop between the rectifiers and batteries, and charging the batteries based on the voltage
at the batteries is more important. In those cases, external sense is used. A pair of external sense \Vires are run to
the positive and negative battery termination buses above the battery stand(s). The rectifier output voltage(s)
are adjusted to provide the float voltage set in the controller so that voltage is produced at those "remote"
battery term buses.
Some older rectifiers, while connected to a controller, cannot have their output voltage finely adjusted by the
controller; thus, each rectifier has its own potentiometer (pot) for setting its output voltage. In such plants,
where they still exist, the rectifiers have to have their "pots" adjusted periodically in order to share the load
fairly equally. S ince each rectifier regulates itself rather than receives a voltage regulation s ignal from the
controller, the voltage of each rectifier may drift ever so slightly over time in relation to the other paralleled
rectifiers. A difference of even a few millivolts can cause huge variations in current sharing between rectifiers;
thus the "pots" have to be periodically adjusted (typically this is done every 6 to 12 months).
7.5.4 Output current limit
Stationary battery chargers are typically constant voltage, designed to limit their output current. When the
charger operates in current limit mode, it is operating as a controlled constant current source to restore the
bulk Ah of the battery and feed the de load. The current should taper as the battery approaches full charge. The
charger can also enter current lim it mode if it is overloaded or a fault appears on the de bus. The current limit
should protect the charger components from premature fai lure due to overload. For some battery technologies
such as VRLA, where charging capacity exceeds the battery's allowable charging limits, dedicated battery
charge current limiting shall be employed.
7 .6 Qualification
The objective of equipment qual ification testing is to demonstrate that the equipment functions with in its
specified range during normal and abnormal operating conditions.
Equipment qualification can normally be divided into the following t\VO major subsets:
Performance qual ification
Environmental qualification
The purpose of environmental qualification is to demonstrate that the charger will operate on demand to meet
system performance requirements wi thin specified environmental parameters. Environmental qualification
covers radiation, electrical environment, temperature, humidity, seismic, and EMI/RFT conditions among
other parameters as defined by each component's application and setting. All chargers/rectifiers are required to
be qualified per performance and environmental standards.
Specific performance and environmental qualification requirements for chargers/rectifiers applicable to all
industries covered in this recommended practice are described belo\v.
7 .6.1 Power generation
Performance and functional qualification for chargers is defined in N EMA PE 5.
29
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
For Class l E, environmental qualification for chargers/ rectifiers is defined in several standards. Radiation
and temperature qualification are dictated by IEEE Std 650. Seismic qualification is defined in IEEE Std 344.
Installation, inspection, and testing requirements for power, instrumentation, and control equipment at nuclear
faci lities are defined in IEEE Std 336 [B8].
For non-Class IE, seismic qualification for chargers/rectifiers are defined by IEEE Std 693-2005 and the
International Building Code (!BC), Chapter 5. EMT/RFI qualification guidance is defined by EPRI TR- I 02323
(B6] among other documents.
7.6.2 Substation
Performance/functional qualification for battery chargers is defined in NEMA PE 5. Seismic qualification for
chargers/rectifiers is defined by IEEE Std 693-2005.
7.6.3 Telecommunication
Rectifiers used in telecommunications applications by many large North American telecommunication
companies may be required to be Network Equipment Building System standards (NEBS) compliant. The
documents applicable to rectifiers and their racks/shelves/cabinets are GR-63, GR- I 089, and SR-3580. Tn
addition to the flammability and seismic requirements ofGR-63 mentioned for batteries, GR-I 089 testing covers
electromagnetic emissions and susceptibil ity, short-circuit testing, power input fault immunity, and safety.
In addition to the basic NEBS specifications produced by Telcordia, the telecommunications company may
require the manufacturer to have their rectifiers and rectifier controllers meet GR-1 51 (central office rectifier
requirements), GR-221 (rectifier controller microprocessor requirements), GR-947 (s"vitchmode rectifiers),
GR-1515 (VRLA thermal runaway detection and control, which is called for by the Fire Codes), GR-3108
(temperature-hardening for outdoor application environments), TR-NWT-000154 (central office po"ver plant
control and distribution equipment), and TA-NWT-000406 (very small de power systems).
Telecommunication rectifiers used in buildings not owned by the telecommunications company are typically
safety-tested and listed to at least UL 60950-1.
Telecommunication operators wanting to specify higher-efficiency rectifiers in order to save money on the
electric bill and reduce carbon emissions may require compliance w ith ANSI/ ATIS-0600015.04.20 I 0 and/
or the US EPA Energy Star Uninterruptible Po\ver Supplies specification (while specifically using UPS in the
title, the specification also covers de power system rectifiers).
In addition to the traditional ISO 9001 quality manufacturing standards required by 1nost users, many
telecommunications companies also require manufacturing compliance to the additional telecommunications
specific quality requirements ofTL9000 [B3 l ].
8. Distribution system
8.1 System layout
8.1.1 Electrical
The de power system layout and connection depend on the site design criteria. The chargers, batteries, and
loads can be connected at various points. Fcodes cFor example:
In substations, chargers can be directly connected to battery terminals, to the load side of batte1y
disconnect S\Vitch (if one exists), or to a de panel branch circuit. Batteries can also be directly connected
to the de panel(s) main lugs, de panel branch circuit, etc., \Vi th or without a disconnect switch.
In power generation applications, all de components are typically connected to a common de
distribution panel.
30
IEE E Std 946-2020
IEE E Recommended Practice for the Desig n of DC Power Systems for Stationary Applications
The layout \Vould also affect the reliability or service of the de power syste1n. For example, ifthe charger is
directly connected to the battery side of the disconnect S\vitch, it could be considered a reliable method of
charging the batte1y, since there are minimal points of failure in between the charger and batt.ery. However,
since the charger also serves to supply power to continuous loads under no1mal operation, a fault on the batte1y
or removal of the battery for replacement (by opening the battery disconnect S\Vitch) may disconnect the
charger from the loads.
8.1.1.1 Power generation
The optimum de supply selection and distribution system design for any given plant is based on the design
criteria established for that plant.
Figure 2 is the key diagram for a typical 125 V de po"ver system.
Battery
Discharge
Test
125Vd c
Battery
125Vdc
N.O.)
Dist. Bus
-j~+---+----~;,___,..--,_~
!
'
.,
, .J Alternate
AC Input
Battery
Charger
L7 ,---·, - ,. - ,_
1--4-- - -·---·------
-
--~
+-
TO
LOADS
,..--,_
-
-
+-
N.O.
Cross-Ti e
...... ----------------Main Distribution Panel
NOTE I-All breakers are norn1ally closed except those marked "N.O."
NOTE 2- 0ptional or alternate features indicated by--------.
NOTE 3-Fuses n1ay be substituted for breakers.
NOTE 4-Diagram is not n1eant to depict redundancy.
Figure 2-125 V de power system key diagram
If replacement of equipment is required in the future \Vithout pe1forming a major shutdown of loads, t\vo
distribution panels or accommodation for connecting alternate connection point to the de bus would be
required. This also applies to systems that require spare, redundant, or temporary system to be connected.
For critical facilities, a dual-de power system design may be considered to help ensure compliance or reliability
requirements. Under normal operation, redundant de po\ver systems are meant to be operated in isolation from
one another. In the case that one de power system fails, the other shall remain available in order to maintain
continuity of power supply to critical loads in spite of a contingency scenario.
31
IEE E Std 946-2020
IEE E Recommended Practice for the Desig n of DC Power Systems for Stationary Applications
Figure 3 is the typical diagram for a class 1Ede power system.
Div. 1
Div. 2
Battery
Discharge
Test
Battery
Discharge
Test
125Vdc
Battery
-J
)
,.-..._
125Vdc
Dist. Bus
125Vdc
Dist. Bus
N .O
,__
AC Input
_,.-...
(
+
+
Batlery
Charger
Spare
Battery
Charger
-; ,__- ,__
--
,__
,.-...
N.O.
,__
t-
Atternate (
) Altemale
AC Input
125Vdc
Batlery
D
I
D
I
v
v
1
2
L
L
0
0
A A
D D
s s
- ~ \
- -,.-...
>- ,.-...
-
~
N.O.
Battery
Charger
AC Input
Spare
Battery
Charger
N.O.
~
AC Input
,.-..._
-- N.0.
N.0.
,__
-
,.-...
Cross-Tie
Cross~Tte
Main Distribution Panel
Main Distribution Panel
NOTE I- All breakers are normally open except those tnarked "N.O."
NOTE 2- Fuses 1nay be substituted for breakers.
NOTE 3- 0ptional or alternate indicated by--------.
NOTE 4- Class IE refers to nuclear safety re lated de po\ver syste1ns.
NOTE 5- Div. I and div. 2 refer to redundant loads required in nuclear systems.
Figure 3-125 V Class 1E de power system key diagram
8.1.1.2 Substation
It is preferable that all de components would be connected to a de distribution panel. In redundant de power
systems, this distribution panel can be fed from another de source through a tie overcurrent protective device
(OCPD). For additional guidance, refer to IEEE Std 1818.
8.1.1 .3 Telecommunication
DC plants for central offices typically are equipped with t\vo or 1nore rectifiers based on the load, capacity to
recharge the battery, and N+ I redundancy as a maintenance spare. The batte1y often consists of two or more
parallel battery strings sized to carry the load for a specified reserve time within the operating voltage \Vindo"v
of the telecommunications systems. The overcurrent protection (OCP) system is made up of pri1nary bays
that in turn feed secondary distribution bays [generally called batte1y distributing fuse bays (BDFB), breaker
bays (BDCBB), power distributing (PD) bays, or similar tenns] (see Figure 4). These secondary distribution
bays feed the telecommunications bays, cabinets, or relay racks by \Vay of a te1tiary level of overcurrent
protection (fuse panels) within bays. A microprocessor-based controller governs de plant operation and
provides 1nonitoring and alarm reporting. The batte1y disconnects shown in Figure 5 are optional based on
system design. The majority of grounded systems will not break the battery to ground connection. Breaking
this connection is typical in non-grounded de po\ver systems. Figure 4 sho\VS a typical central office de power
plant, and Figure 5 shows a typical redundant de design.
32
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
:--1
RECT
RECT
CONTROL & ALARM
,..r
......r-
...
~-
RECT
t:;::
-- --=-=-- -..::..:::..
---=-=--~
-~
~
...~
...
~-
~
t:;::
PRIMARY
DIST
PRIMARY
DIST
L
0
A
D
SECONDARY
DIST
_--;:::...
-
>->->--
--
I==
-Figure 4-Typical central office de power plant
,. -
I
--
CONTROL & ALARM
.
REC T
RECT
RECT
,..,
~
''
1-o,..,_
-
'I
-
I
I
--
I
I
I
-
--
I
I
I
I
I
L
0
2-POLE
PRIMARY
DIST
A
D
I
,
,
.
--Figure 5-Redundant rectifier design
DC plants for small facilities, such as cell or microwave sites, typically are equipped \Vith two or more rectifiers
based on the load, capacity to recharge the battery, and n+ 1 redundancy as a maintenance spare (see Figure 5).
The battery consists of one or more parallel battery strings sized to carry the load for a specified reserve time
wi thin the operating voltage w indow of the telecommunications systems. DC power distribution consists of
primary panels of fuses, circuit breakers, or a combination of devices that feed the telecommunications bays,
cabinets, or relay racks, usually by way of a second level of overcurrent protection (fuse panels) \Vithin bays.
A microprocessor-based controller governs de plant operation and provides monitoring and alarm reporting.
33
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
While a larger telecommunications site may have multiple de plants, almost a11 de plants only have one set
of power distribution panels or bays connected to the common de bus to \Vhich the batteries and rectifiers are
connected. From this single point, typica11y dual (A and B) feeds are derived to po\ver seconda1y and tertiary
fuse or breaker panels. Most loads are dual fed so that loss of a single feed does not dis1upt service.
8.1.2 Physical
Consideration needs to be given to where the de distribution equipment is to be located. It is usually advisable
to place the equipment (battery, charger, and main distribution panel) as close to the electrical load as possible,
thus reducing voltage drop issues and accommodating maintenance and testing activities.
DC control and instrumentation cables should be routed in separate raceways from power cables in order
to reduce the impact of surges or transients. These surges and transients can be present in ac and de power
systems and/or grounding cables as a result of S\Vitching activities, lightning strikes, or undetected ground
faults.
8.1.2.1 Power generation
Ifthe ac auxiliary po"ver system is separated into two or more independent divisions with a de power system
for each ac division, then the equipment and cables for each de po"ver system division should be separated
from the equipment and cables at the other division to the same extent as employed for the ac system.
Sizing and routing of cables should take into consideration inrush currents from motor loads that occur during
breaker or switch mechanism operation, and their impact on voltage drop throughout the de power system.
In any arrangement, it is recommended to run the positive and negative main cables in separate conduits so
that any fault on these cables will first be polarity-to-ground before a polarity-to-polarity fault can develop.
The use of nonmagnetic conduit should be considered so as to reduce the inductance of this circuit. A circuit
wi th lower inductance will reduce the magnih1de of voltage spikes generated and reflected into the de power
system when high-current load circuits (such as motors, inverters, and faults) are intern1pted. In addition, a
highly inductive circuit may adversely affect the performance of current-limiting fuses if utilized.
A throw-over or transfer switch should be considered to a11ow removal of one of the batteries (and chargers)
from service for maintenance purposes.
Depending on the application, other standards related to cable specifics (e.g., installation, splicing, flame
resistance) such as IEEE Std 383 [89] and IEEE Std 1202 [B19] may also be required.
8.1.2.2 Substation
For additional guidance refer to IEEE Std 1818.
8.1.2.3 Telecommunication
In typical telecommunication sites with only a single de plant using lead-acid batteries, the plant is typica11y
located on the ground floor or basement because of the additional floor loading capacity provided by a s labon-grade floor to support the heavy \Veight requirements of lead-acid batteries. When multiple plants exist in
a site, they may be located 1nuch closer to the loads they serve in order to reduce voltage drop, po\ver losses,
copper cable size, and cost. The batteries, rectifiers, and prima1y distribution bays are typica11y located in
a compartmentalized "power room" for medium and larger plants, but there are now sma11er distributed de
plants co11ocated with the loads they serve.
Batte1y positive and negative leads should be 1un adjacent to each other on the same cable trays, conduits, or
racev;ays throughout the de distribution system to provide electromagnetic noise field cance11ation.
34
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
In typical customer premises applications (commercial customer-owned building \Vhere the teleco1nmunications
company provides the equipment), the de plant (including its batteries) takes up a portion of a bay in the
com1nunications room of the customer, or in small applications may even be wall-1nounted batteries, rectifiers,
and miniature distribution.
In modern remote terminal (RT) outdoor cabinets , the de power plant may be in the same chamber as the
electronic equipment, or in an end chamber of the cabinet. The batteries are usually located in a completely
separate (and separately-ventilated) compartment, generally in the lowest level of the cabinet.
" Unfused" de cabling (the cabling that connects the rectifiers, batteries, and d istribution systems in parallel)
wi thin the de power plant is typically run on separate overhead open frame cable racks in telecommun ications
company faci lities. The primary distribution cable to the BDFB or larger runs to equipment bays (these feeds
are typically fused at I 00 A or greater) are typically run on a separate overhead "power" cable rack from the
rest of the communications type cables in larger facil ities. Secondary power distribution from the BDFB to the
equipment bays or smaller runs directly from the PBDs to equipment bays (these feeds are typically fused at
ground smaller than I 00 A) are typically run on cable hangers or horns attached to the outside of the overhead
cable rack used for the copper communications cables. DC grounding system cables are also typically n1n in
this way (on hangers or horns attached to overhead cable rack).
In huts, customer premises, and sim ilar facili ties, the de power cable is typically run on the same overhead
rack with the rest of the commun ications cables, but is usually segregated on the rack from the rest of the
communications cables.
8.2 Distribution panels
The number of distribution panels is determined by the anticipated number of loads served from the de power
system and system design requirements. Consideration should be given to probable growth requirements over
the expected useful life of the site as well. It may be desirable in certain cases to employ one distribution panel
for each independent de power system. A minimum of one d istribution panel should be provided for each de
power system. It should be located as close as possible to the battery.
The buses in the main distribution panel should be protected for personnel safety and to reduce the probability
of bus faults. However, these buses must be designed to handle mechanical and electrical stresses caused by
transient faults or during normal operation. Proper spacing or barrier rated for the maximum system voltage
should be provided between the positive and negative line leads at the main distribution panel. The design
should follow applicable safety standards and meet or exceed expected short-circuit currents.
8.2.1 Substation
For additional guidance refer to IEEE Std 1818.
8.3 Available short-circuit current
T he maximum available short-circuit current for the de power system is the sum of that delivered by the battery,
charging system (one or more chargers in parallel), and inductive load, such as motors (as applicable). This
available short-circuit current must not exceed the intenupting capacity of feeder breakers and fuses, as \Veil
as the withstand capability of the distribution buses and disconnecting devices. When a more accurate value
of maximum available short-circuit current is required, the analysis should account for interconnecting cable
impedance. When a fault occurs on a de po\ver syste1n and depending on the impedance of the fault circuit,
a substantial voltage drop can occur that could affect or dis1upt operating connected loads. As a result, the
overcurrent protective device (OCPD) should be selected and coordinated to minimize the de pov;er system
lo"v voltage and the duration caused by a fault. DC busses and panels should also meet or exceed expected
short-circuit currents. Only the OCPD feeding the fault and the closest to the fault should trip, leaving the rest
35
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
of the system intact to continue supplying pov;er to unaffected areas. Refer to Annex C and Annex D for details
and examples of battery and charger short-circuit current contributions.
8.3.1 Batteries
The short-circuit current across a battery terminal is a function of its actual voltage, internal impedance,
intercell connectors impedance, and other series connected components. Note that internal impedance is
significantly affected by the chemistry and internal design for all battery types. Also, the rise time \Vhen a
short-circuit occurs is relatively fast. If connecting multiple battery strings in parallel, the fault current
contribution from each string needs to be considered. It is recommended to contact the battery manufacturer
for the maximum short-circuit current value. Refer to Annex C for calculations.
As an example, the fault current from a large vented lead-acid storage battery resulting from a bolted short at
the battery terminals will typically exh ibit a rate-of-rise that delivers the peak current within 17 ms. The fault
current for a short at the de distribution switchgear or panelboard peaks later (typically within 34 ms to 50 ms)
depending on the cable size and layout, due to the inductance of the de power system in series with the fault.
The magnitude of the fault current for a short at the distribution bus will also be lower than the value at the
battery due to the resistance of the cables between the battery terminals and the bus.
8.3.2 Chargers
The available short-circuit current at the charger output depends on the charger design topology. T\VO
predominant designs are used:
Low-frequency rectifying chargers such as SCR, ferro -resonant, and magnetic amplifier
High-frequency rectifying chargers commonly called switched mode chargers
Tests on current-limited low-frequency battery chargers have shown that the initial short-circuit current
contributed from the battery charger can exceed the current-limited value. A large transient current spike may
occur during a certain time before the charger current limit mode activates or a protective device opens.
Whether the battery charger is isolated from the battery or is operating in parallel with the battery can change
the response and contribution to the fault current.
The following considerations should be taken in account:
a)
The stored energy in filter circuits (capacitors): This instantaneous peak short-circuit current may
approach a value 200 times the charger rated current. However, the time duration of the initial transient
current is sho1t (in the order of 2 ms due to the inter connections impedance) and generally does not
affect the ratings of equipment and protective devices.
b)
The rectifier and the upstream circuit impedance: After the stored energy in the filter circuits is
dissipated, the magnitude of the transient short-circuit current is dependent on the L/R ratio of the
ac supply as \Veil as the inductance and resistance of the transformer-rectifier-fault circuit (rectifiers,
filter inductors, shunt, and other series co1nponents).
c)
Response time to current limit: This time is typically 300 ms. for a controlled SCR type charger, and
the transient current amplitude is typically l 0 to 12 times the rating of the charger. The time to current
limit for a controlled ferro resonant charger is less due to its inherent design. In all cases, consult the
battery charger's manufacturer for short-circuit capabilities, response to a short-circuit, and time to
current limit under sho1t-circuit conditions. The battery charger's time to current limit under a shortcircuit condition should be considered along with the impedance and L/R ratio of the fault circuit
when selecting and coordinating OCPDs.
36
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
High-frequency battery chargers behave in a different manner. Because of the high-frequency controllers,
current limiting and voltage regulation are relatively faster, so their contribution in the fault is usually limited
to no more than 150o/o of the rated charger output current.
Generally, at the distribution panel charger connection, external cable gauge and length may reduce the fault
contribution to levels where charger internal protection may not operate. In all cases, consult the charger's
manufacturer for the charger time constant and its short-circuit capabilities at its output terminals.
It may be appropriate to select a battery charger technology with higher short-circuit current capabilities if
needed to more efficiently trip the protection device (OCPD) closest to the faul t.
8.3.3 Other equipment
Operating inductive loads, such as de motors, contribute to the total fault current. T he maximum current that a
de motor delivers to a short-circuit at its terminals is limited by the effective transient armah1re resistance (Rd')
of the motor. For de motors of the type, speed, voltage, and size is typically used in generating stations. Rd' is
in the range ofO. I to 0.15 per unit. Thus, the maximum fault current for a short at the motor terminals typically
ranges from seven to ten times the motor's rated armature current. Therefore, it is conservative to estimate
the maximum current that a motor contributes to a fault as ten times the motor's rated full load current. When
a more accurate value is required, the short-circuit contribution should be calculated using specific Rd' data
for the specific motor, or actual test data should be obtained from the motor manufacturer. For additional
accuracy, the calculation should account for the resistance of the cables betvveen the motor and the fault.
8.4 Protective device description and rating
Selective protection coordination of the OCPDs in de po"ver system components is of the utmost importance. Tn a
fault condition, the closest OCPD to the fault should trip first to isolate the fault so that less equipment is affected.
An OCPD may have different rat ings for ac and de applications. In de application, it is h ighly recommended
to only use de rated OCPD. Although empirical methods can be found to convert OCPD ac characteristics to
de equivalent values, it is recommended that the OCPDs and associated hard\vare such as fuse holders, are
certified to w ithstand the appl ication maximum operating voltage (not nominal voltage) and current.
An OCPD, such as a circuit breaker or a fuse, is recommended between the de sources [e.g., battery, charger(s),
etc.] and the rest of the connected load equipment. Associated OCPD status monitoring (e.g., breaker open
relay or blown fuse indication) may be needed to provide annunciation of alarm conditions.
In addition, if a h ighly inductive load or circuit is connected, it may adversely affect the pe1formance of
OCPDs. OCPD de ratings are usually based on a fixed maximtun time constant (L/R).
All protective and disconnecting devices should be properly rated for short-circuit current and 1naxi1num
operating voltage of the de application. (e.g., float, equalize).
Also, a 1nanual isolation device is recommended between the battery terminals and the main distribution buses
and panels.
If more than one batt.e ry string is used, a manual isolation device is recommended betvveen each string and
the main distribution buses and panels. When using an OCPD on each string, OCPD size and characteristics
should to be selected based on the \Vorst-case scenario operation, such as if one string is disconnected and the
balance of the battery is subject to cany a heavy load, or in case of a fau lt on the de bus. Refer to IEEE Std 1375
for the guide of battery protection.
T he distribution bus and battery OCPD should have interrupting capacity or short-circuit current \Vithstanding
capability that exceeds the maximum short-circuit current available for the system voltage and ambient
tem perahU"e.
37
IEE E Std 946-2020
IEE E Recommended Practice for the Desig n of DC Power Systems for Stationary Applications
The continuous current rating of the batte1y OCPD should be selected to accommodate the maximum sustained
current in the batte1y duty cycle, and should have instantaneous current rating and delayed current rating (12t) to:
Help prevent the undesired operation of the battery OCPD during the highest duty-cycle current
magnitude and duration
Facilitate proper coordination with downstream OCPDs
Protect the main distribution bus and cabling
When cross-tie OCPD and/or battery test-OCPDs are used, then protection and coordination for transfer inrush
currents from all energy sources and short-circuit currents shall be considered. It is important to understand
that there is a difference between protection and coordination (see IEEE Standards Dictionary Online).
8.4.1 Power generation
For the battery to meet the demands of momentary or random loads (see definition in IEEE standards such
as IEEE Std 485, IEEE Std 1115, etc.), the protective device's interrupting rating should be sized adequately.
Consult the battery manufacturer for ratings for discharge duration less than I min.
The main protective device should coordinate with all downstream protective devices. The distribution bus
and any manual disconnecting device should have a short-circuit current \Vithstand capability (i.e., bracing)
that exceeds the maximum short-circuit current available.
NOTE- For grounded syste1ns, protective devices should be coordinated. For ungrounded syste1ns, these devices 1nay
not be fi.t lly coordinated; hoi.vever, the devices should be selected to protect the associated infrastructure, including cables.
8.4.2 Substation
Same as 8.4. 1.
8.4.3 Telecommunication
Same as 8.4. 1.
8.5 Voltage ratings for loads
A confirmation that each component powered by de power systems can operate \Vithout damage over the
system voltage window (e.g., from equalize to the final end-of-discharge voltage) at the location of its input
terminals should be done.
Voltage drop due to large loads, such as motor starting and charging capacitor inrush, may result in a 1nomentary
voltage dip between the terminals of the battery and the terminals of the de loads, and should be addressed.
Ensure that cables are sized to reduce excessive voltage drop during transitory load conditions.
External transient events such as lightning and transmission faults affecting the ground grid should be
considered as applicable; control cable shielding, surge protective devices and filters may need to be
e1nployed. Standards such as IEEE Std C37.90. l [B23] should be consulted for 1nore information about
external transients.
8.5.1 Power generation
Equipment specifications for components po\vered by de power systems should require the equipment to
operate as designed and without damage over the input terminal voltage window corresponding to the variation
in system voltage. For designs in \vhich the batte1y is equalized while connected to the load, this range should
38
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
cover the variation from equalization to the final end-of-discharge voltage (e.g., from 140 V de to 105 V de
for a nominal 125 V de power system, or from 280 V to 2 10 V de for a nominal 250 V de power system). This
is representative of the system operating voltage window. Table 2 provides the recommended voltage range
of some (typical) de powered components for those designs in which the batte1y is equalized while connected
to the load. Note that the de voltage ratings of components may not have a plus or minus 10% tolerance that is
typical ofac rated components.
Table 2-Recommended voltage range of 125 V de and 250 V de (nominally rated
components for designs in which the battery is equalized while connected to the load ;
voltages do not include transient events)
Voltage range
Component
125 V de (nominal)
250 V de (nominal)
Circuit-breaker trip coil
90- 140
70- 140
180- 280
140- 280
Motor-starter coil
90- 140
Solenoid valve
90- 140
90- 140
100- 140
100-140
180- 280
180- 280
Circu it-breaker close coil
Valve-operator n1otor
Auxiliary motor
Electro1nechanical relay coil
lnstru1nentation including protection relays
100- 140
100-140
Indication larnp
100-140
Solid-state relay
180- 280
200- 280
200-280
200- 280
200-280
200- 280
For ungrounded de po"ver systems, external transients, such as lightning and trans111ission faults affecting the
ground grid, may result in significant voltage increases, line to chassis.
In addition to ampacity considerations, supply cables for de-powered components should be sized to provide
adequate voltage for proper operation during the individual component's worst-case operating condition. The
\VOrst-case condition for a constant po\ver load, such as an inverter, may occur at a reduced battery-terminal
voltage in \Vhich case the load current increases. A de valve-operator motor 111ay have a locked-rotor (starting)
cu1Tent of four to ten times the rated full-load current; therefore, voltage drop during the starting of the valve
operator motor is typically the "vorst-case condition. In addition, the voltage drops in four (rather than two)
feeder conductors (from starter to 111otor) should be included in the total voltage drop, due to the necessity
of s"vitching the series field "vhen reversing the valve motor. For small horsepo\ver motors, the voltage drop
across the thermal-overload relay element may be significant and should be considered in the cable sizing.
8.5.2 Substation
Same as po\ver generation. Additional guidance on cables and de po"ver system can be found in IEEE Std 525
and IEEE Std 1818.
8.5.3 Telecommunication
Equipment specifications for co111ponents po"vered by de po\ver systems should require the equipment to
operate, as designed and without damage, over the input terminal voltage range corresponding to ANSI/ATIS
0600315.2013 [B2] andANSI/ATIS 0600315.01.2015 [B3].
8.6 Qualification
Distribution equipment should be qualified for the application.
39
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
8.6.1 Power generation
For Class 1E nuclear applications, distribution equipment shall be qualified in accordance \Vith IEEE Std 649
(as applicable). Cable, field splices, and connections shall be qualified in accordance \Vith IEEE Std 383 (89).
8.6.2 Telecom
Telecom distribution systems shall be designed to meet NEBS compliance.
9. DC power system instrumentation, controls, and alarms
9.1 General
Control devices, instrumentation, and alarms should be provided in order to both enable monitoring and
control of the de po"ver system (e.g., voltages, currents, OCPD operation, temperatures), and to annunciate
during abnormal conditions.
Figure 6 is a one-line diagram showing the recommended instrumentation and alarms for the 125 V de power
system shown in Figure 2 and Figure 3. The recommended instruments, controls, alarms, and their locations
are described in Table 3.
AC1
AC2
I)
I)
Charger 1
Charger 2
Rectifier
Rectifier
Battery
+
Battery
Discharge Test
r)
Breaker..l )
Oµen
Alarm
J_Tesl breaker
Closed
Alar1l'I
, -,
~----<>----------------------
,-,
'
OpenT
Alam1
Breaker ..l )
Open
Alarm T
T
T
Breake'.J_
"'
Blocking
Diode
)
Breaker _l ,.· 1
Open
Alarn1
T'
'~
--
'
'
Oplional
~
Cro~~
Tie
Alternate
~
Breaker..l (
Open
Alarm
Battery ..l
Closed T
Alam1
T
A: Ammeter
v Vollme1er
GV: Ground test voltrneter
ACF: AC fail alarm
HV: High volts alarm
LV: Low voUs alarm
GF Ground fault
RF: Rectifier fail
ST: circuit breaker shunt Tnp
Ii Ii Ii
To Loads
rv1ain Distrut1on Bus
Figure 6-125 V de power system instrumentation and alarms
40
I)
•
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Appl ications
Table 3-lnstruments, controls, and alarms
Location
Instrument/alarm/control
Main control
room(MCR)
Local
Battery current (an1111eter, charge/discharge)
x·
Battery charger output current (ammeter)
x
x
DC bus voltage (voltn1eter)
x
Battery charger output voltage (voltmeter)
x
Ground detector
x
DC bus undervoltage alarm
Xb
DC po\ver systen1 ground alarn1
X<
Battery breaker/s\vitch open alarm
Xb
Battery-charger output-breaker open alarn1
Xb
Battery-charger de output failure alarm
Xb
Cross-tie breaker closed alam1
Battery-charger ac po\ver failure alarm
X'
Xb
Charger lo\v de voltage alann
Xb
x
Charger high de voltage shutdo•vn relay (opens
main ac supply breaker to the charger)
Battery test breaker closed alann
X'
•Hall-effect instrun1ent, or a jack (connected to the battery an1n1eter shunt) for use \v ith a portable test instru1nent
(microvoltmeter) n1ay be provided to read battery charging current, and thus detenn ine the state of charge as described in
IEEE Std 450. Other n1ethods for 1neasuring float current n1ay be used.
"These alam1s should not be con1bined \vith others.
<These alam1s 111ay be tied to a con1n1on alam1. Co1n1non alann n1ay not reflect a less i1nportant event and has to be resolved.
To confirm proper battery voltage, it is i1nportant to rnonitor the voltage specifically at the battery tenninals,
rather than at the output terminals of the charger or throughout the de distribution system.
Additional instru1nentation rnay be considered to detennine the battery capability for trending purposes.
Refer to applicable IEEE standards (e.g., IEEE Std 450, IEEE Std 1106, IEEE Std 1188) to identify which
parameters to monitor (e.g., float cun·ent, te1nperature). For details on how to rnonitor these pararneters, refer
to IEEE Std 1491.
9.1.1 Power generation
Cybersecurity concerns and access control rnust be considered when detennining whether or not to e1nploy
rernote operation of de power system components. Figure 6 shows a typical 125 V de power systern
instrurnentation and ala1ms.
The controls for the battery should be principally located in the battery area. All switching associated with the
battery system should be performed at the local equip1nent. Careful considerations should be made when using
rernote controls.
It is recom1nended that in new de power systerns all alanns should be individually reported.
Additional instru1nentation (e.g., battery monitoring syste1ns, float cu1Tent monitoring, AH 1netering) may be
incorporated into the de power system design as needed to i1nprove its reliability.
41
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
9.1.2 Substation
Cybersecurity concerns and access control must be considered \Vhen determining \Vhether or not to employ
remote operation of de power system components. As a minimum, the follo"ving alarms and readings should
be remotely monitored: de bus voltage, de bus low-voltage alarm, battery current (float, charge/discharge),
breaker closed alarm for test/service battery (if appl icable), and ground detection.
The following alarms and readings may be remotely monitored: charger output current, de bus voltage, charger
output voltage, battery OCPD open alarm, charger OCPDs open alarm, charger de output failure alarm, crosstie breaker closed alarm (if applicable), charger ac power failure alarm, charger low de voltage alarm, charger
high de voltage shutdown relay (opens main ac supply breaker to the charger), and ambient temperature.
Transmission entities might have additional requirements imposed by transmission control agencies or other
governmental entities, such as NERC in the US.
9.1.3 Telecommunication
The typical controller used in telecommunications de plants is a microprocessor device that governs the power
plant and provides alarm, monitor, and control functionality. Some controllers also provide features such as
thermal management to protect VRLA battery cells from thermal runaway. The controller uses Form C contacts
for local audio/visual annunciators as well as points for local site monitoring equipment. Many controllers also
provide graph ical user interface (GUI) and SNMP access for remote monitoring through a network connection.
Figure 7 shows a typical de plant controller, wh ile Table 4 shows typical telecom instruments and alarms.
I
Alpha Numeric Display
0
0
0
0
0
0
0
0
Panel
Controls
Status LEDs
D
Controller
I
-
J
Form C to Aud I Visual Annunciators
-
J
Form C to Scan Points (Remote Mon)
<' T
) Telephone RJ11 (if equipped)
(~ H
('
G~
<:
) Networking RJ45
(;
('
·..:
Laptop
c omputer
c onnection
( Maintenance)
'
)
\....___ _ _ _____,/
Wiring to DC Plant
and Ancillary Systems
Figure 7-DC plant controller (typical)
Table 4 lists typical instruments and alarms.
42
IEEE Std 946-2020
IEEE Recommended Practice for the Desig n of DC Power Systems for Stationary Appl ications
Table 4-lnstruments and alarms
Location
Instrument/alarm/control
Controller local
Remote
Individua l rectifier output current
x
x
x
x
x
x
Thermal management (VRLA)
x
Load discharge CLUTent (an1n1eter, discharge)
DC bus voltage (voltmeter)
Load sharing/energy 1nanagen1ent
x
x
Forni C Contacts
x
Critical alarm/major alarm
x
x
Minoralam1
Battery on discharge alam1 (BOD)
x
x
x
x
x
x
LOV\'-voltage alarm (may be one or l\vo voltage levels)
x
x
Distribution alarm
High-vo ltage shutdo\vn alann (HVSD)
x
x
x
x
Rectifier fail alann (RFA)/Jnultiple rectifier alam1 (MRFA)
x
More than one rectifier fail (critical)
x
x
x
x
x
x
x
x
x
Feeder drain monitoring (overcurrent protection > 100 A [typical])
Alanns
AC fail/rnultiple ac fail alarm
Discharge fuse alann (DFA)
High-voltage alarm (HV)
Alarm battery supply fuse fail (ABS) (if app licable)
Most ala1ms in a telecom1nunications de plant are organized into two master alam1s: MAJOR and MINOR,
which have their own Fonn C contacts. This does not preclude any alam1 from having its o>vn Fom1 C contact.
9.2 DC power system ground detection
Depending on the application design and requirements, de power systems can be designed to operate as
grounded or ungrounded (floating) systems.
Low-voltage cominunications and back-up generator circuits usually have one pole grounded and do not
require ground detection. Refer to A1u1cx E for discussion about the effect of unintentional grounds on the
operation of de power system.
9.2.1 Power generation
In typical power generation, a de power syste1n is designed as an ungrounded system instead of a grounded
system so that a low-resistance ground fault on one of its two polari ties cannot affect the operation of the
system, thus increasing system reliability and continuity of service. Negatively grounded systerns are used
in older generation wireless co1nmunication. Positive and negative gro unded systerns do not require ground
detection circuits.
A gro unded syste1n design may be used if there is a desire to isolate low-resistance gro und faults by means of
protective devices.
While protected for single line-to-gro und faults, ungrounded systerns are sensitive to relative changes in the
grounding plane as a result of sto1m activity or transmission line faults. Also, when a ground occurs, the
voltages to ground in the system adjust and the capacitive charges redistribute. Sensitive relays have been
43
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
known to energize momentarily, while the cable and capacitive charge to ground shifts. Electronic loads,
such as inverters and protective relays with instantaneous undervoltage trips, have operated erroneously.
Sufficient redundancy and confirmation signals should be used in any logic which is dependent on the voltage
of the de pov;er system. Many relays with de coils have lo"v dropout voltages and are not affected by the
momentary line voltage dips. Ground detection should be provided for an ungrounded de power system. It is
recommended that ground fault resistance magn itude also be mon itored so as to lessen the likelihood of a lowresistance (polarity-to-polarity) fault caused by multiple grounds occurring and affecting the operation of the
de load(s). A symmetrical deterioration occurs when the insulation resistance of all conductors in the affected
system decreases at a s imilar rate. An asymmetrical deterioration occurs \Vhen the insulation resistance of one
conductor decreases substantially more than that of another conductor. Insulation deterioration may lead to a
leakage current to ground.
Symmetrical and asymmetrical deterioration of insulation shall be checked. Multiple high resistive grounds
are frequently present on a distribution system. A ground detection system that actively and continuously
evaluates both the positive and the negative leg of the de power system is preferred.
Figure 8 shO\VS a balanced ground detection design for an ungrounded system. A galvanometer or milliammeter provides indication and recording capability. The ground detector should provide a high polarity-toground resistance so that a s ingle ground occurring on the system can not affect the operation of that system.
Consideration should be given to the individual load (device) characteristics to determine the magn itude of
ground resistance that could initiate operation of normally de-energized loads or inhibit drop-out of normally
energized loads. A conservative approach to determine the ground detector alarm setpoint for a de power
system is to measure the normal ground leakage current of the system and set the ground detector to alarm
at that value plus a margin to be determined by engineering judgment. This approach should result in a very
sensitive ground detection system that alarms on a high resistance ground. A suitable, less sensitive ground
detector alarm setpoint may be determined by the method provided in Annex E.
GF alarm circuit
R1
DC
Alternate
SYSTEM
..-- . /
•
1----','
mA
•
~,r---'>--.~~~--1
...... -- :
~
...'
•••
-··
••
•••
R3 .............·
••
• ••
• • ••
••
R2
NOTE- Resistance val ue 111ay change \vith different app lication.
Figure 8-Typical ground detection for an ungrounded de power system
44
GF
alarm
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
NOTE-High resistance to ground used in son1e ground detector designs i1nplies nlinin1al and balanced current leakage
to ground, so the floating de po,ver syste1n is not l OOo/o floating. Strong consideration 1night be needed \vhen 1nultiple
detectors are used (e.g., in a redundant configuration, one in each parallel charger and one on the con1n1on de bus) to
reduce nuisance alam1s or sporadic systen1 tripping since all resistor nel\vorks \viii be connected in parallel resulting in an
increased leakage current to ground.
9.2.2 Substation
Same as power generation.
9.2.3 Telecom
Traditional telecommunication systerns are positive grounded, e.g., -48 V de. Negative grounded systerns
are so1netimes used in wireless telecom1nunication systems, e.g., +24 V de. Neither positive nor negative
grounded systerns require ground detection circuits.
9.3 DC bus undervoltage alarm
The function of the de bus undervoltage alann is to alert the operator that the battery is being discharged. The
de bus undervoltage relay should be adjustable and set to alann at a voltage slightly less than the open circuit
voltage of the battery (e.g., approxirnately 119 V for a 58-cell, 125 V battery or 49 V for 24 cell, 48 V battery)
rather than at the mini1num allowable systern voltage (typically 105 V for a 125 V system).
This undervoltage value setting alerts the operator whenever the battery is supplying energy to the de bus load
(e.g., rnore load than the charger can handle), sufficiently early to take appropriate corrective action.
It is also recom1nended that this ala1m is equipped with a time delay (e.g., 1 s) to reduce nuisance alanns
caused by a sudden voltage drop following an inrush on the de bus.
9.4 Special loading considerations
The following equipment characteristics and systern design features should be given consideration when
sizing and selecting equipment.
9.4.1 Load transfers
Ifthe de power systern design is such that a load group can be auto1natically or manually transferred to another
de source during equalizing, testing, etc., that source (battery, charger, and distribution equipment) should be
sized to supply both (original plus transferred) load groups. This would be typical of a systern with a cross-tie
feature.
When load transfe1Ting fro1n one system to another, the voltage window of both rnust be compatible.
9.4.2 Constant-power de loads
Many de loads are constant power, especially loads that power electronics such as computer servers, inverters,
and de-de conve1ters. Constant power loads draw more current as the battery voltage decreases. For syste1ns
where 1nost of the load is constant power, as rnany telecornmunications loads are, constant power rating tables
should be used when sizing the battery. If the de load is a mixture of constant power and cun·ent loads, using
constant power rating tables 1nay result in a significant oversizing of the battery. In these instances, it rnay
be 1nore realistic to conve1t the constant power loads into constant current values by using a voltage value
that approximates the average voltage during the duty cycle. If the average voltage is unknown, then we
recommend using open circuit voltage of the fully charged battery.
45
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
9.5 Design features to assist in battery testing
The design features of each de power systen1 should provide an effective and safe nleans to periodically
perfonn a capacity discharge test on each battery. Considerations should also be made to feed the de load by
another de source, such as a parallel battery pem1anently connected to the de bus or a dispatch service battery.
In the latter case, a service-battery s"vitching device, such as a nom1ally open circuit breaker or a disconnectfused switch 1nay be required. Figure 3 provides one configuration that allows the battery to be isolated fron1
the de power systen1 for testing via a dedicated test circuit breaker in the associated distribution panel. This
test breaker should be maintained in the open position during all 1nodes ofsysten1 operation, except during the
batte1y test, with an ala1n1 in the 1nain control room when this breaker is closed. Cables fro1n this test circuit
breaker should be routed to a convenient location and terminated so that connection to a discharge load bank
is convenient. This nlethod (or similar nleans) for a safe connection often1porary discharge test cables should
be provided.
The design can also provide 1neans for partial battery discharge test, "vhere the rectifier output voltage is
lo"vered below open circuit, but the rectifiers are not shut off to reduce the possibility that a battery failure
would cause a catastrophic load failure.
9.6 Cross-tie between buses
Cross-ties between de distribution buses may be utilized to supply critical loads when a battery or charger is
taken out of service for maintenance or testing. Cross-ties can also provide beneficial switching flexibility
during such situations and can aid in acco1nplishing orderly plant shutdowns. A cross-tie to any independent
de power system is acceptable provided that the independent de po"ver system nleets the sizing requirements
of 6.2. One acceptable design provides a 1nanually operated circuit breaker at each end of the cross-tie. The
cross-tie circuit breakers should be normally open and should activate an alarm in the main control roon1 if
either is closed. Operating procedures should clearly define the operation of the cross-tie breakers. If crosstie operation results in paralleling two batteries, the duration of parallel operation should be lin1ited to the
time required for s"vitching so as to reduce the impact that circulating currents may have on battery capacity.
Increased available short-circuit current resulting fron1 the parallel sources should also be considered.
WARNING
OCPDs including circuit breakers and fuses, de panels, and related equipment must be sized to withstand
the momentary surge current flowing when two de po\ver systems are paralleled.
9.6.1 Power generation
For Class IE nuclear applications, a cross-tie to any independent de power system, other than the de po\ver
system in the redundant safety division, is acceptable only during cold shutdo\vn or refueling modes, and only
if it can be shown that the cross-tie does not impair the ability of the Class IE de po"ver system to perform its
safety function. In multi-unit nuclear stations, Class lE de po"ver syste111s shall not be shared bet>veen units
unless it can be sho"vn that such sharing does not impair their ability to perform their safety functions.
10. Protection against electrical noise, lightning, and switching surges
Electrical noises, s"vitching surges, and lightning effects should be adequately addressed for a reliable de po\ver
system and its connected load operation. These disturbances are generally caused by lightning, inverters, dcdc converters, inductive loads S\Vitching, arcing, radio equipment, grid transients, voltage sags from large
connected loads, equipment failure, relay actuation, cable crosstalk, etc. If not suppressed or filtered, they may
impress voltage spikes on the de po\ver system of magnitudes that are dangerous for connected equipment,
and in addition to damages, they can also cause reset and disturb sensitive loads.
46
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
Precautionary techniques should be used to eliminate, reduce, and to damp out the noise and surge a1nplitude
with surge suppressors and/or low and band pass filter circuits. Segregation of power and signal cables can
also be deployed to help assuring better immunity.
In ungrounded de power systems, suppressing and filtering circuits should not establish additional conductive
paths, in the power frequency range, to ground.
10.1 Electromagnetic interference (EMI), radio-frequency interference (RFI)
EMI and RFI are propagated by conducted and radiated effects. Adequate equipment filtering should be
considered to help assuring necessary immunity to disturbances such as:
Radiated noise, radio-frequency, electromagnetic field
Conducted disturbances induced by radio-frequency fields
Conducted and common mode disturbances
For guidance refer to EPRI TR-I 02323 [B6], IEC 61000-4 series [8 7], FCC, and ATIS 0600013 [85].
10.2 Lightning and switching surges
For applications where the power and control cables connected to the de system are exposed to transient
voltages, considerations should be made to protect electrical equipment against their damaging effects.
Examples of typical disturbances are as follows:
Lightning
Voltage surges, high-frequency disturbances
Oscillatory and fast transients, and bursts generated by inductive loads and power equipment switching
Electrostatic discharges
Damped oscillatory magnetic fields
Voltage dips, short interruptions, and voltage variations
Adequate surge protective devices (SPD) networks [e.g., 1netal oxide varistors (MOVs), avalanche diodes,
spark gap protection devices, etc.] should be installed at the distribution panel level and at each connected
equipment level. In addition to SPDs, t\visted and shielded cables may also be needed for signaling and control
circuits.
For protected indoor installations, equipment should typically be able to withstand 2 kV to ground momentarily
without damage, while for outdoor equipment, the level is typically 4 kV.
Depending on the site conditions and needed protection, refer to standards such as IEEE C62 family on surge
protection (e.g., IEEE Std C62.41.I [8 24], IEEE Std C62.41.2 [825]), IEEE Std C37.90.l [8 23], IEEE Std
525, IEC 61000-4 series [87], etc.
47
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
11. Spare equipment
The need for spare equipment/components depends on system design features, as \Vell as the criticality of
the system. Other factors, such as operating experience, the availability of replacement equipment/parts,
manufacturing lead time, the capabilities of pe1forming in-house repairs, and component failure rates should
be given consideration in determining the specific components that should be maintained as spare parts. For
example, if the system design provides a readily available backup battery charger, the need for maintaining
spare charger components could be reduced and possibly eliminated.
Other considerations, such as shelf life, 1nay make it undesirable to obtain spares at an early stage in plant life.
For exa1nple, spare battery containers can typically be maintained for only one year if they are of the dry type
(or three to six months ifv;et type) and thereafter should be charged and 1naintained the sa1ne as a battery in
service. With an adequate design margin, the battery may be able to fulfill the service require1nents \Vith one or
more cells electrically removed. The quantity of batteries provided and the capability to utilize backup battery
capacity through the use of cross-tie circuits for example, may also be factors in determining the need and
urgency for obtaining spare cells.
In any case, an evaluation should be performed to determine the need for spare equipment based on the
co1nbination of system design features, operating requirements, etc.
48
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
Annex A
(informative)
Bibliography
Bibliographical references are resources that provide additional or helpful material but do not need to be
understood or used to implement this standard. Reference to these resources is made for informational use
only.
[BI] ANSI/ ATIS 0600329.2008: Net\vork Equipment- Earthquake Resistance. 8
[B2] ANSVATlS 0600315.2013: Voltage Levels for DC- Po\vered Equipment Used in the Telecommunications
Environment.
[B3] ANSVATIS 0600315.01.2015: Voltage Levels for 380V DC-Po\vered Equipment Used 1n the
Telecommunications Environment.
[B4] Accredited Standards Committee C2-2017, National Electrical Safety Code® (NESC®). 9
[BS] ATIS 0600013: Electro1nagnetic Compatibility (EMC) And Electrical Protection used for
telecommunication equipment.
[B6] EPRI TR-102323, Guidelines for Electromagnetic Interference Testing of Po\ver Plant Equipment. 10
[B7] IEC 61000-4 (all parts) Electro1nagnetic Compatibility Testing and Measurement series of publications. ''
[B8] IEEE Std 336TM, IEEE Recommended Practice for Installation, Inspection, and Testing for Class IE
Power, Instnunentation, and Control Equipment at Nuclear Facilities. 12•13
[B9] IEEE Std 383TM, IEEE Standard for Qualifying Electric Cables and Splices for Nuclear Facilities.
[BIO] IEEE Std 384TM, IEEE Standard Criteria for Independence of Class IE Equipment and Circuits.
[B 11] IEEE Std 446TM, IEEE Recommended Practice for Emergency and Standby Power Systems for
Industrial and Commercial Applications (The Orange BookTM).
[B 12] IEEE Std 535TM, IEEE Standard for Qualification of Class IE Vented Lead Acid Storage Batteries for
Nuclear Po\ver Generating Stations.
[B 13] IEEE Std 603TM, IEEE Standard Criteria for Safety Systems for Nuclear Po\ver Generating Stations.
[B 14] IEEE Std 627TM, IEEE Standard for Qualification of Equipment Used in Nuclear Facilities.
[B 15] IEEE Std 666TM, IEEE Design Guide for Electric Power Service Systems for Generating Stations.
8ATIS
publications are availab le from the Alliance lor Telecommunications Industry Solutions (https://ww\v.atis.org/).
is avai lable from the Institute of Electrical and Electronics Engineers (https://standards. ieee.org/).
10
EPRI pub lications are avai lable from the Electric Power Research Institute (https://w\vw.epri .com).
" IEC publications are available from the International Electrotechnical Commission (https://wW\v.iec.ch) and the American Nationa l
Standards Institute (https://www.ansi.org/).
'Z'fhe IEEE standards or products referred to in this annex are trademarks owned by The Institute of Electrical and Electronics Engineers,
Incorporated.
i;IEEE pub lications are available from The Institute of Electrical and Electronics Engineers (https://standards.ieee.o rg/).
9The NESC
49
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
[B 16] IEEE Std 690TM, IEEE Standard for Design and Installation of Cable Systems for Class IE Circuits in
Nuclear Power Generating Stations.
[B l7] IEEE Std 74JTM, IEEE Standard for Criteria for the Protection of Class IE Po\ver Systems and
Equipment in Nuclear Power Generating Stations.
(B 18) TEEE Std 1184™, IEEE Guide for Batteries for Un interruptible Power Supply Systems.
(B 19) TEEE Std 1202TM, IEEE Standard for Flame-Propagation Testing of Wire and Cable.
(B20] TEEE Std 188 1TM, IEEE Standard Glossary of Stationary Battery Terminology.
(B2 I] TEEE Std C37. l 6TM, IEEE Standard for Preferred Ratings, Related Requirements, and Application
Recommendations for Low-Voltage AC (635 V and below) and DC (3200 V and belo"v) Po\ver C ircuit
Breakers.
(B22] IEEE Std C37.90TM, TEEE Standard for Relays and Relay Systems Associated \Vith Electric Power
Apparatus.
(B23) fEEE Std C37.90. 1TM, IEEE Standard for Surge Withstand Capability (SWC) Tests for Relays and
Relay Systems Associated with Electric Power Apparatus.
(B24] IEEE Std C62.41.1 TM, IEEE Guide on the Surge Environment in Lo\v-Voltage ( I 000 V and less) AC
Power Circuits.
(B25] IEEE Std C62.4 l .2TM, IEEE Recommended Practice on Characterization of Surges in Low-Voltage
( 1000 V and less) AC Power Circuits.
(B26] NEMA MG 1, Motors and Generators.
(B27] NEMA PE 1, Uninterruptible Power Systems.
(B28] NEMA PE 7, Communication Type Battery Chargers.
(B29] Stationary Battery Short-Circuit Test Report, AETTest No. 059 1- 1, May 1991, Alber Engineering, Inc.
(B30) Stationary Battery Short-Circuit Test Report, ATT Test No. 0792-1 , July I 992, Alber Technologies, Tnc.
[B3 l] TL9000: Quality Management System (QMS).
[B32] US NRC NUREG/CR-7229: Testing to Evaluate Battery and Battery Charger Sho1t -Circuit Current
Contributions to a Fault on the DC Distribution System.
50
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
Annex B
(informative)
Battery charger sizing-Sample calculations
B.1 General
This annex outlines the method, including sample calculations, for determining the required rating of battery
chargers.
B.2 Equation
The facility battery charging equipment should be sized in accordance with the following equation:
(B. l)
Refer to 7.3 for the equation and variable definition.
B.3 Example 1: Power generation
Determine the rating of the charger required for a battery \Vhere the continuous de load is 100 A, the largest
combination of non-continuous loads is 80 A; ampere-hours discharged from the VLA battery is 400 Ah "vith
12 h to recharge the battery (no abnormal service conditions) at an altitude of 1500 m (5000 ft) above sea level,
and 10% future load gro\vth:
(B.2)
Altitude de-rating factor value at 1500 m provided by the manufacturer is 1.07.
K = altitude factor x future factor: K = I . 07 x I . I 0 = I . 18
(B.3)
(B.4)
To account for the largest con1bination of non-continuous loads, the follo\ving applies:
12 = l.18x(l00+80) = 212.4A
(B.5)
Recommended charger rated outputl3 :
(B.6)
51
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
B.4 Example 2: Substation
Simple calculation for an existing 150 Ah VLA battery with 12 h recharge and 25 A of constant load including
the future growth. The charger is installed at I 00 m (333 ft) above sea level. The load profile is not known.
I'"= I (
150 x l.l
)
+25
12
=
38.8 A: use40Aoutputcharger
(B.7)
Caution
Refer to 7 .3 for special considerations.
B.5 Example 3: Telecommunication
Some telecommunication rectifier sizes are given in nominal rated de output amperes (e.g., 50 A, 200 A) for
typical non1inal 48 V de or 24 V de rectifiers. While some of these rectifiers may put out slightly more current
(if the current limit is set to allow it) than the nominal ampere rating, the non1inal rated output number is used
for rectifier sizing calculations.
More and more telecomn1unication rectifiers are no"v rated in constant-po\ver "vatts, \Vhich means that their
output varies by voltage. For sizing purposes, this number may be converted to an1peres at the expected battery
float voltage. For example, a 3500 W rectifier floating 24 series-connected 1.215 s.g. cells at an average of
2.20 V/cell, "vould be an approximately 66.3 A rectifier (3500 W/52.8 V). Or, the calculations nlay be done
\Vholly in \Vatts (see examples belo"v).
While individual telecomn1unication equipment shelf loads may vary more \Videly, overall load on a modern
telecommunication de po\ver plant is relatively constant throughout the day and year, unless end-use
equipment is added or removed. This load may vary slightly, but at nominal float voltage, the average plant
current during the busiest hour of the year is typically described as an average List I load (Telcordia GR-513
gives nlore exact definitions of List 1 drains for overall de plants and individual pieces of telecommunication
equipment). List 1 drain (which can be sin1ply described as a peak average \Vell belo"v the worst-case peak)
is used to size rectifiers and batteries in teleconlmunications de plants. Because most of the loads are constant
po\ver, and battery voltage drops during a full discharge by up to about 20o/o, GR-513 recommends that the
total rectifier capacity be at least 20% greater than the List 1 load (most telecommunication net"vork operators
specify a minimum rectifier sizing guideline of20% to 40% excess capacity). In addition to helping to ensure
that the rectifiers can handle recharge \Vhen the batteries have been drained to the minimum operating voltage,
this excess capacity sometimes helps to verify that the GR-513 requirement to recharge the batteries to 80% of
capacity \Vithin 24 h is met (returning the first 80% of capacity is a near 100% efficient operation-returning
the last 20% becomes less and less efficient). Ho\vever, calculations must be done to prove this, and if not, then
additional rectifiers may be needed.
As noted in above in this document, and per Telcordia GR-513 and FCC Best Practices, multiple rectifiers in
parallel are commonly used in telecommunications, designed "vith n+ 1 redundancy so that if one rectifier fails,
the load is not partially on the battery. Some telecommunications companies even require n+2 redundancy in
certain types of sites (such as the most critical ones, or the ones that are hardest to reach in certain climatic
conditions or furthest a"vay), or in all sites. Note that the "redundant" rectifier(s) is really "vhichever rectifier
fails first, as all installed rectifiers are typically set up to share the load fairly equally. If one rectifier is lost, the
output current of all of the remaining rectifiers that are turned on and properly set "vill increase to cover the
capacity that was being carried by the rectifier that failed.
52
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
Depending on the recharge time desired, typical rectifier sizing ranges from 120% to l 40o/o minimum of the
typical relatively constant current load. Both rules (n+ I, and recharge) must be satisfied.
T he paralleled chargers in a modern telecommunication system are connected to a com1non microprocessor
controller that has the ability to shut down and restart the rectifiers. The controller should shut down the
rectifier(s) in case of a high voltage (kno\vn as high-voltage shutdown [HVSD]). Modern rect ifiers used in
charger systems also have their own HVSD protection scheme. Some have a max imum value set by hardware
in the unit as well as a software threshold that is able to be configured by the charger system controller. The
controller may turn off and restart individual rectifiers in order to help ensure that the working rectifiers are
operating near the highest point on the ir energy efficiency curve (this is typically called energy management).
The controller also has the ab i1ity to control the rectifier output voltages (usually for temperature-compensated
charging of YR LA batteries based on input from one or more battery temperature sensors, or to tum down the
voltage to a level that allows a partial discharge test of the batteries).
Illustrating telecommunication rectifier sizing is probably best done \Vith an example. A typical
telecommun ication central office-48 V de po"ver plant might be designed with the following inputs/standards:
I 0 500 W nom inal -48 V rectifiers
Plant float voltage of-52.80 V
Float busy-day peak average (List 1 dra in) load of2400 A (this includes future growth over the relevant
planning horizon)
M inimum recharge sizing rule of 20o/o extra capacity above and beyond List 1 for this particular
telecommun ication company
N+ 1 rectifier s izing rule for th is particular telecommunication company
80% recharge w ith in 24 h
12 parallel strings of flooded 1680 Ah cells designed for an approx imate backup time of 4 h
The4 h rate of these particular cells to the designed minimum voltage per cell (m Vpc) is 295 A, making
their capacity at the 4 h rate 1180 Ah
Given these inputs, the m inimum number of rect ifiers can no"v be designed.
First, the de current per rectifier for sizing purposes must be determined:
l 0 500 W ~
A
199
52.8 v
(B.8)
To barely meet the future planning horizon List I load, 13 rectifiers are required (decimals must be rounded up
when sizing rectifiers):
2400 A =
12 06
199 A
.
(B.9)
Due to the N+ 1 reliability r ule, this means that 14 rectifiers are required. This 1neans that there is 2786 A of
total rectifier capacity:
(B. l 0)
14 x 199 A = 2 786 A
53
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
This total rectifier capacity must now be compared against the 20% excess recharge capacity require1nent and
the 80% recharge w ithin 24 h rules. In order to meet the recharge capacity n1les, the projected List l drain is
multiplied by the excess capacity factor:
2 400 A x I . 2 = 2 880 A
(B.11)
As can be seen, the minimum recharge capacity rule is going to requ ire 15 rectifiers:
2 880 A =
14 47
199 A
.
(B.12)
This means that the total rectifier capacity is now:
15 x 199 A = 2 985 A
(B.13)
T he spare capacity available for recharge is:
2 985 A - 2 400 A = 585 A
(B.14)
This excess capacity needs to return 80% of the capacity of a fully discharged battery plant within 24 h per the
1u les of this particular telecommunication company. The capacity of the battery plant is:
12 strings x I 180 Ah = 14 160 Ah
(B.15)
80o/o of this capacity is:
14 160 Ah x 80% = 11 328 Ah
(B.16)
T he number of hours required to return 80% of the capacity of a fully discharged batte1y is:
11 328 Ah
.
A ::::: 19 hand 20 n11n
585
(B.17)
This is less than the maximum 24 h requirement, so no additional rectifiers are needed.
Because the modern telecommunications loads are primarily constant power (as has been described previously
in th is document) and the rectifiers themselves are rated in \Vatts, all of the calculations from the preceding
example can be done in watts.
The first step is to convert the List I drain to watts. While technically, L ist I drains are calculated by equipment
manufacturers at 52.0 V, it is more accurate with actual loads to run the calculations at the actual float voltage.
Assuming constant power loads, the L ist I load in \Vatts is:
2 400 A
x
52 . 80 V = 126 700 W
(B.18)
To barely meet the load, 13 rectifiers are required.
126700 w
.
W = 12 . 07 rectifiers
10 500
(B.19)
54
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
Due to then+ I rectifier redundancy/reliability requirement, this means that 14 rectifiers are required. Total
rectifier capacity is thus:
14 x IO 500 W = 147 000 W
(B.20)
This tota l rectifier capacity must then be compared against the 20o/o recharge capacity rule, and the 80%
recharge in 24 h rules. Tn order to meet the recharge capacity rules, the List 1 drain is multiplied by the excess
capacity factor:
126 700 W x I . 2 - 152 000 W
(B.21)
To meet this rule, 15 rectifiers are requ ired:
152000 w
IO 500 W
14 48
.
=
(B.22)
With 15 rectifiers, the total rectifier capacity is now:
15 x 10500
w
= 157500
w
(B.23)
The spare capacity available for recharge is then:
157 500
w-
126 700
w
= 30 800
w
(B.24)
To determine if80% of battery capacity can be returned \Vithin 24 h (the first 80% of lead-acid battery capacity
returned is almost I 00% coulombically efficient). The full capacity of all the batteries is:
12 x I 180 Ah x 52.80 V = 747600 Wh
(B.25)
80% of that capacity is:
747600 Wh x 80% = 598 100 Wh
(B.26)
T he number of hours required to return 80% of the capacity, using the spare rectification is:
598 100 Wh
30 800 w
=
19 4
.
h
(B.27)
This is less than the 24 h required by the specification, and thus no more rectification is needed to nleet this
final rule.
55
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
Annexe
(informative)
Battery available short-circuit current-Sample calculations
C.1 Introduction
The current that a batte1y delivers on sho1t -circuit depends on the total resistance of the sho1t-circuit path.
The batte1y nominal voltage should be used when calculating the maximtun short-circuit current. Tests have
shown that an increase in electrolyte temperature (above 25 °C) or elevated battery terminal voltage (above
nominal voltage) has no appreciable effect on the magnitude ofshort-circuit current delivered by a batte1y (see
Stationa1y Battery Sho1t-Circuit Test Repo1t 059 1- 1 [B29] and Stationary Battery Short-Circuit Test Report
0792- 1 [B30]).
For Ni-Cd cells the short-circuit capability can range between 7 and 50 ti1nes the rated Ah capacity. Refer to
the manufacturer for the Ni-Cd battery sho1t -circuit capability.
Although an increase in temperature results in an increase in the chemical activity of the battery, it also
increases the resistance of the metallic components of the batte1y, thereby offsetting any appreciable change in
the magnitude of short-circuit current the batt.e1y can deliver. However, the elevated temperature results in the
batte1y's capability to deliver the short-circuit current for a longer duration.
Elevated battery terminal voltage (above nominal voltage) during float and equalize charge does not increase
the che1nical energy available from the batte1y during sho1t-circuit. For lead-acid batteries, the effective
voltage driving the sho1t -circuit current is dependent on the acid concentration in direct contact with the active
material in the plates. Therefore, the batte1y nominal voltage (2.00 V per cell for lead-acid and 1.2 V for NiCd) should be used when calculating the maximtun short-circuit current.
The fault current from a large lead-acid batt.e ry resulting from a bolted short at the batte1y terminals typically
exhibits a rate-of-rise that delivers the peak current \Vi thin typically I 0 ms to 50 ms. For the actual time,
consult the battery manufacturer. Refer to references [B27], (B29], and [B30].
C.2 Discussion
The total resistance is made up oft\vo major parts as follows:
The apparent internal resistance of the battery
The external circuit resistance
The total internal resistance of the battery is equal to the sum of the internal resistance of the cells plus the
resistances of the intercell connections. The value of internal cell resistance is a variable quantity that is
significantly influenced by 1nany factors, e.g., the temperature, the age, and the state of charge of the cell. The
total external circuit resistance is the sum of the resistances of the various components, e.g., the connecting
cables and the fault resistance.
The following sa1nple calculations illustrate one method of calculating the internal resistance of any cell
(utilizing manufacturer's published discharge characteristic curves for that cell) and then calculating the
current that cell can deliver:
Through a bolted short-circuit at the cell terminals (i.e., zero external resistance)
Through a short-circuit at the main distribution bus \Vi th a specific (0.0 I) exte1nal resistance and \Vi th
no charger or motor contribution
56
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
C.3 Sample calculations for a lead-acid battery
Sample calculations for a lead-acid battery include the follo\ving:
a)
Internal resistance of a cell can be calculated from the slope of the initial volts line. See Figure C. l ,
which shows a discharge characteristic fan curve for a typical 7 through 15 plate (total) cell. (When
fan curves are not available or other battery technologies are employed, contact the manufacturer
for Rt values for specific cell types and recommended voltages to be used for short-circuit current
calculations.)
Rp
Rt= -
(C. I)
1Vp
\Vhere
Rt
Rp
Np
Rp
=
is the total internal resistance of cell (ohms)
is the resistance per positive plate (ohms)
is the quantity of positive plates
Vi - V2
_I, Q I positive plate
12
(C.2)
Rp is the resistance per positive plate for any two voltage and current points along the line.
If we p ick 1.90 Vas V,, and 1.50 Vas V2 , 11 is found to be 60 A/positive plate and 12 is 370 A/positive
plate. The calculation is given in Equation (C.3):
R _ 1.9-1.5
p - 270-60 -
~ i~
=
0. 00129 Q / positive plate
(C.3)
Assume that the cell being investigated has 15 (total) plates. Since the cell has seven positive plates
(connected in parallel), the total internal resistance is:
Rt
b)
= 0.0~ 129 = 0.000 18 Q
(C.4)
The short-circuit current available at the cell terminals is found from Ohm's law as follows:
Ee
Jc =-Rt
=
2
-0-.0-00_1_8
=
(C.5)
ll lll A
where
Jc
Ee
Rt
c)
is the available short-circuit current (in amperes)
is the nominal cell voltage (2.00 V)
is the total internal resistance of cell (ohms)
The short-circuit current available at the main distribution bus (the load term inals of the 1nain/ battery
circuit breaker) from a battery made up of 58 cells with an average internal resistance of 0.00018 Q
as calculated in Equation (B.5), with 0.0 I 0 Q total external resistance, and \Vi th no charger or motor
contribution, is found from Ohm 's lav; as follov;s:
Eb = 58 cells = 2 Vpc = 116 V
£ 8 is the nominal batte1y voltage
R8 is the total internal resistance ofbatte1y:
Rb = Rt percell*numberofcells = 0.00018*58 = 0.01044 Q
57
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
Rx is the total external circuit resistance (including resistance of main lead cables; intercell [if not
included in R8 ], inter-tier, and inter-rack connections; main circuit breaker or fuse; and the fault)
Rx
= 0.0100 Q
Rr is the total circuit resistance
RT = RB + Rx = 0.0 1044+0.0 100 - 0.02044 Q
IB
Eb
=
Rt =
116
0.02044
=
(C.6)
5675 A
where
IB
is the available short-circuit current (amperes) at load terminals of main/battery circuit breaker
The short-circuit currents calculated in Equation (B.6) may, when combined with the charger and motor
contribution, be used to determine required interrupting capacity of the circuit breakers or fuses.
NOTE- For comparison, I 0 ti1nes the I 1nin a1npere rating (to 1.75 V per cell at 25 °C) of the battery is I 0 x 1139 = 11 390 A.
RP=
v, - V2 12
v,
-
I,
- ~·~g =
1.90 - 1.50
370 - 60
0.00129
~2/POSITIVE PLATE
2.0
7 T HROUGH 15 TOTAL PLAT ES
1.210 SPECIFIC GRAVITY 77 °F
I,
w
1.9
1.8
CONNECTOR DROP INCLUDED
~
_J
1.7
a_
w
1.6
>
1-(j)
15
80
0a_
-
z
( j)
0:::
::J
0
--1
)>
60
r
<
I
0
~
w'
0:::
(/)
w 40
a_
:a:
4:
1 min
0
0
40
80
120
160
200
240
280
AM PERES/ POSITIVE PLAT E
Figure C.1-Typical lead-acid battery fan curve
58
320
360
400
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
Annex D
(informative)
Battery charger and de power system, short-circuit current
contribution
D.1 Introduction
This annex provides a rationale for the selection of the maximtun value of batte1y charger sho1t -circuit
current that occurs coincident with the maximtun battery short-circuit current. The reason for determining
the maximum combined short-circuit current is to specify equipment that is suitable for the expected fau lt
current. It is necessary to include the contributions from connected motors, battery chargers, and batteries
when calculating the total short-circuit current for a fault in a de po\ver system.
D.2 Combined effect of currents from battery, charger, and connected
inductive load
The fault current for a short at the de distribution switchgear or panel board peaks later (typically within
34 ms to 50 ms) due to the inductance of the de pov;er system in series with the fault. The magnitude of the
fault current for a short at the distribution bus will also be lower than the value at the co1nponent due to the
impedance of the cables between this component's terminals and the bus.
Due to the battery time constant, the maximum coincident short-circuit current can be conservatively
calculated as the sum of the peak sho1t-circuit current from the battery and the peak sho1t -circuit current value
from the charger (Figure D. l ).
Inductive loads such as de motors, if operating, \Vil! contribute to the total fault current. The maximtun current
that a de 1notor delivers at its terminals is limited by the effective transient armature resistance Rd of the range
ofO. l to 0. 15 per unit. Thus, the maximum fault current for a fault at the motor terminals typically ranges from
seven to ten times the motor's rated armature current. Therefore, it is conservative to estimate the maximum
current that a motor contributes is ten times the motor's rated full-load current. When a more accurate value is
required, the sho1t -circuit current should be calculated using specific rd data or from test data obtained from
the motor manufacturer. (IEEE Std 446 [B 11] : Emergency and standby systems for industrial and commercial
applications and IEEE Std 666.)
Each installation should be evaluated by the design engineer to determine the magnitude of the sho1t -circuit
currents from the battery, charger, motors, etc. Any non-typical installation should be evaluated by the design
engineer to verify that the peak values of the batte1y and charger short-circuit currents are not coincident.
D.3 Sample evaluation
The follov;ing example illustrates one method for determining the relative fault-current contributions of a
batte1y and a current-limited charger to a fault at the distribution panel bus. (Note that the follo\ving values
may not be typical for any given charger type, design or installation.)
Charger and feeder cables
Charger current rating: I = 300 A
Charger transient current: Jc = I 0 x 300 - 3000 A
59
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
Cable and breaker resistance: Re = 0.0228
Battery and feeder cables
Battery: 60 cells, rated 1950 Ah at 8 h rate.
Battery resistance under fault:
Rb = 0. 000113 I
x
60 cells = 0. 006786
Q
Time to peak short-circuit current @ five-time constants = 11 ms
Battery cables: 2-1 IC 350 kcmil cables/leg, each 12.3 m (40.5 ft)
Total calculated loop resistance: R2 = 0.00 17
Total calculated inductance: L= 36.2 ~tH
The battery ti1ne constant and apparent inductance under short-circuit conditions are calculated as follows:
.
Battery tune constant: T =
11 ms
5
= 2. 2 ms
Battery inductance: lb = T x Rb = 2. 2£ - 3
x
0. 006786 = 14. 9 µH
The time constant for the cornbination of the battery and cables is calculated as follows:
14.9£-6 + 36.2£-6
lb + l
Tl = Rb +R2 0. 006786 + 0. 0017
=
6 ms
Fault at the distribution panel bus
Battery fault current:
2 Vpc*60
120 v
17-) - 14 141 A
lb = Rb +R2 - -(0- .-00_6_7_86_+_0_.0-0-
Batte1y fault current peaks at (5) (Tl) = (5) (6 ms) = 30 ins
The charger short-circuit contribution: le = 3000 A
Conclusion
As illustrated in Figure D. l , the maximum total combined short-circuit current is:
I = 14 141 + 3 000 = I 7 141 A
17 141 A is a maximwn current \vhen the battery current peaks at 30 ms after the fault.
60
(D.l)
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
18000
16000
_
-------~
14000
------------=
----,,
,,,
~----
,,
12000
,,''
,
10000
I
I-
<(
6000 - - I
I
4000
I
·'.·.
~• •
. ·-••
2000
0
I
I
I
~
...
• •
••
••
••• •
•
••
•• ••••
• ••
•• ••
.•··•·•···••··••·•••·••·· ····· • ••
..•
Time (ms)
• • • • • Charger current (A) - - - Battery current (A) -
- Total (A)
Figure D.1- Typical short-circuit current evolution at the de bus connected to a battery
in parallel with current-limited charger
61
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
Annex E
(informative)
Effect of unintentional grounds on the operation of de power
systems
E.1 Introduction
This annex provides guidance in determining: a) the threshold value of ground-fault resistance that may affect
equipment operation if a second ground occurs in an ungrounded de power system and b) a suitable, less
sensitive, ground-detector alarm setpoint.
Note that unintentional grounds in grounded systems (positively or negatively grounded) 1nake themselves
obvious by creating a quite large sho1t -circuit event.
E.2 Discussion
A single low-resistance ground on one of the two polarities of an ungrounded de po\ver system should not
affect the operation of the system. However, a ground of sufficiently lo"v resistance on one polarity follo\ved
by a second ground can produce ground currents of sufficient magnitude to initiate operation of de-energized
de loads (devices) or inhibit dropout of energized de loads (devices). The grounding configuration shown in
pa1t A of Figure E. l illustrates how multiple grounds could initiate operation of a de-energized load. This
condition could result in false operation of a normally de-energized load (device).
(+)
(+)
I
(-)
I
(-)
A
B
Figure E.1-Ground faults may energize a normally de-energized device
or prevent de-energizing a normally energized device
The grounding configuration shown in part 8 of Figure E. l could exist in a normally energized logic circuit,
such as an engineered safety feature or a reactor protection system, and could inhibit the drop-out ofenergized
loads (devices). Figure E.2 shows three ground co1nbinations that could short out an actuating coil and/or the
de source. For low-resistance grounds, a fuse or circuit breaker \Vould trip a circuit designed with overload
and fau lt protection, clearing the faulted circuit and dropping out all energized loads (part 8 of Figure E.2),
or preventing operation of de-energized equipment (pa1t A and pa1t C of Figure E.2) on that circuit. A fau lt
configured as shown in part A or part C of Figure E.2 could inhibit the trip of a breaker for switchgear,
generator, or a large motor. The ground configuration shov;n in part A or part 8 of Figure E.3 could cause
the relay contacts associated with one piece of equipment in one circuit to affect the operation of a similarly
grounded device in another circuit.
62
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
(+)
(+)
(+)
I
(-)
I
(-)
A
(-)
B
I
c
Figure E.2-Ground faults may short out an actuating coil and/or de source
(+)
(+)
I
(-)
I
(-)
A
B
Figure E.3- Ground faults may cause contacts in one circuit to actuate devices
in a different circuit
In order to detennine the threshold resistance ofa ground fault that, if follo>ved by a solid ground, could initiate
operation of a normally de-energized load or could inhibit dropout of a nonnally energized load; the most
sensitive de loads (devices) should be identified and their nlinimu1n pickup current and nlaxin1um dropout
cu1rent should be evaluated.
E.3 Sample evaluations
E.3.1 Example 1 (low-resistance device)
The control devices for a Type XYZ switchgear breaker nip device is rated at 125 V de. The closing coil and
trip coil are as given below.
Assuming a de power syste1n operating (float) voltage of 130 V, the current through the closing coil or n·ip
coil having 20.83 n, in series with a 20 k.Q ground leakage in parallel with the control contacts (so when the
control contact is the open position, the coil will be fed through the 20 k.Q ground leakage resistance) so the
current as per pa1tA of Figure E. l would be:
130 v
20 k.Q + 20. 83 .Q = 6 . 5 mA
(E. l)
Mini1num pickup closing current:
90 v
20. 83 .Q = 4. 32 A
(E.2)
63
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
Mini1num pickup tripping current:
70 v
20. 83 Q = 3. 36 A
(E.3)
(ln this example we assumed that the values of90 V de mini1num pickup closing voltage and 70 V de minimum
pickup tripping are provided by the coil data sheets.)
Since the minimum pickup current for the coils is 4.32 A (closing) and 3.36 A (trip), these values are \Yell
above the value of current (6.5 mA) that \vould be experienced with a 20 kQ total ground-fault resistance.
Therefore, a 20 kQ ground, followed by a solid ground, would not cause spurious operation of the switchgear
breaker. The threshold value of ground-fault resistance is:
130 v
3 .36 A
20.83 Q = 17.86 Q
(E.4)
neglecting the effect of the ground detector resistance.
Assu1ning that this circuit breaker trip coil has been identified as the most sensitive (lowest pickup or dropout
current) device connected to the de power system, the ground detector alarm can be set at any value above
17.86 Q ; the appropriate margin (above the threshold value) being based on engineering j udgment. A setpoint
of20 Q will alarm at a ground fault current (3.18 A) that is 5.4% below the 1ninimum pickup current (3.36 A).
This example is for illustrative purposes only since most ground detection systems detect grounds several
orders of magnitude greater than 20 n.
E.3.2 Example 2 (high-resistance device)
The operating characteristics of a nonnally energized Type XYZ 125 V de relay are as given in the following
paragraph.
Assuming a de power syste1n operating (float) voltage of 130 V, the current passing through the 2 kQ relay coil
and the 20 kQ ground leakage resistance (in parallel with the control contacts) (part B of Figure E. l ), after they
open, would be:
130 v
-(2_0_kQ_+_2_k_Q_) - 5 . 91 1nA
(E.5)
Since the maximtun dropout current for the relay is 6.25 mA, a 20 kQ ground follo\ved by a solid ground
would be of high enough total resistance to produce a low enough ground current that \vould not prevent the
relay fro1n dropping out when the control contacts open. T he threshold value of ground-fau lt resistance is
( 130 V/0.00625 A) - 2 000 Q = 18 800 Q , neglecting the effect of the ground detector resistance. Assuming
that this relay has been identified as the most sensitive (lowest pickup or dropout current) device connected to
the de power system, the ground detector alann can be set at any value above 18 800 Q ; the appropriate margin
(above the threshold value) being based on engineering judgment. A setpoint of20 kQ will alann at a ground
fault current (5.91 mA) that is 5.4% below the 1naxi1num dropout current (6.25 mA).
64
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
Annex F
(informative)
Telecommunication-specific considerations
F.1 Telecommunication qualifications
Network Equipment-Building System (NEBS) is a fami ly of Ericsson/Telcordia (Telcordia is a vestige of the
Bell Labs research division of the old Bell syste1n) documents that helps to ensure safety (similar to UL), as
well as reliability at the highest levels. The documents applicable to batteries and their racks/trays/cabinets
are GR-63 and SR-3580, \vhere various flammability (the NEBS flammability requirements are based in part
on UL 94, ANSI/ATIS-0600307 .2007, and ANSI/ ATIS-06003 19.2008). Transpo1tation, vibration, seismic
(Telcordia earthquake testing requirements are more similar to the old Uniform Building Code (UBC)
requirements than the International Building Code (IBC) requirements and are based in part on ANSl/ATIS0600329.2008 (B 1]), and other requirements applicable to batteries are covered in GR-63, and the various
levels (Level 1 being basic safety and Level 3 being the highest level of reliability) are specified in SR-3580.
Most telecommunication companies require the NEBS testing (and determination of the applicability of the
various NEBS requirements) are to be done by a third-party NRTL (OSHA Nationally Recognized Testing
Laborato1y), or at least have the tests witnessed by a third-party NRTL.
In addition to the basic NEBS specifications produced by Telcordia, the telecommunications company may
require the manufacturer to have their batteries tested to at least po1tions of GR-232 for vented lead-acid
batteries, GR-4228 ("vhich also references GR- 1200 for accelerated life testing) for VRLA batteries, GR3150 for Li-based batteries, GR-3168 for NiMH batteries, and GR-3020 for Ni-Cd batteries used in outdoor
cabinets.
Finally, 1nost major telecommunication companies require batte1y (and other telecommunication equipment)
manufacturers to be ISO 9001 :2008 or TL9000 (B3 I] (a telecommunication-specific set of manufacturing
quality standards produced by the Quest Forum) certified by a registrar.
F.2 Telecommunication de power system distribution design
While a larger telecommunications site may have multiple de plants, almost all de plants only have one set
of power distribution panels or bays connected to the common de bus to \Vhich the batteries and rectifiers are
connected. From this single point, typically dual (A and B) feeds are derived to po\ver seconda1y and tertiary
fuse or breaker panels. Most loads are dual fed so that loss of a single feed does not dis1upt service.
Figure F. I shows a typical de power system layout in telecom1nunication plant and in the information and
co1nmunication technologies (ICT) equipment room including:
Rectifiers
Batteries formed of multiple strings
Primary, secondary, and tertiary distribution panels
Interconnection cable runs bet\veen different pa1ts of the de power system
In huts, customer premises (Customer Prems) and similar facilities, the de power cable is typically 1un on the
same overhead rack \Vith the rest of the communication cables, but is usually segregated on the rack from the
rest of the communications cables.
65
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
n+1
rectifiers
main
PBD(s)
(primary DC
distribution)
BDFB
(secondary DC
distribution)
(\
distribution
fed from
essential
AC bus
- backed by
engine(\_
alternator(s)
(\_
these cables
are typically run
overhead in
dedicated "un-fused"
power cable ladder rack
DC Power Room
/'\fl -
iii iii
to
DC
~ - loads
J'\J<> -
~
~
-
"\.;<> -
~ "\~
misc fuse panel
@top of an
individual
equipment bay
(tertiary DC
distribution)
"""
toA/B-fed
loads (power
supplies) in
the eqpt bay
A
B
side side
~ -
these cables
these cables
are typically run overhead
are run overhead in
In hangers on the
dedicated "fused"
outside of small-gauge
DC power
twisted pair and coax
cable ladder rack switchboard cable ladder rack
multiple
-48 v
battery
- · strings in parallel
-
ICT Equipment Room
Figure F.1-Typical telecom de power system and distribution equipment
The de distribution panels/bays are similar to a power distribution PDU or wall-1nount cabinet in a traditional ac
distiibution system, except they are typically full bays, although there are mini- and micro-battery distribution
fuse bays (BDFB) that are smaller and mounted in a cabinet or relay rack. The 1niscellaneous fuse or breaker
panels typically sit at the top of each bay/relay rack to feed the individual equip1nent shelves. In smaller sites,
the BDFBs 1nay not exist. In addition, son1e shelves are fed directly fron1 the power boards (PBDs) of the main
de plant, or directly fro1n a BDFB, bypassing the miscellaneous fuse/breaker panel.
F.3 Telecommunication power plant location
In typical teleconununication sites with only a single de plant using lead-acid batteries, the plant is typically
located on the ground floor or base1nent because of the additional floor loading capacity provided by a slab-ongrade floor to support the heavy weight rcquire1nents of lead-acid batteries. When n1ultiplc plants exist in a site,
they 1nay be located n1uch closer to the loads they serve to reduce voltage drop, power losses, and 1ninin1ize
copper cable size and cost. The batteries, rectifiers, and pri1nary distribution bays are typically located in a
co1npartJnentalized "power roo1n" for n1ediun1 and larger plants, but there arc now sn1aller distributed de
plants collocated with the loads they serve.
In typical custon1er pren1ises applications (co1nn1ercial custo1ner-owned building where the teleco1rununications
co1npany provides the equip1nent), the de plant (including its batteries) takes up a po1tion of a bay in the
co1rununications roo1n of the custo1ner, or in s1nall applications nlay even be wall-1nounted batteries, rectifiers,
and nliniature disti·ibution.
In 1nodern re1note tenninal (RT) outdoor cabinets, the de power plant 1nay be in the sa1ne chan1ber as the
electi·onic equipn1ent, or in an end cha1nber of the cabinet. The batteries are usually located in a con1pletely
separate (and separately-ventilated) con1part111ent, generally in the lowest level of the cabinet.
66
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
F.4 Telecommunication battery sizing considerations
Most telecommunication de plants designed for three or more hours of battery reserve use a battery designed
for long-duration discharge. If the site is po\vered by renewable energy, or has poor grid quality, the cycling
duty of the battery should be considered for battery type selection.
Battery sizing in telecommunication typically de-rates from the constant-current manufacturer battery capacity
tables due to generally decreasing capacity as the battery ages. Since the generally accepted knee of the life
curve of a lead-acid battery is 80% capacity, many users simply de-rate the batte1y capacity from manufacturer
tables by 20%. In a larger site, there may be many paralleled battery strings in a de plant of differing ages, so
the deration 1night be averaged to be less then 20%.
Typical nominal de syste1n bus voltages are +24 V, - 24 V, and-48 V. There are a fe"v other rarer types, such as
+48 V, - 130 V, + 130V,+140 V, 220 V, 380 V, and 575 V (the latter three are relatively new de architectures for
data centers, with limited deployments so far). Batteries are placed in series in a string to achieve the desired
float voltage (for example, - 52.80 V is typical for 24 cell strings of 1.215 s.g. lead-calcium vented lead-acid
cells, and - 54.48 V might be more typical for 38 cell strings of Ni-Cd batteries). For more advanced batteries
(such as larger format Li-ion batteries), the entire battery module typically operates near the nominal plant
voltage (while a typical Li-ion cell has an operating voltage of about 4 V, most Li-ion battery blocks sold into
telecommunication markets come pre-packaged in a box with 1nultiple internally series-paralleled cells and
a batte1y management system that accepts a telecommunication charger float voltage of - 52.08 V, - 52.21 V,
- 52.80 V, - 53.52 V, - 54.00 V, or - 54.50 V, for example). The maximum allowable float voltage of a plant is
typically determined by the lowest maximum operating voltage of all the connected loads, minus a volt or so to
help ensure that any fluctuations do not cause proble1ns (guidelines for telecommunication equipment voltage
operating windows are found in ANSI/ATIS-06003 15 [B2]). Batte1y strings (or the more advanced modules)
are typically paralleled to achieve the desired hour reserve based on the de-rated constant current capacity of
the string from the manufacturers' tables. For reliability reasons (especially \Vith VRLA batteries), in general,
even if a single string would meet the capacity requirements, the telecommunication power engineer may put
in more than one string. For example, ifthe design load is 200 A and the desired reserve time is 8 h, two strings
with de-rated 8 h rates (to the desired minimum volts per cell or mYpc, or minimum plant voltage) of l 00 A
apiece could be used. This approach means that loss of a string due to an open cell or connector significantly
reduces the available reserve time (to less than half of the design), but there may still be plenty of reserve
time. Them Ypc for multi-cell strings in telecommunication applications is determined by taking the highest
minimum operating voltage of the many pieces of load equipment, adding the designed voltage drop of the
cabling between the batteries and the loads (which is typically no more than a volt or t\vo for the low-voltage
nominal 24 V and 48 V plants), then dividing by the number of cells in a series string. The design load is
detennined by taking the float load and increasing it by either the maximum current at the end voltage of the
plant (the aforementioned highest connected load minimum operating voltage), or increasing it to a current
that would be found in the middle of the discharge window. For example, ifthe designed reserve \Vere 4 h, the
current of a constant power load at a plant voltage 2 h into the discharge curve might be used.
Illustrating telecommunication battery sizing is probably best done \Vith an exa1nple. A typical
telecommunication central office-48 V de po\ver plant might be designed with the following inputs/standards:
Lead-calcium long-duration flooded 1680 Ah cells with 1.215 s.g. from manufacturer XYZ
Plant float voltage of- 52.80 V (average of2.20 V/cell for a 24-cell string)
Equipment voltage operating window of-42.75 V to - 56.00 V
4 h minimum of designed battery reserve
Float busy-day peak average (kno\vn as List I drain in telecommunication power lingo, per Telcordia
GR-513) load of2400 A (this includes future growth over the relevant planning horizon)
67
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
Designed voltage drop from the batte1y terminals to the load during discharge of2.0 V loop (co1nbined
voltage drop of the batte1y and return cables)
Battery capacity de-rating for end-of-life (80o/o of manufacturer's listed capacity)
100% constant-power loads assumed
Average discharge voltage of 47.50 V (2 h into the rated 4 h discharge)
Given these inputs, the battery reserve can now be designed. First, the mVpc must be determined. Adding the
designed voltage drop to the minimum equipment operating voltage, and dividing by 24 yields:
v
42.75+2.0
= 44.75
v
(F. I)
44. 50 v
- 1.865 Vpc
24 cells
(F.2)
Because battery manufacturers do not have a table describing battery capacity at 1.865 V/cell, rounding to
the nearest table value of 1.86 V/cell is the prudent choice. An end-of-life de-rating (multiply by 80%) must
then be applied to the 4 h rate for the 1.86 nlVpc for 1680 Ah 1.215 s.g. lead-calcium long-duration cells from
manufacturer XYZ. Manufacturer XYZ lists a lOOo/o current capacity figure of295 A at the 4 h rate for their
1680 Ah 1.215 s.g. lead-calcium long duration battery for an mVpc of 1.86 V. De-rating this yields:
295 Ax 80%
=
236 A/ string
(F.3)
The next step is to detem1ine the average discharge load. Because the loads are constant power, and we kno"v
the fl.oat voltage load, we can calculate the average discharge load, given the average discharge voltage of
47.50 V.
2400 Ax 52.80 V
126720 W
47. 50 v
=
=
126720 W
(F.4)
2668 A
(F.5)
Knowing the de-rated string capacity, and the future protected constant power average load, the number of
strings of 1680 Ah 1.215 specific gravity flooded lead-calcium batteries can now be computed:
2 668 A
236 A l string
=
.
11 . 3 stnngs
(F.6)
Rounding up to help ensure adequate capacity for the future load yields a result of 12 parallel strings of
1680 Ah flooded lead-calcium long-duration 1.215 s.g. battery strings.
F.5 Filter of charging/rectification equipment
The good output filtering circuitry found on teleconlmunication rectifiers \Vas previously referred to as
"battery eliminator" circuitry, since the rectifiers could be feed telecommunication voice circuit equipment
\Vith no connected battery. Many rectifiers even have high ripple pulldown (HRPD) circuitry that senses
\Vhen the :filters are failing and shut then1selves do"vn to keep from introducing ripple to the loads. While
technology has changed (analog plain old telephone service or POTS is slo\vly dying), thus lessening the need
68
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
to fi lter the low-frequency ripple as strictly in telecommunication plants, the legacy has carried through, and
telecommunication rectifiers remain well filtered, producing very little ultimate de bus output ripple in either
the lo\ver frequency or higher frequency bands. While this 1nay no longer be necessary for many of the loads, it
is definitely beneficial to the batteries since they are not used as primary capacitive filters.
F.6 Telecommunication rectifier/charger controls and signals
The paralleled chargers in a modern telecommunication system are connected to a common microprocessor
controller that has the ability to shut down and restart the rectifiers (typically only in case of a runaway high
voltage in an indiv idual rectifier, or for energy management purposes in order to make the rectifiers left on
run at their most economical loading percentage). This type of installation also has the ability to control the
rectifier output voltages (usually for temperature compensated charging of VRLA batteries based on input
from one or more battery temperature sensors, or to turn down the voltage to a level that allo\VS a partial
discharge test of the batteries).
Many of the charger system controllers have external voltage sense that monitors the de battery bus. This
provides the ability to control the rectifier output voltages for voltage regulation as \Veil as for temperature
compensated charging of VRLA or other battery types based on input from one or more battery temperature
sensors. Features such as lowering the de voltage to a level that allows a partial discharge test of the batteries
can be achieved.
The rectifier controller can set the output voltage of the rectifiers based on internal sense, or external sense.
Internal sense is when the output voltage setting is matched to the common "hot" de output bus voltage in that
bay or shelf relative to the grounded return bus in that bay or shelf. This is commonly used in small de plants
where the batteries are very near the rectifiers. In larger de plants where the batteries are a little further away,
there is usually more voltage drop between the rectifiers and batteries, and it is more important to charge the
batteries based on the voltage at the batteries. In those cases, external sense is used. A pair of external sense
wires are run to the positive and negative battery termination buses above the battery stand(s), and the rectifier
output voltage(s) are adjusted to provide the float voltage set in the controller so that voltage is produced at
those "remote" battery term buses.
As partially described above, telecommunication rectifiers need to generally be capable of having their output
voltage controlled (typically in the range of 100% to 117% of the nominal plant voltage) by the de plant
controller in order to do the following:
Have the proper float voltage setting for the various types of batteries and number of series-connected
cells that could be used.
Temperature-compensated charging ofVRLA batteries.
To adjust their voltage in order to meet the required voltage at the remote sense points. This adjustment
is load dependent, as the changing current changes the voltage drop bet\veen the rectifiers and the
remote voltage sense points nearer the batteries.
Potential partial discharge tests (where the rectifier output voltage is lo\vered below open circuit, but
the rectifiers are not shut off to reduce the possibility that a battery failure would cause a catastrophic
load failure).
69
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
AnnexG
(informative)
Load sharing of chargers
Operating considerations such as increased power, reliability, and serviceability 1nay dictate that more than
one charger is connected in parallel on the same de bus. Active and passive load sharing can be available:
Active load sharing mode:
In this mode, each charger controller actively regulates its output voltage based on its output current
in order that the load current is equally shared between parallel connected chargers. Load sharing
bet\veen two parallel chargers can be achieved as follows:
-
Load sharing control through a common connection control cable.
-
Independent load sharing control with no control \Vire connection between the t\vo parallel
chargers.
Passive load sharing mode:
In this mode, the load-current is passively split bet\veen chargers based on each charger output voltage
static settings.
Active load sharing is normally only done \Vith chargers of the same brand, model, and output rating.
In telecom1nunications systems where, controlled energy management is used, so1ne rectifiers may be shut
down (or placed in "hot" standby) to make the rest of the rectifiers load-share at a level that achieves the
highest energy conversion efficiency.
70
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
Annex H
(informative)
Center tapped battery design considerations
For ungrounded 250 V systems that also supply 125 V de components in a split battery arrange1nent. Some
installations 1nay utilize a center-tapped design where the battery is separated into two sections. The battery
can be used as a single higher voltage de po\ver syste1n for larger voltage loads, (e.g., a 250 V de configuration),
or separated into two lo\ver voltage de pov;er systems for a control system or other loads (e.g., 125 V de
configurations). When using a center-tapped design, each half of the batte1y shall have its own charger(s). Care
must be taken to help ensure that the batte1y is properly sized to cany any and all loads for the anticipated cycle
of all elements. Since any bus could become grounded, the 125 V components, including any surge protection
and fi lters, should be capable ofv;ithstanding the full syste1n voltage at equalize. Refer to Figure H.1 .
_L
CHARGER
1/2 SYSTEM
VOLTAGE
-
LOAD
LO AD
-
CHARGER
1/2 SYSTEM
VOLTAGE
LOAD
I
Figure H.1-Typical center-tap battery, chargers, and load connections
71
IEEE Std 946-2020
IEEE Recommended Practice for the Design of DC Power Systems for Stationary Applications
Annex I
(informative)
Additional batteries in nuclear power generation applications
For Class IE nuclear applications, as a minimum, a separate battery shall be provided for each Engineered
Safety Feature (ESF) Division in each unit in order to provide the required independence between redundant
Class IE pov;er systems. For increased operating flexibility in designs where the reactor protection system
channels are dependent on de, the number of safety-related batteries provided on each unit should equal the
number of independent and redundant reactor protection system (instnunentation and control) channels. For
example, in a unit \Vith four reactor protection channels, four batteries should be provided. The rated capacity
of each batte1y should be determined as described above in this guide.
72